Research ArticleManufacturersrsquo Green Decision Evolution Based onMulti-Agent Modeling

Zhen Li1 Hongming Zhu1 Qingfeng Meng 1 Changzhi Wu 2 and Jianguo Du 1

1School of Management Jiangsu University Zhenjiang 212013 China2School of Management Guangzhou University Guangzhou 510006 China

Correspondence should be addressed to Qingfeng Meng mqfujseducn

Received 10 January 2019 Accepted 27 February 2019 Published 1 April 2019

Academic Editor Dehua Shen

Copyright copy 2019 Zhen Li et al This is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The development of green products is gradually becoming important due to serious ecological issues In this study an agent-basedmodel is developed to visualize and analyze the evolution of green decision-making in the manufacturing industry Based on thisagent-based model the macrobehaviors of manufacturers consumers and products are analyzed and simulated Our results showthat first themanufacturing industry emerges showing a ldquoconvergencerdquo effectThemanufacturersmay overestimate the consumersrsquogreen degree demand but this gradually gets corrected through the mechanisms of market competition Second as consumerincome increases it becomes easier for manufacturers to adapt to the marketrsquos supply and demand as impacted by the productsrsquogreen degrees and it becomes unfavorable for them to form a monopoly in the market Furthermore the profit of manufacturersis mainly derived from the sales and gradually gets more influenced by the productsrsquo green degree when the consumer incomeincreases

1 Introduction

The development and promotion of green products is grad-ually becoming important for governments due to seri-ous ecological issues Green products can be defined asproducts that are composed of recyclable materials and areremanufactured using water and energy-saving methods toreduce waste packaging and the amount of toxic materialsdisposed [1] The concept of green degree was developed todistinguish green products from other competitive productsand highlight the degree of environmental-friendliness ofgreen products It reflects the general impact of differentgreen attributes on a green product that is transparent toconsumersThe firms can then evaluate the greenness of theirproducts to consumers in a credible way for instance interms of emission levels and gas-mileage of vehicles [2]

We take Chinarsquos split air conditioners as an example toillustrate this concept In practice the green degree of splitair conditioners is divided into three levels based on theirenergy efficiency If the energy efficiency is valued above 36it is ranked as the best The values of energy efficiency for

the other two levels are [34 36) and [32 34) If the energyefficiency of an air conditioner is lower than 32 it cannot besold in the market If manufacturers create products with ahigher green degree it will not only benefit the environmentbut also enhance their corporate social image and create newcompetitive advantages [3]

Manufacturersrsquo green degree decisions concerning theirproducts can be triggered by the greendemands of consumersand markets [4] The green purchase behavior of consumersis of great significance to the decision-making of manufac-turers regarding the productrsquos green degree [5] Howeverconsumersrsquo green buying decisions are very complex Firstwhen consumers choose products most tend to evaluatewhether they can afford the product based on their incomewhich determines their purchasing power [6] Thereforeconsumer income is an important factor that affects theirpurchase intention and behavior concerning green products[7] Second an increasing number of consumers value greenproducts because these can save more energy and are envi-ronmentally friendly among other reasons [8] Besides con-sumers are affected by their social networks they may choose

HindawiComplexityVolume 2019 Article ID 3512142 14 pageshttpsdoiorg10115520193512142

2 Complexity

products based on information shared with their friends andsomenetwork platforms [9] Interactions between consumersand the influence of ldquoword-of-mouthrdquo significantly impactconsumersrsquo purchase decisions

Additionally a manufacturerrsquos green degree decisionconcerning its product is also complicated When manu-facturers make decisions on this issue they are accustomedto considering market competition feedback and resultswhich are determined by consumersrsquo purchasing behaviorsFeedback includes a manufacturerrsquos current and historicalsales and profit among others [10] Besides manufacturersmay refer to the green decisions of other manufacturersand especially learn from the most competitive company[2] Thus a manufacturerrsquos green-related decision changesdynamically and the green degrees of all products in themarket evolve at different times

Due to the complexity of the decision-making and inter-action between manufacturers and consumers the evolutionmechanism of a manufacturerrsquos green degree decision con-cerning its product is still unclearThus we find the followingproblems worth investigating

(1) How is a manufacturerrsquos green degree decision con-cerning its product affected by consumer income andpurchase decision

(2) How does the green degree of products evolve due tocompetition within the manufacturing industry

(3) At which level of green degree can products gener-ate more profits given different levels of consumerincome

Themarket system is composed of the agents of manufac-turers and consumers It is a typical complex adaptive systeminwhich agents are heterogeneous and behave adaptively [11]the interactions and transactions among manufacturers andconsumers change dynamically and the system performanceis complex and nonlinear

Agent-based modeling (ABM) is a form of ldquobottom-uprdquomodeling which is well suited for studying complex adaptivesystems comprising autonomous interacting agents whosebehaviors are based on the current state of their interactionswith other agents and the environment [12 13] ThereforeABM was adopted in this study to construct interactionsamong consumers and manufacturers We used small-worldnetworks [14] to construct interactions among consumers tomodel the ldquoword-of-mouthrdquo influence The particle swarmoptimization (PSO) algorithm [15] was applied to modeldynamic decisions of manufacturers concerning their prod-uctsrsquo green degrees Using ABM this study was able toinvestigate the emerging macrobehaviors of the system fromthe microbehaviors of consumers and manufacturers [13]Based on the model we analyzed how the manufacturersrsquogreen degree decisions are derived from consumers howthe industryrsquos green degree changes differently under varyingpurchasing powers of consumers and why products ofdifferent levels of green degrees can survive in competitivemarkets

The novelty and distinctive features of this study are asfollows First unlike previous studies that paidmore attention

to the optimal decision-making of an individual manufac-turer this study focuses more on the evolution of decision-making concerning productsrsquo green degree and the compet-itive performance of products with different green degreesin the market Second this study constructs a multiagentmodel which takes into account more complex features ofagents for instance by fully considering consumersrsquo incomeconsumersrsquo preference heterogeneity related to price andgreen degree consumer interactions and the competitionimitation and learning among manufacturers Third thisstudy presents a visual analysis of the product green degreeevolution under different consumer income scenarios and ananalysis of multidimensional performance indicators such asthe product greendegree product salesmanufacturer profitsproduct concentration degree and the matching degree ofsupply and market demand

The rest of the paper is organized as follows Section 2presents a review and analysis of the literature on decision-making related to manufacturersrsquo green products Section 3provides a systematic analysis of the agents and their interac-tionmechanisms and discusses the construction of the agent-based model Section 4 presents an analysis of the decision-making process of agents and their interactions and the set-up of simulated scenarios Section 5 discusses the simulationsand their results together with managerial insights Section 6presents the conclusions and suggestions for future research

2 Literature Review

The decision-making of manufacturers of green productsgenerally can be categorized based on two dimensions (1)market characteristics and (2) the supply chain environment

Regarding market characteristics Liu et al (2012) [16]studied the effects of competition and consumersrsquo environ-ment consciousness on the profitability of manufacturerswith different green efforts Swami and Shah (2013) [17]showed that greening efforts lead to an increase in marketdemand while the optimal greening efforts of the manufac-turers and retailers are determined by their green sensitivityand green costs Nouira et al (2014) [18] showed that byproviding variations of products with different green degreesfor ordinary and environmentally conscious consumersenterprises can gain more profit than by providing only oneproduct Zhang et al (2014) [19] studied the pricing strategyof manufacturers under a hybrid production mode wheregreen and nongreen products compete in the market andrevealed that the difference in production costs exerts aninfluence on the production choice of manufacturers whenconsumers are heterogeneous in their product evaluation Xiand Lee (2015) [20] analyzed the pricing and investment deci-sions on green product innovation in the scenario where con-sumer demand is dependent onprice and investment of greeninnovation Zhu and He (2017) [2] analyzed the influenceof various types of green products and market competitionon the green degree decision of competing enterprises Ulkuand Hsuan (2017) [21] investigated the pricing decisions andprofits of two competing enterprises under different levels ofmodularity of green products and consumersrsquo environmental

Complexity 3

sensitivity Yenipazarli and Vakharia (2017) [22] addressedthe question of to what extent a specific green productstrategy can benefit the environment while simultaneouslyresulting in profits Raza et al (2018) [23] proposed anintegrated revenue management framework to deal with themanufacturerrsquos green effort pricing and inventory decisionsin the market where consumers are heterogeneous in theirwillingness-to-pay

Focusing on the related decision-making of green prod-uct manufacturers in the supply chain environment Ghoshand Shah (2012) [24] analyzed the influence of channelstructures greening costs and consumer sensitivity towardsgreen apparels on greening levels prices and profits of thegreen supply chain Zhang and Liu (2013) [25] showed thatcooperative decision-making can guarantee that the supplychain and its members gain optimal profit and that supplychain members become more active in producing and pro-moting green products under a revenue-sharing mechanismGhosh and Shah (2015) [26] explored the impact of cost-sharing contracts on the green degree price and profit ofgreen supply chain under circumstances where consumersare relatively sensitive to the environment Huang et al (2016)[27] proposed a game-theory model to study the effect ofproduction lines supplier selection transportation selectionand pricing of green supply chain on the profitability andgreenhouse gases emissions of the supply chain Liu and Yi(2017) [28] took into consideration target advertising andshowed that the optimal price is inversely associated withthe green degree and investment of target advertising Yangand Xiao (2017) [29] examined how price green degreeand profit of green supply chains are affected by channelleadership and government subsidy under a fuzzy environ-ment where production costs and consumer demands areunclear Madani and Rasti-Barzoki (2017) [30] revealed thatcooperation among supply chain members can encouragethese members to produce greener products and gain moreeconomic benefits for themselves Xing et al (2017) [31]found that in competition with traditional manufacturersgreen manufacturers are more likely to choose the integratedchannel strategy to make rational green degree decisionsand quickly make changes in response to market demandsSong and Gao (2018) [32] showed that a revenue-sharingcontract is relatively effective in enhancing the green degreeand profit of the whole green supply chain Hong et al (2018)[33] explored an effective way to optimize the service timeand production selection decision of the green supply chainconsidering service time and emission regulations

The existing study (on manufacturersrsquo decision-makingconcerning their productrsquos green degree) mainly focuses onthe optimal decision-making of individual manufacturersunder different circumstances and less on the decision ofmanufacturers and the industry as a whole In actualityin market competitions among manufacturers there areinteractions such as mutual imitation and learning whichexert influence on the productsrsquo green degree decision madeby manufacturers Moreover the existing study describesmarket demand as a linear function of the productrsquos priceand green degree but ignores the interactions among con-sumers and consumersrsquo learning and imitation behaviors in

the process of making purchase decisions Meanwhile theconsumersrsquo purchasing decisions are not only influenced bythe multiattributes of the product (such as price and quality)but are also impacted by other consumers in the interactionnetwork

3 Agent-Based Modeling

31 Model Description and Analysis The market includesconsumer andmanufacturer agents as well as green productsManufacturers and consumers are connected via productsthey interact based on product supply and demand Therelationships among agents are described in Figure 1 Theagentsrsquo decision-making and interaction mechanisms areconsidered and shown in this model

As shown in Figure 1 the products have different priceand green degrees which are provided by different man-ufacturers and diversely meet the needs of different con-sumers During market transactions the consumers makepurchasing decisions and the manufacturers make productgreen degree decisions The factors influencing consumersrsquopurchase decisions include consumer income price andgreen degree expectations and their sensitivity to the priceand green degree among others Additionally the ldquoword-of-mouthrdquo recommendation and imitation tendencies basedon interactions among consumers also affect the purchasedecisions of consumers The manufacturers may adjust theirgreen degree decisions based on feedback from productcompetition including current and historical experiencesMoreover the manufacturer may learn from other bench-mark manufacturers through interactions among manufac-turers All agentsrsquo decisions may adjust dynamically withthe productsrsquo green degree changing accordingly [34] As aresult the greendegree of themanufacturing industry evolvesnonlinearly

The systemrsquos macroperformance such as the industryrsquosproduct green degree and competition performance can bestudied through the interactions of these agents in themarketand indicators including manufacturersrsquo sales and profit aswell as the product concentration degree constantly emergein different curves under various scenarios

32 Consumer Agents

321 Consumer Interaction Network During the purchaseand consumption process consumers are connected throughone or more formal and informal relationships called theconsumer interaction network [35] The consumers areembedded in the interaction network therefore their pur-chase decisions are influenced by other consumers in thisnetwork [36] Newman (2000) [37] suggested that manyreal-world social networks have the famous ldquosmall-worldrdquoproperty therefore ldquosmall worldrdquo is used to describe thestructure of consumer interaction networks Watts and Stro-gatz (1998) [38] suggested that the small-world networksare characterized by a high degree of local clustering (likeregular lattices) but also possess a short diameter or vertex-to-vertex distances which can be constructed as follows

4 Complexity

IncomePrice expectationGreen degree expectationPrice sensitivityGreen degree sensitivityHerding tendency

Consumers

Consumer social network

Small-world networkNumber of neighbour nodesRewiring probability

Feedback from consumers

Demand levelAverage green degree weighted by salesSales volumeMarket share

ConsumersrsquoInteraction

Word-of-mouthImitationHerding effect

Products

PriceGreen degree

ManufacturersProductCostProfitSelf-learning abilitySwarm learning ability

Manufacturersrsquointeraction

CompetitionMutual learning

Figure 1 The system structure diagram

Starting from a ring lattice with 119873 vertices and 119870 edges pervertex each edge is rewired at random with probability 119875119903This construction produces graphs between regularity (119875119903 =0) and disorder (119875119903 = 1) such that 119875119903 in the intermediateregion 0 lt 119875119903 lt 1 produces some degree of both Specificallyit is required that 119873 ge 119870 ge ln119873 ge 1 where 119870 ge ln119873guarantees that a random graph will be connected Henceconsumers will be affected by the characteristics of the small-world network such as the number of neighbor nodes 119870which can characterize the number of consumers interactingwith one consumer in making purchase decisions and therewiring probability 119875119903 which represents the probabilitythat the consumerrsquos neighbors will change their purchasedecisions

322 Consumer Purchase Decision-Making The manufac-turers and consumers can be denoted as 119894 and 119895 and theirnumbers are 119872 and 119873 respectively 119880119895119894 is the utility ofthe product 119894 for the consumer 119895 (119895 isin (1 sdot sdot sdot 119873)) whichcan indicate the consumersrsquo motivation and willingness topurchase products The consumers make purchase decisionsaccording to the utility of the product The motivationalfunction proposed by Zhang and Zhang (2007) [39] is used asthe decision basis for consumers to make purchase decisionsThe utility 119880119895119894 is shown in

119880119895119894 = 120583119895119894 lowast 119901119895119894 + 120588119895119894 lowast 119892119895119894 (1)

Theutility119880119895119894 of the product 119894 for the consumer 119895 ismainlyinfluenced by two factors the price and quality of the productIn (1) 120583119895119894 represents the price sensitivity of consumer 119895 tothe product 119894 Price sensitivity which shows the consumerrsquos

cognition and perception of the goodsrsquo value includingpractical and emotional values is one of the attributes ofconsumers and varies from consumer to consumerThe pricesensitivity distribution model [40] suggests that a consumerrsquosprice sensitivity is an exponential function of the differencebetween the real price of a product and the expected price ofthe product as shown in (2) where 120582119895 gt 1 and 119901(119890)119895119894 is theconsumer 119895rsquos expected price of the product 119894

120583j119894 = 1

1 + 120582119895(119875119895119894 minus119875(119890)

119895119894 )

(2)

On the other side 120588119895119894 is the green degree sensitivity ofconsumer 119895 to the product 119894 In thismodel the greendegree ofthe product refers to the general impact from multiple greenattributes of the product while multiple green attributesare equivalent to qualities of the product at a certain leveltherefore the sensitivity of the consumer to the productrsquosgreen degree can be regarded as the sensitivity to the productqualityThe outlier avoidance consumer psychological theory[41] suggests that the more closely the quality of the productapproximates the consumerrsquos expected quality of this kind ofproduct the more sensitive the consumer is to the qualityof the product The quality sensitivity 120588119895119894 is as shown in (3)where 0 lt 120573119895 lt 1 and 119892(119890)119895119894 is consumer 119895rsquos expected greendegree of the product 119894

120588119895119894 = 120573119895|119892119895

119894 minus119892(119890)119895

119894 | (3)

In reality consumers are embedded in the correspondingsocial networks therefore their purchase decisions for thesame kind of products with similar utilities are not only

Complexity 5

influenced by the price and green degree of products butare also affected by other consumers in the network such asfriendsrsquo recommendation and word of mouth among othersBesides whenmaking purchase decisions consumers show acertain degree of herdmentality [42] Herding behavior refersto the phenomenon where an agentrsquos decision-making in themarket is influenced by what others around him are doing[43] When consumer 119895 interacts with other consumers theutility of the product for himself will change correspondinglyunder the influence of the herd mentality The changing rulesof product utility for consumer 119895 under the influence of herdeffects are shown in formula (4) as follows

119867119864119895 = 120579119895 lowast infl119895119897 (4)

where 119867119864119895 refers to the herd effect on consumer 119895 andthe parameter 120579119895 which is subject to uniform distributiondenotes the herd tendency of consumer 119895The closer the valueof 120579119895 is to 0 the less likely a consumer is to be influencedby the surrounding people and the weaker the herd effect iswhile the larger the value of 120579119895 the more likely a consumer isto be influenced by the surrounding people and the strongerthe herd effect is The parameter infl119895119897 which denotes theinfluence of other consumers in the consumer interactionnetwork as perceived by consumer 119895 can be obtained fromthe average utility of product 119894 for neighboring consumersas shown in formula (5) where ℎ is the number of neighborswho interact with consumer 119895 in the consumer interactionnetwork and 119880119894119895119897 is the utility of product 119894 for the neighboringconsumer 119897 (119897 isin (1 sdot sdot sdot ℎ))

infl119895119897 =sumℎ119897=1119880119894119895119897

ℎ(5)

Accordingly on the basis of formulas (1) to (5) theproductrsquos utility can be further computed as shown in

119880119895119894 = 11 + 120582119895(119901

119895119894 minus119901(119890)

119895119894 )

lowast 119901119895119894 + 120573119895|119892119895119894minus119892(119890)

119895119894 | lowast 119892119895119894 + 120579119895 lowast infl119895119897 (6)

According to related literature [44] there is a positivecorrelation between consumerrsquos expected green degrees ofthe product and hisher income that is a consumer with ahigher income places more emphasis on the environmentalperformance of the product and hisher expected greendegree of the product is much higherTherefore a consumerrsquosexpected green degree of the product is shown in formula(7) where 119894119899119888119900119898119890119895 is the income of consumer 119895 119898 is theregression coefficient between a consumerrsquos expected greendegree and hisher income and 119898 gt 0 119886 is a scalingconstant indicating the fixed expectation for the green degreeof the product when a consumerrsquos income is 0 and 120596119895 is thestochastic disturbance drawn from a uniform distributionindicating the deviation of a consumerrsquos expected greendegree of the product

119892(119890)119895119894 = 119898 lowast 119894119899119888119900119898119890119895 + 119886 + 120596119895 (7)

In addition it is assumed that consumer 119895 can estimatethe expected price 119901(119890)119895119894 of product 119894 based on hisher own

expected green degree 119892(119890)119895119894 and 119901(119890)119895119894 increases with 119892(119890)119895119894 Moreover following Eppstein et al (2011) [45] it is alsoassumed that if the product price exceeds 100 of theexpected price consumers will consider the product notaffordable and refuse to purchase it Therefore the utilityfunction of consumersrsquo evaluation of different products isshown in formula (8) below

119880119895119894 = 11 + 120582119895(119901

119895

119894 minus119901(119890)119895

119894 )lowast 119901119895119894 + 120573119895|119892

119895

119894minus119898lowast119894119899119888119900119898119890119894minus120596119895| lowast 119892119895119894

+ 120579119895 lowast infl119895119897

(8)

To obtain the maximum utility consumers compare theutilities of products from all manufacturers in the market andfollow the rules below to make purchase decisions product 1will be purchased when 1198801198951 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product2 will be purchased when 1198801198952 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product119872 will be purchased when 119880119895119872 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 andthe product is purchased randomly when 1198801198951 = 1198801198952 = sdot sdot sdot =119880119895119872

33 Manufacturersrsquo Agents

331 Manufacturersrsquo Profits In this study there are 119872competing manufacturers in the market It is assumed thatproducts from different manufacturers are similar in theirfunction and performance but differ in price and greendegrees therefore manufacturers compete based on the priceand green degrees of the products

The production process of products with higher greendegrees may be more complex and require higher levels oftechnology therefore it is inevitable for the manufacturer toinvest additional costs in certain aspects of the productionto improve the green degree of products The unit costfunction of green products can be separated into two partsone is the fixed regular unit production cost 119888119894 the other isthe additional margin cost 1198881015840119894 Based on relevant literature[16] the additional margin cost is a quadratic function withthe product green degree as such 119888119894 can be set as 1198881015840119894 =(12)1199031198941198921198942 where 119903119894 represents the cost factor related to thegreen production efforts and 119903119894 gt 0 Therefore the unit costfunction of green products is shown in formula (9) below

119888119894 = 119888119894 + 121199031198941198921198942 (9)

With regard to product pricing it is assumed that themanufacturer applies the cost-plus pricing method that isthe price of the product 119894 is set as 119901119894 = (1 + 119900119894)119888119894 where119900119894 gt 0 represents the mark-up on the product Accordinglythe product price 119901119894 is shown in formula (10) below

119901119895119894 = (1 + 119900119894) 119888119894 = (1 + 119900119894) (119888119894 + 121199031198941198921198942) (10)

6 Complexity

Moreover we set the parameter 119901119888119895119894 to indicate whetherconsumer 119895 purchases the product 119894 and if the answer is yes119901119888119895119894 = 1 otherwise 119901119888119895119894 = 0 then the sales volume of product119894 can be calculated as in formula (11) below where 119902119894 is thesales volume of product 119894

119902119894 =119873

sum119895=1

119901119888119895119894 (11)

Thus the profit of the manufacturer is shown in formula(12) below

120587119894 = (119901119895119894 minus 119888119895119894 ) 119902119894 = (119901119895119894 minus 119888119895119894 )119873

sum119895=1

119901119888119895119894 (12)

332Manufacturersrsquo Decision-Making Manufacturers adjustthe green degrees of their products based on their profits Tobetter analyze the evolution of manufacturersrsquo product greendegree decision-making this study assumes that there are notechnical barriers and production performance limitationsfor manufacturers indicating manufacturers can adjust thegreen degree of their products based on their profits withouthaving to bear additional costs from production conversionMoreover the algorithm PSO [15] is applied to simulate themanufacturersrsquo productsrsquo green degree adjustment behaviorso that the mutual imitation and learning among manu-facturers can be better considered A manufacturer learnsand imitates from the ldquobenchmark manufacturerrdquo with thebest profit in the market Manufacturer 119894 determines thesubsequent green degree of its products based on its cur-rent product green degree (119892119895119894 (119905)) the historical productgreen degree (119901119887119890119904119905119895119894 ) that resulted in its best profit thusfar and the current product green degree of the ldquobench-mark manufacturerrdquo in the market (119892119887119890119904119905119895119894 ) The productgreen degree of manufacturer 119894 is thus updated by thefollowing

Step 1 Initialize the manufacturer parameters that is set-ting the number of manufacturers 119872 and the initialproduct green degree 119892119895119894 (119905) randomly generating velocityV119894(119905)Step 2 Calculate each manufacturerrsquos current profit 120587119894(119905)under its current product green degree 119892119895119894 (119905)Step 3 If the manufacturerrsquos current profit 120587119894(119905) is largerthan the manufacturerrsquos previous best profit 120587119894119901 then the120587119894119901 is replaced with the 120587119894(119905) and the product green degreecorresponding to the previous best profit 119901119887119890119904119905119895119894 is updated tothe current green degree 119892119895119894 (119905)Step 4 It is assumed that the manufacturer can learn fromthe ldquobenchmarkmanufacturerrdquoTheprofit of the ldquobenchmarkmanufacturerrdquo in the current market is recorded as 120587119887119890119904119905which can be shown as 120587119887119890119904119905 = max(120587119894(119905)) and the productgreen degree of the ldquobenchmark manufacturerrdquo is updated to119892119887119890119904119905119895119894

Step 5 Thevelocity and the product greendegree are updatedas shown in formulas (13) and (14) below

V119894 (119905 + 1) = 1198880V119894 (119905) + 11988811199031198861198991198891 (119901119887119890119904119905119895119894 minus 119892119895119894 (119905))+ 11988821199031198861198991198892 (119892119887119890119904119905119895119894 minus 119892119895119894 (119905))

(13)

119892119895119894 (119905 + 1) = 119892119895119894 (119905) + V119894 (119905 + 1) (14)In formula (13) 1198880 is the weight coefficient whose value

set in the range of [09 12] and enables the PSO algorithmto have good convergence based on relevant literature [46]therefore we set 1198880 = 1 1198881 is the manufacturerrsquos learning coef-ficient which reflects its self-learning ability from its history1198882 is the manufacturerrsquos swarm cognition coefficient whichindicates the degree of collaboration between manufacturersand the degree of themanufacturerrsquos acceptance of the grouprsquoscommon information [47] and 1199031198861198991198891 1199031198861198991198892 are randomnumbers drawn from a uniform distribution in the range(01)

Step 6 Return to Step 2 until all manufacturers have beenlooped through

4 Simulation Scenarios Design

We collected information about manufacturers and annualsales of split air conditioners from Taobao and JD andreferred to the ratio between the number of manufacturersand their annual sales We assumed that there are 10 man-ufacturers and 10000 consumers in the market and theirdecision-making processes and interactions are shown inFigure 2 In the initial period of simulation manufacturerssell their products with different prices and green degreesThe consumers calculate the utilities of all products thatmeet their purchasing power and make purchase decisionsbased on the principle of maximum utility After consumerspurchase products manufacturers are able to obtain theirown competitive performance indicators such as sales andprofit and they seek out the ldquobenchmark manufacturerrdquo bycomparing the profit of other manufacturers Subsequentlythey adjust the productsrsquo green degree and prices in thenext business cycle based on the PSO algorithm and otherinfluencing factors

Therefore we keep the basic parameters of the modelunchanged and change 119868119899119888119900119898119890119895 in three different scenariosas well as the parameter 119868119899119888119900119898119890119895 values in three ranges119873(400 75)119873(500 75) and119873(600 75) correspondinglyThevariances of consumer income levels are kept constant andthemeans of consumer income levels are different in the threescenarios

To reduce randomness in the simulations and improvethe statistical stability and validity of simulation results eachsimulation scenario is ran 10 times and the total numberof ticks per simulation scenario 119879max is set to 150 theaverage results of which is then statistically analyzed Theinitial value-setting of the basic model parameters is shownin Table 1

To analyze the manufacturersrsquo decision-making concern-ing their productsrsquo green degree in a more detailed manner

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 3

sensitivity Yenipazarli and Vakharia (2017) [22] addressedthe question of to what extent a specific green productstrategy can benefit the environment while simultaneouslyresulting in profits Raza et al (2018) [23] proposed anintegrated revenue management framework to deal with themanufacturerrsquos green effort pricing and inventory decisionsin the market where consumers are heterogeneous in theirwillingness-to-pay

Focusing on the related decision-making of green prod-uct manufacturers in the supply chain environment Ghoshand Shah (2012) [24] analyzed the influence of channelstructures greening costs and consumer sensitivity towardsgreen apparels on greening levels prices and profits of thegreen supply chain Zhang and Liu (2013) [25] showed thatcooperative decision-making can guarantee that the supplychain and its members gain optimal profit and that supplychain members become more active in producing and pro-moting green products under a revenue-sharing mechanismGhosh and Shah (2015) [26] explored the impact of cost-sharing contracts on the green degree price and profit ofgreen supply chain under circumstances where consumersare relatively sensitive to the environment Huang et al (2016)[27] proposed a game-theory model to study the effect ofproduction lines supplier selection transportation selectionand pricing of green supply chain on the profitability andgreenhouse gases emissions of the supply chain Liu and Yi(2017) [28] took into consideration target advertising andshowed that the optimal price is inversely associated withthe green degree and investment of target advertising Yangand Xiao (2017) [29] examined how price green degreeand profit of green supply chains are affected by channelleadership and government subsidy under a fuzzy environ-ment where production costs and consumer demands areunclear Madani and Rasti-Barzoki (2017) [30] revealed thatcooperation among supply chain members can encouragethese members to produce greener products and gain moreeconomic benefits for themselves Xing et al (2017) [31]found that in competition with traditional manufacturersgreen manufacturers are more likely to choose the integratedchannel strategy to make rational green degree decisionsand quickly make changes in response to market demandsSong and Gao (2018) [32] showed that a revenue-sharingcontract is relatively effective in enhancing the green degreeand profit of the whole green supply chain Hong et al (2018)[33] explored an effective way to optimize the service timeand production selection decision of the green supply chainconsidering service time and emission regulations

The existing study (on manufacturersrsquo decision-makingconcerning their productrsquos green degree) mainly focuses onthe optimal decision-making of individual manufacturersunder different circumstances and less on the decision ofmanufacturers and the industry as a whole In actualityin market competitions among manufacturers there areinteractions such as mutual imitation and learning whichexert influence on the productsrsquo green degree decision madeby manufacturers Moreover the existing study describesmarket demand as a linear function of the productrsquos priceand green degree but ignores the interactions among con-sumers and consumersrsquo learning and imitation behaviors in

the process of making purchase decisions Meanwhile theconsumersrsquo purchasing decisions are not only influenced bythe multiattributes of the product (such as price and quality)but are also impacted by other consumers in the interactionnetwork

3 Agent-Based Modeling

31 Model Description and Analysis The market includesconsumer andmanufacturer agents as well as green productsManufacturers and consumers are connected via productsthey interact based on product supply and demand Therelationships among agents are described in Figure 1 Theagentsrsquo decision-making and interaction mechanisms areconsidered and shown in this model

As shown in Figure 1 the products have different priceand green degrees which are provided by different man-ufacturers and diversely meet the needs of different con-sumers During market transactions the consumers makepurchasing decisions and the manufacturers make productgreen degree decisions The factors influencing consumersrsquopurchase decisions include consumer income price andgreen degree expectations and their sensitivity to the priceand green degree among others Additionally the ldquoword-of-mouthrdquo recommendation and imitation tendencies basedon interactions among consumers also affect the purchasedecisions of consumers The manufacturers may adjust theirgreen degree decisions based on feedback from productcompetition including current and historical experiencesMoreover the manufacturer may learn from other bench-mark manufacturers through interactions among manufac-turers All agentsrsquo decisions may adjust dynamically withthe productsrsquo green degree changing accordingly [34] As aresult the greendegree of themanufacturing industry evolvesnonlinearly

The systemrsquos macroperformance such as the industryrsquosproduct green degree and competition performance can bestudied through the interactions of these agents in themarketand indicators including manufacturersrsquo sales and profit aswell as the product concentration degree constantly emergein different curves under various scenarios

32 Consumer Agents

321 Consumer Interaction Network During the purchaseand consumption process consumers are connected throughone or more formal and informal relationships called theconsumer interaction network [35] The consumers areembedded in the interaction network therefore their pur-chase decisions are influenced by other consumers in thisnetwork [36] Newman (2000) [37] suggested that manyreal-world social networks have the famous ldquosmall-worldrdquoproperty therefore ldquosmall worldrdquo is used to describe thestructure of consumer interaction networks Watts and Stro-gatz (1998) [38] suggested that the small-world networksare characterized by a high degree of local clustering (likeregular lattices) but also possess a short diameter or vertex-to-vertex distances which can be constructed as follows

4 Complexity

IncomePrice expectationGreen degree expectationPrice sensitivityGreen degree sensitivityHerding tendency

Consumers

Consumer social network

Small-world networkNumber of neighbour nodesRewiring probability

Feedback from consumers

Demand levelAverage green degree weighted by salesSales volumeMarket share

ConsumersrsquoInteraction

Word-of-mouthImitationHerding effect

Products

PriceGreen degree

ManufacturersProductCostProfitSelf-learning abilitySwarm learning ability

Manufacturersrsquointeraction

CompetitionMutual learning

Figure 1 The system structure diagram

Starting from a ring lattice with 119873 vertices and 119870 edges pervertex each edge is rewired at random with probability 119875119903This construction produces graphs between regularity (119875119903 =0) and disorder (119875119903 = 1) such that 119875119903 in the intermediateregion 0 lt 119875119903 lt 1 produces some degree of both Specificallyit is required that 119873 ge 119870 ge ln119873 ge 1 where 119870 ge ln119873guarantees that a random graph will be connected Henceconsumers will be affected by the characteristics of the small-world network such as the number of neighbor nodes 119870which can characterize the number of consumers interactingwith one consumer in making purchase decisions and therewiring probability 119875119903 which represents the probabilitythat the consumerrsquos neighbors will change their purchasedecisions

322 Consumer Purchase Decision-Making The manufac-turers and consumers can be denoted as 119894 and 119895 and theirnumbers are 119872 and 119873 respectively 119880119895119894 is the utility ofthe product 119894 for the consumer 119895 (119895 isin (1 sdot sdot sdot 119873)) whichcan indicate the consumersrsquo motivation and willingness topurchase products The consumers make purchase decisionsaccording to the utility of the product The motivationalfunction proposed by Zhang and Zhang (2007) [39] is used asthe decision basis for consumers to make purchase decisionsThe utility 119880119895119894 is shown in

119880119895119894 = 120583119895119894 lowast 119901119895119894 + 120588119895119894 lowast 119892119895119894 (1)

Theutility119880119895119894 of the product 119894 for the consumer 119895 ismainlyinfluenced by two factors the price and quality of the productIn (1) 120583119895119894 represents the price sensitivity of consumer 119895 tothe product 119894 Price sensitivity which shows the consumerrsquos

cognition and perception of the goodsrsquo value includingpractical and emotional values is one of the attributes ofconsumers and varies from consumer to consumerThe pricesensitivity distribution model [40] suggests that a consumerrsquosprice sensitivity is an exponential function of the differencebetween the real price of a product and the expected price ofthe product as shown in (2) where 120582119895 gt 1 and 119901(119890)119895119894 is theconsumer 119895rsquos expected price of the product 119894

120583j119894 = 1

1 + 120582119895(119875119895119894 minus119875(119890)

119895119894 )

(2)

On the other side 120588119895119894 is the green degree sensitivity ofconsumer 119895 to the product 119894 In thismodel the greendegree ofthe product refers to the general impact from multiple greenattributes of the product while multiple green attributesare equivalent to qualities of the product at a certain leveltherefore the sensitivity of the consumer to the productrsquosgreen degree can be regarded as the sensitivity to the productqualityThe outlier avoidance consumer psychological theory[41] suggests that the more closely the quality of the productapproximates the consumerrsquos expected quality of this kind ofproduct the more sensitive the consumer is to the qualityof the product The quality sensitivity 120588119895119894 is as shown in (3)where 0 lt 120573119895 lt 1 and 119892(119890)119895119894 is consumer 119895rsquos expected greendegree of the product 119894

120588119895119894 = 120573119895|119892119895

119894 minus119892(119890)119895

119894 | (3)

In reality consumers are embedded in the correspondingsocial networks therefore their purchase decisions for thesame kind of products with similar utilities are not only

Complexity 5

influenced by the price and green degree of products butare also affected by other consumers in the network such asfriendsrsquo recommendation and word of mouth among othersBesides whenmaking purchase decisions consumers show acertain degree of herdmentality [42] Herding behavior refersto the phenomenon where an agentrsquos decision-making in themarket is influenced by what others around him are doing[43] When consumer 119895 interacts with other consumers theutility of the product for himself will change correspondinglyunder the influence of the herd mentality The changing rulesof product utility for consumer 119895 under the influence of herdeffects are shown in formula (4) as follows

119867119864119895 = 120579119895 lowast infl119895119897 (4)

where 119867119864119895 refers to the herd effect on consumer 119895 andthe parameter 120579119895 which is subject to uniform distributiondenotes the herd tendency of consumer 119895The closer the valueof 120579119895 is to 0 the less likely a consumer is to be influencedby the surrounding people and the weaker the herd effect iswhile the larger the value of 120579119895 the more likely a consumer isto be influenced by the surrounding people and the strongerthe herd effect is The parameter infl119895119897 which denotes theinfluence of other consumers in the consumer interactionnetwork as perceived by consumer 119895 can be obtained fromthe average utility of product 119894 for neighboring consumersas shown in formula (5) where ℎ is the number of neighborswho interact with consumer 119895 in the consumer interactionnetwork and 119880119894119895119897 is the utility of product 119894 for the neighboringconsumer 119897 (119897 isin (1 sdot sdot sdot ℎ))

infl119895119897 =sumℎ119897=1119880119894119895119897

ℎ(5)

Accordingly on the basis of formulas (1) to (5) theproductrsquos utility can be further computed as shown in

119880119895119894 = 11 + 120582119895(119901

119895119894 minus119901(119890)

119895119894 )

lowast 119901119895119894 + 120573119895|119892119895119894minus119892(119890)

119895119894 | lowast 119892119895119894 + 120579119895 lowast infl119895119897 (6)

According to related literature [44] there is a positivecorrelation between consumerrsquos expected green degrees ofthe product and hisher income that is a consumer with ahigher income places more emphasis on the environmentalperformance of the product and hisher expected greendegree of the product is much higherTherefore a consumerrsquosexpected green degree of the product is shown in formula(7) where 119894119899119888119900119898119890119895 is the income of consumer 119895 119898 is theregression coefficient between a consumerrsquos expected greendegree and hisher income and 119898 gt 0 119886 is a scalingconstant indicating the fixed expectation for the green degreeof the product when a consumerrsquos income is 0 and 120596119895 is thestochastic disturbance drawn from a uniform distributionindicating the deviation of a consumerrsquos expected greendegree of the product

119892(119890)119895119894 = 119898 lowast 119894119899119888119900119898119890119895 + 119886 + 120596119895 (7)

In addition it is assumed that consumer 119895 can estimatethe expected price 119901(119890)119895119894 of product 119894 based on hisher own

expected green degree 119892(119890)119895119894 and 119901(119890)119895119894 increases with 119892(119890)119895119894 Moreover following Eppstein et al (2011) [45] it is alsoassumed that if the product price exceeds 100 of theexpected price consumers will consider the product notaffordable and refuse to purchase it Therefore the utilityfunction of consumersrsquo evaluation of different products isshown in formula (8) below

119880119895119894 = 11 + 120582119895(119901

119895

119894 minus119901(119890)119895

119894 )lowast 119901119895119894 + 120573119895|119892

119895

119894minus119898lowast119894119899119888119900119898119890119894minus120596119895| lowast 119892119895119894

+ 120579119895 lowast infl119895119897

(8)

To obtain the maximum utility consumers compare theutilities of products from all manufacturers in the market andfollow the rules below to make purchase decisions product 1will be purchased when 1198801198951 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product2 will be purchased when 1198801198952 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product119872 will be purchased when 119880119895119872 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 andthe product is purchased randomly when 1198801198951 = 1198801198952 = sdot sdot sdot =119880119895119872

33 Manufacturersrsquo Agents

331 Manufacturersrsquo Profits In this study there are 119872competing manufacturers in the market It is assumed thatproducts from different manufacturers are similar in theirfunction and performance but differ in price and greendegrees therefore manufacturers compete based on the priceand green degrees of the products

The production process of products with higher greendegrees may be more complex and require higher levels oftechnology therefore it is inevitable for the manufacturer toinvest additional costs in certain aspects of the productionto improve the green degree of products The unit costfunction of green products can be separated into two partsone is the fixed regular unit production cost 119888119894 the other isthe additional margin cost 1198881015840119894 Based on relevant literature[16] the additional margin cost is a quadratic function withthe product green degree as such 119888119894 can be set as 1198881015840119894 =(12)1199031198941198921198942 where 119903119894 represents the cost factor related to thegreen production efforts and 119903119894 gt 0 Therefore the unit costfunction of green products is shown in formula (9) below

119888119894 = 119888119894 + 121199031198941198921198942 (9)

With regard to product pricing it is assumed that themanufacturer applies the cost-plus pricing method that isthe price of the product 119894 is set as 119901119894 = (1 + 119900119894)119888119894 where119900119894 gt 0 represents the mark-up on the product Accordinglythe product price 119901119894 is shown in formula (10) below

119901119895119894 = (1 + 119900119894) 119888119894 = (1 + 119900119894) (119888119894 + 121199031198941198921198942) (10)

6 Complexity

Moreover we set the parameter 119901119888119895119894 to indicate whetherconsumer 119895 purchases the product 119894 and if the answer is yes119901119888119895119894 = 1 otherwise 119901119888119895119894 = 0 then the sales volume of product119894 can be calculated as in formula (11) below where 119902119894 is thesales volume of product 119894

119902119894 =119873

sum119895=1

119901119888119895119894 (11)

Thus the profit of the manufacturer is shown in formula(12) below

120587119894 = (119901119895119894 minus 119888119895119894 ) 119902119894 = (119901119895119894 minus 119888119895119894 )119873

sum119895=1

119901119888119895119894 (12)

332Manufacturersrsquo Decision-Making Manufacturers adjustthe green degrees of their products based on their profits Tobetter analyze the evolution of manufacturersrsquo product greendegree decision-making this study assumes that there are notechnical barriers and production performance limitationsfor manufacturers indicating manufacturers can adjust thegreen degree of their products based on their profits withouthaving to bear additional costs from production conversionMoreover the algorithm PSO [15] is applied to simulate themanufacturersrsquo productsrsquo green degree adjustment behaviorso that the mutual imitation and learning among manu-facturers can be better considered A manufacturer learnsand imitates from the ldquobenchmark manufacturerrdquo with thebest profit in the market Manufacturer 119894 determines thesubsequent green degree of its products based on its cur-rent product green degree (119892119895119894 (119905)) the historical productgreen degree (119901119887119890119904119905119895119894 ) that resulted in its best profit thusfar and the current product green degree of the ldquobench-mark manufacturerrdquo in the market (119892119887119890119904119905119895119894 ) The productgreen degree of manufacturer 119894 is thus updated by thefollowing

Step 1 Initialize the manufacturer parameters that is set-ting the number of manufacturers 119872 and the initialproduct green degree 119892119895119894 (119905) randomly generating velocityV119894(119905)Step 2 Calculate each manufacturerrsquos current profit 120587119894(119905)under its current product green degree 119892119895119894 (119905)Step 3 If the manufacturerrsquos current profit 120587119894(119905) is largerthan the manufacturerrsquos previous best profit 120587119894119901 then the120587119894119901 is replaced with the 120587119894(119905) and the product green degreecorresponding to the previous best profit 119901119887119890119904119905119895119894 is updated tothe current green degree 119892119895119894 (119905)Step 4 It is assumed that the manufacturer can learn fromthe ldquobenchmarkmanufacturerrdquoTheprofit of the ldquobenchmarkmanufacturerrdquo in the current market is recorded as 120587119887119890119904119905which can be shown as 120587119887119890119904119905 = max(120587119894(119905)) and the productgreen degree of the ldquobenchmark manufacturerrdquo is updated to119892119887119890119904119905119895119894

Step 5 Thevelocity and the product greendegree are updatedas shown in formulas (13) and (14) below

V119894 (119905 + 1) = 1198880V119894 (119905) + 11988811199031198861198991198891 (119901119887119890119904119905119895119894 minus 119892119895119894 (119905))+ 11988821199031198861198991198892 (119892119887119890119904119905119895119894 minus 119892119895119894 (119905))

(13)

119892119895119894 (119905 + 1) = 119892119895119894 (119905) + V119894 (119905 + 1) (14)In formula (13) 1198880 is the weight coefficient whose value

set in the range of [09 12] and enables the PSO algorithmto have good convergence based on relevant literature [46]therefore we set 1198880 = 1 1198881 is the manufacturerrsquos learning coef-ficient which reflects its self-learning ability from its history1198882 is the manufacturerrsquos swarm cognition coefficient whichindicates the degree of collaboration between manufacturersand the degree of themanufacturerrsquos acceptance of the grouprsquoscommon information [47] and 1199031198861198991198891 1199031198861198991198892 are randomnumbers drawn from a uniform distribution in the range(01)

Step 6 Return to Step 2 until all manufacturers have beenlooped through

4 Simulation Scenarios Design

We collected information about manufacturers and annualsales of split air conditioners from Taobao and JD andreferred to the ratio between the number of manufacturersand their annual sales We assumed that there are 10 man-ufacturers and 10000 consumers in the market and theirdecision-making processes and interactions are shown inFigure 2 In the initial period of simulation manufacturerssell their products with different prices and green degreesThe consumers calculate the utilities of all products thatmeet their purchasing power and make purchase decisionsbased on the principle of maximum utility After consumerspurchase products manufacturers are able to obtain theirown competitive performance indicators such as sales andprofit and they seek out the ldquobenchmark manufacturerrdquo bycomparing the profit of other manufacturers Subsequentlythey adjust the productsrsquo green degree and prices in thenext business cycle based on the PSO algorithm and otherinfluencing factors

Therefore we keep the basic parameters of the modelunchanged and change 119868119899119888119900119898119890119895 in three different scenariosas well as the parameter 119868119899119888119900119898119890119895 values in three ranges119873(400 75)119873(500 75) and119873(600 75) correspondinglyThevariances of consumer income levels are kept constant andthemeans of consumer income levels are different in the threescenarios

To reduce randomness in the simulations and improvethe statistical stability and validity of simulation results eachsimulation scenario is ran 10 times and the total numberof ticks per simulation scenario 119879max is set to 150 theaverage results of which is then statistically analyzed Theinitial value-setting of the basic model parameters is shownin Table 1

To analyze the manufacturersrsquo decision-making concern-ing their productsrsquo green degree in a more detailed manner

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

4 Complexity

IncomePrice expectationGreen degree expectationPrice sensitivityGreen degree sensitivityHerding tendency

Consumers

Consumer social network

Small-world networkNumber of neighbour nodesRewiring probability

Feedback from consumers

Demand levelAverage green degree weighted by salesSales volumeMarket share

ConsumersrsquoInteraction

Word-of-mouthImitationHerding effect

Products

PriceGreen degree

ManufacturersProductCostProfitSelf-learning abilitySwarm learning ability

Manufacturersrsquointeraction

CompetitionMutual learning

Figure 1 The system structure diagram

Starting from a ring lattice with 119873 vertices and 119870 edges pervertex each edge is rewired at random with probability 119875119903This construction produces graphs between regularity (119875119903 =0) and disorder (119875119903 = 1) such that 119875119903 in the intermediateregion 0 lt 119875119903 lt 1 produces some degree of both Specificallyit is required that 119873 ge 119870 ge ln119873 ge 1 where 119870 ge ln119873guarantees that a random graph will be connected Henceconsumers will be affected by the characteristics of the small-world network such as the number of neighbor nodes 119870which can characterize the number of consumers interactingwith one consumer in making purchase decisions and therewiring probability 119875119903 which represents the probabilitythat the consumerrsquos neighbors will change their purchasedecisions

322 Consumer Purchase Decision-Making The manufac-turers and consumers can be denoted as 119894 and 119895 and theirnumbers are 119872 and 119873 respectively 119880119895119894 is the utility ofthe product 119894 for the consumer 119895 (119895 isin (1 sdot sdot sdot 119873)) whichcan indicate the consumersrsquo motivation and willingness topurchase products The consumers make purchase decisionsaccording to the utility of the product The motivationalfunction proposed by Zhang and Zhang (2007) [39] is used asthe decision basis for consumers to make purchase decisionsThe utility 119880119895119894 is shown in

119880119895119894 = 120583119895119894 lowast 119901119895119894 + 120588119895119894 lowast 119892119895119894 (1)

Theutility119880119895119894 of the product 119894 for the consumer 119895 ismainlyinfluenced by two factors the price and quality of the productIn (1) 120583119895119894 represents the price sensitivity of consumer 119895 tothe product 119894 Price sensitivity which shows the consumerrsquos

cognition and perception of the goodsrsquo value includingpractical and emotional values is one of the attributes ofconsumers and varies from consumer to consumerThe pricesensitivity distribution model [40] suggests that a consumerrsquosprice sensitivity is an exponential function of the differencebetween the real price of a product and the expected price ofthe product as shown in (2) where 120582119895 gt 1 and 119901(119890)119895119894 is theconsumer 119895rsquos expected price of the product 119894

120583j119894 = 1

1 + 120582119895(119875119895119894 minus119875(119890)

119895119894 )

(2)

On the other side 120588119895119894 is the green degree sensitivity ofconsumer 119895 to the product 119894 In thismodel the greendegree ofthe product refers to the general impact from multiple greenattributes of the product while multiple green attributesare equivalent to qualities of the product at a certain leveltherefore the sensitivity of the consumer to the productrsquosgreen degree can be regarded as the sensitivity to the productqualityThe outlier avoidance consumer psychological theory[41] suggests that the more closely the quality of the productapproximates the consumerrsquos expected quality of this kind ofproduct the more sensitive the consumer is to the qualityof the product The quality sensitivity 120588119895119894 is as shown in (3)where 0 lt 120573119895 lt 1 and 119892(119890)119895119894 is consumer 119895rsquos expected greendegree of the product 119894

120588119895119894 = 120573119895|119892119895

119894 minus119892(119890)119895

119894 | (3)

In reality consumers are embedded in the correspondingsocial networks therefore their purchase decisions for thesame kind of products with similar utilities are not only

Complexity 5

influenced by the price and green degree of products butare also affected by other consumers in the network such asfriendsrsquo recommendation and word of mouth among othersBesides whenmaking purchase decisions consumers show acertain degree of herdmentality [42] Herding behavior refersto the phenomenon where an agentrsquos decision-making in themarket is influenced by what others around him are doing[43] When consumer 119895 interacts with other consumers theutility of the product for himself will change correspondinglyunder the influence of the herd mentality The changing rulesof product utility for consumer 119895 under the influence of herdeffects are shown in formula (4) as follows

119867119864119895 = 120579119895 lowast infl119895119897 (4)

where 119867119864119895 refers to the herd effect on consumer 119895 andthe parameter 120579119895 which is subject to uniform distributiondenotes the herd tendency of consumer 119895The closer the valueof 120579119895 is to 0 the less likely a consumer is to be influencedby the surrounding people and the weaker the herd effect iswhile the larger the value of 120579119895 the more likely a consumer isto be influenced by the surrounding people and the strongerthe herd effect is The parameter infl119895119897 which denotes theinfluence of other consumers in the consumer interactionnetwork as perceived by consumer 119895 can be obtained fromthe average utility of product 119894 for neighboring consumersas shown in formula (5) where ℎ is the number of neighborswho interact with consumer 119895 in the consumer interactionnetwork and 119880119894119895119897 is the utility of product 119894 for the neighboringconsumer 119897 (119897 isin (1 sdot sdot sdot ℎ))

infl119895119897 =sumℎ119897=1119880119894119895119897

ℎ(5)

Accordingly on the basis of formulas (1) to (5) theproductrsquos utility can be further computed as shown in

119880119895119894 = 11 + 120582119895(119901

119895119894 minus119901(119890)

119895119894 )

lowast 119901119895119894 + 120573119895|119892119895119894minus119892(119890)

119895119894 | lowast 119892119895119894 + 120579119895 lowast infl119895119897 (6)

According to related literature [44] there is a positivecorrelation between consumerrsquos expected green degrees ofthe product and hisher income that is a consumer with ahigher income places more emphasis on the environmentalperformance of the product and hisher expected greendegree of the product is much higherTherefore a consumerrsquosexpected green degree of the product is shown in formula(7) where 119894119899119888119900119898119890119895 is the income of consumer 119895 119898 is theregression coefficient between a consumerrsquos expected greendegree and hisher income and 119898 gt 0 119886 is a scalingconstant indicating the fixed expectation for the green degreeof the product when a consumerrsquos income is 0 and 120596119895 is thestochastic disturbance drawn from a uniform distributionindicating the deviation of a consumerrsquos expected greendegree of the product

119892(119890)119895119894 = 119898 lowast 119894119899119888119900119898119890119895 + 119886 + 120596119895 (7)

In addition it is assumed that consumer 119895 can estimatethe expected price 119901(119890)119895119894 of product 119894 based on hisher own

expected green degree 119892(119890)119895119894 and 119901(119890)119895119894 increases with 119892(119890)119895119894 Moreover following Eppstein et al (2011) [45] it is alsoassumed that if the product price exceeds 100 of theexpected price consumers will consider the product notaffordable and refuse to purchase it Therefore the utilityfunction of consumersrsquo evaluation of different products isshown in formula (8) below

119880119895119894 = 11 + 120582119895(119901

119895

119894 minus119901(119890)119895

119894 )lowast 119901119895119894 + 120573119895|119892

119895

119894minus119898lowast119894119899119888119900119898119890119894minus120596119895| lowast 119892119895119894

+ 120579119895 lowast infl119895119897

(8)

To obtain the maximum utility consumers compare theutilities of products from all manufacturers in the market andfollow the rules below to make purchase decisions product 1will be purchased when 1198801198951 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product2 will be purchased when 1198801198952 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product119872 will be purchased when 119880119895119872 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 andthe product is purchased randomly when 1198801198951 = 1198801198952 = sdot sdot sdot =119880119895119872

33 Manufacturersrsquo Agents

331 Manufacturersrsquo Profits In this study there are 119872competing manufacturers in the market It is assumed thatproducts from different manufacturers are similar in theirfunction and performance but differ in price and greendegrees therefore manufacturers compete based on the priceand green degrees of the products

The production process of products with higher greendegrees may be more complex and require higher levels oftechnology therefore it is inevitable for the manufacturer toinvest additional costs in certain aspects of the productionto improve the green degree of products The unit costfunction of green products can be separated into two partsone is the fixed regular unit production cost 119888119894 the other isthe additional margin cost 1198881015840119894 Based on relevant literature[16] the additional margin cost is a quadratic function withthe product green degree as such 119888119894 can be set as 1198881015840119894 =(12)1199031198941198921198942 where 119903119894 represents the cost factor related to thegreen production efforts and 119903119894 gt 0 Therefore the unit costfunction of green products is shown in formula (9) below

119888119894 = 119888119894 + 121199031198941198921198942 (9)

With regard to product pricing it is assumed that themanufacturer applies the cost-plus pricing method that isthe price of the product 119894 is set as 119901119894 = (1 + 119900119894)119888119894 where119900119894 gt 0 represents the mark-up on the product Accordinglythe product price 119901119894 is shown in formula (10) below

119901119895119894 = (1 + 119900119894) 119888119894 = (1 + 119900119894) (119888119894 + 121199031198941198921198942) (10)

6 Complexity

Moreover we set the parameter 119901119888119895119894 to indicate whetherconsumer 119895 purchases the product 119894 and if the answer is yes119901119888119895119894 = 1 otherwise 119901119888119895119894 = 0 then the sales volume of product119894 can be calculated as in formula (11) below where 119902119894 is thesales volume of product 119894

119902119894 =119873

sum119895=1

119901119888119895119894 (11)

Thus the profit of the manufacturer is shown in formula(12) below

120587119894 = (119901119895119894 minus 119888119895119894 ) 119902119894 = (119901119895119894 minus 119888119895119894 )119873

sum119895=1

119901119888119895119894 (12)

332Manufacturersrsquo Decision-Making Manufacturers adjustthe green degrees of their products based on their profits Tobetter analyze the evolution of manufacturersrsquo product greendegree decision-making this study assumes that there are notechnical barriers and production performance limitationsfor manufacturers indicating manufacturers can adjust thegreen degree of their products based on their profits withouthaving to bear additional costs from production conversionMoreover the algorithm PSO [15] is applied to simulate themanufacturersrsquo productsrsquo green degree adjustment behaviorso that the mutual imitation and learning among manu-facturers can be better considered A manufacturer learnsand imitates from the ldquobenchmark manufacturerrdquo with thebest profit in the market Manufacturer 119894 determines thesubsequent green degree of its products based on its cur-rent product green degree (119892119895119894 (119905)) the historical productgreen degree (119901119887119890119904119905119895119894 ) that resulted in its best profit thusfar and the current product green degree of the ldquobench-mark manufacturerrdquo in the market (119892119887119890119904119905119895119894 ) The productgreen degree of manufacturer 119894 is thus updated by thefollowing

Step 1 Initialize the manufacturer parameters that is set-ting the number of manufacturers 119872 and the initialproduct green degree 119892119895119894 (119905) randomly generating velocityV119894(119905)Step 2 Calculate each manufacturerrsquos current profit 120587119894(119905)under its current product green degree 119892119895119894 (119905)Step 3 If the manufacturerrsquos current profit 120587119894(119905) is largerthan the manufacturerrsquos previous best profit 120587119894119901 then the120587119894119901 is replaced with the 120587119894(119905) and the product green degreecorresponding to the previous best profit 119901119887119890119904119905119895119894 is updated tothe current green degree 119892119895119894 (119905)Step 4 It is assumed that the manufacturer can learn fromthe ldquobenchmarkmanufacturerrdquoTheprofit of the ldquobenchmarkmanufacturerrdquo in the current market is recorded as 120587119887119890119904119905which can be shown as 120587119887119890119904119905 = max(120587119894(119905)) and the productgreen degree of the ldquobenchmark manufacturerrdquo is updated to119892119887119890119904119905119895119894

Step 5 Thevelocity and the product greendegree are updatedas shown in formulas (13) and (14) below

V119894 (119905 + 1) = 1198880V119894 (119905) + 11988811199031198861198991198891 (119901119887119890119904119905119895119894 minus 119892119895119894 (119905))+ 11988821199031198861198991198892 (119892119887119890119904119905119895119894 minus 119892119895119894 (119905))

(13)

119892119895119894 (119905 + 1) = 119892119895119894 (119905) + V119894 (119905 + 1) (14)In formula (13) 1198880 is the weight coefficient whose value

set in the range of [09 12] and enables the PSO algorithmto have good convergence based on relevant literature [46]therefore we set 1198880 = 1 1198881 is the manufacturerrsquos learning coef-ficient which reflects its self-learning ability from its history1198882 is the manufacturerrsquos swarm cognition coefficient whichindicates the degree of collaboration between manufacturersand the degree of themanufacturerrsquos acceptance of the grouprsquoscommon information [47] and 1199031198861198991198891 1199031198861198991198892 are randomnumbers drawn from a uniform distribution in the range(01)

Step 6 Return to Step 2 until all manufacturers have beenlooped through

4 Simulation Scenarios Design

We collected information about manufacturers and annualsales of split air conditioners from Taobao and JD andreferred to the ratio between the number of manufacturersand their annual sales We assumed that there are 10 man-ufacturers and 10000 consumers in the market and theirdecision-making processes and interactions are shown inFigure 2 In the initial period of simulation manufacturerssell their products with different prices and green degreesThe consumers calculate the utilities of all products thatmeet their purchasing power and make purchase decisionsbased on the principle of maximum utility After consumerspurchase products manufacturers are able to obtain theirown competitive performance indicators such as sales andprofit and they seek out the ldquobenchmark manufacturerrdquo bycomparing the profit of other manufacturers Subsequentlythey adjust the productsrsquo green degree and prices in thenext business cycle based on the PSO algorithm and otherinfluencing factors

Therefore we keep the basic parameters of the modelunchanged and change 119868119899119888119900119898119890119895 in three different scenariosas well as the parameter 119868119899119888119900119898119890119895 values in three ranges119873(400 75)119873(500 75) and119873(600 75) correspondinglyThevariances of consumer income levels are kept constant andthemeans of consumer income levels are different in the threescenarios

To reduce randomness in the simulations and improvethe statistical stability and validity of simulation results eachsimulation scenario is ran 10 times and the total numberof ticks per simulation scenario 119879max is set to 150 theaverage results of which is then statistically analyzed Theinitial value-setting of the basic model parameters is shownin Table 1

To analyze the manufacturersrsquo decision-making concern-ing their productsrsquo green degree in a more detailed manner

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 5

influenced by the price and green degree of products butare also affected by other consumers in the network such asfriendsrsquo recommendation and word of mouth among othersBesides whenmaking purchase decisions consumers show acertain degree of herdmentality [42] Herding behavior refersto the phenomenon where an agentrsquos decision-making in themarket is influenced by what others around him are doing[43] When consumer 119895 interacts with other consumers theutility of the product for himself will change correspondinglyunder the influence of the herd mentality The changing rulesof product utility for consumer 119895 under the influence of herdeffects are shown in formula (4) as follows

119867119864119895 = 120579119895 lowast infl119895119897 (4)

where 119867119864119895 refers to the herd effect on consumer 119895 andthe parameter 120579119895 which is subject to uniform distributiondenotes the herd tendency of consumer 119895The closer the valueof 120579119895 is to 0 the less likely a consumer is to be influencedby the surrounding people and the weaker the herd effect iswhile the larger the value of 120579119895 the more likely a consumer isto be influenced by the surrounding people and the strongerthe herd effect is The parameter infl119895119897 which denotes theinfluence of other consumers in the consumer interactionnetwork as perceived by consumer 119895 can be obtained fromthe average utility of product 119894 for neighboring consumersas shown in formula (5) where ℎ is the number of neighborswho interact with consumer 119895 in the consumer interactionnetwork and 119880119894119895119897 is the utility of product 119894 for the neighboringconsumer 119897 (119897 isin (1 sdot sdot sdot ℎ))

infl119895119897 =sumℎ119897=1119880119894119895119897

ℎ(5)

Accordingly on the basis of formulas (1) to (5) theproductrsquos utility can be further computed as shown in

119880119895119894 = 11 + 120582119895(119901

119895119894 minus119901(119890)

119895119894 )

lowast 119901119895119894 + 120573119895|119892119895119894minus119892(119890)

119895119894 | lowast 119892119895119894 + 120579119895 lowast infl119895119897 (6)

According to related literature [44] there is a positivecorrelation between consumerrsquos expected green degrees ofthe product and hisher income that is a consumer with ahigher income places more emphasis on the environmentalperformance of the product and hisher expected greendegree of the product is much higherTherefore a consumerrsquosexpected green degree of the product is shown in formula(7) where 119894119899119888119900119898119890119895 is the income of consumer 119895 119898 is theregression coefficient between a consumerrsquos expected greendegree and hisher income and 119898 gt 0 119886 is a scalingconstant indicating the fixed expectation for the green degreeof the product when a consumerrsquos income is 0 and 120596119895 is thestochastic disturbance drawn from a uniform distributionindicating the deviation of a consumerrsquos expected greendegree of the product

119892(119890)119895119894 = 119898 lowast 119894119899119888119900119898119890119895 + 119886 + 120596119895 (7)

In addition it is assumed that consumer 119895 can estimatethe expected price 119901(119890)119895119894 of product 119894 based on hisher own

expected green degree 119892(119890)119895119894 and 119901(119890)119895119894 increases with 119892(119890)119895119894 Moreover following Eppstein et al (2011) [45] it is alsoassumed that if the product price exceeds 100 of theexpected price consumers will consider the product notaffordable and refuse to purchase it Therefore the utilityfunction of consumersrsquo evaluation of different products isshown in formula (8) below

119880119895119894 = 11 + 120582119895(119901

119895

119894 minus119901(119890)119895

119894 )lowast 119901119895119894 + 120573119895|119892

119895

119894minus119898lowast119894119899119888119900119898119890119894minus120596119895| lowast 119892119895119894

+ 120579119895 lowast infl119895119897

(8)

To obtain the maximum utility consumers compare theutilities of products from all manufacturers in the market andfollow the rules below to make purchase decisions product 1will be purchased when 1198801198951 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product2 will be purchased when 1198801198952 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 product119872 will be purchased when 119880119895119872 = max1198801198951 1198801198952 sdot sdot sdot 119880119895119872 andthe product is purchased randomly when 1198801198951 = 1198801198952 = sdot sdot sdot =119880119895119872

33 Manufacturersrsquo Agents

331 Manufacturersrsquo Profits In this study there are 119872competing manufacturers in the market It is assumed thatproducts from different manufacturers are similar in theirfunction and performance but differ in price and greendegrees therefore manufacturers compete based on the priceand green degrees of the products

The production process of products with higher greendegrees may be more complex and require higher levels oftechnology therefore it is inevitable for the manufacturer toinvest additional costs in certain aspects of the productionto improve the green degree of products The unit costfunction of green products can be separated into two partsone is the fixed regular unit production cost 119888119894 the other isthe additional margin cost 1198881015840119894 Based on relevant literature[16] the additional margin cost is a quadratic function withthe product green degree as such 119888119894 can be set as 1198881015840119894 =(12)1199031198941198921198942 where 119903119894 represents the cost factor related to thegreen production efforts and 119903119894 gt 0 Therefore the unit costfunction of green products is shown in formula (9) below

119888119894 = 119888119894 + 121199031198941198921198942 (9)

With regard to product pricing it is assumed that themanufacturer applies the cost-plus pricing method that isthe price of the product 119894 is set as 119901119894 = (1 + 119900119894)119888119894 where119900119894 gt 0 represents the mark-up on the product Accordinglythe product price 119901119894 is shown in formula (10) below

119901119895119894 = (1 + 119900119894) 119888119894 = (1 + 119900119894) (119888119894 + 121199031198941198921198942) (10)

6 Complexity

Moreover we set the parameter 119901119888119895119894 to indicate whetherconsumer 119895 purchases the product 119894 and if the answer is yes119901119888119895119894 = 1 otherwise 119901119888119895119894 = 0 then the sales volume of product119894 can be calculated as in formula (11) below where 119902119894 is thesales volume of product 119894

119902119894 =119873

sum119895=1

119901119888119895119894 (11)

Thus the profit of the manufacturer is shown in formula(12) below

120587119894 = (119901119895119894 minus 119888119895119894 ) 119902119894 = (119901119895119894 minus 119888119895119894 )119873

sum119895=1

119901119888119895119894 (12)

332Manufacturersrsquo Decision-Making Manufacturers adjustthe green degrees of their products based on their profits Tobetter analyze the evolution of manufacturersrsquo product greendegree decision-making this study assumes that there are notechnical barriers and production performance limitationsfor manufacturers indicating manufacturers can adjust thegreen degree of their products based on their profits withouthaving to bear additional costs from production conversionMoreover the algorithm PSO [15] is applied to simulate themanufacturersrsquo productsrsquo green degree adjustment behaviorso that the mutual imitation and learning among manu-facturers can be better considered A manufacturer learnsand imitates from the ldquobenchmark manufacturerrdquo with thebest profit in the market Manufacturer 119894 determines thesubsequent green degree of its products based on its cur-rent product green degree (119892119895119894 (119905)) the historical productgreen degree (119901119887119890119904119905119895119894 ) that resulted in its best profit thusfar and the current product green degree of the ldquobench-mark manufacturerrdquo in the market (119892119887119890119904119905119895119894 ) The productgreen degree of manufacturer 119894 is thus updated by thefollowing

Step 1 Initialize the manufacturer parameters that is set-ting the number of manufacturers 119872 and the initialproduct green degree 119892119895119894 (119905) randomly generating velocityV119894(119905)Step 2 Calculate each manufacturerrsquos current profit 120587119894(119905)under its current product green degree 119892119895119894 (119905)Step 3 If the manufacturerrsquos current profit 120587119894(119905) is largerthan the manufacturerrsquos previous best profit 120587119894119901 then the120587119894119901 is replaced with the 120587119894(119905) and the product green degreecorresponding to the previous best profit 119901119887119890119904119905119895119894 is updated tothe current green degree 119892119895119894 (119905)Step 4 It is assumed that the manufacturer can learn fromthe ldquobenchmarkmanufacturerrdquoTheprofit of the ldquobenchmarkmanufacturerrdquo in the current market is recorded as 120587119887119890119904119905which can be shown as 120587119887119890119904119905 = max(120587119894(119905)) and the productgreen degree of the ldquobenchmark manufacturerrdquo is updated to119892119887119890119904119905119895119894

Step 5 Thevelocity and the product greendegree are updatedas shown in formulas (13) and (14) below

V119894 (119905 + 1) = 1198880V119894 (119905) + 11988811199031198861198991198891 (119901119887119890119904119905119895119894 minus 119892119895119894 (119905))+ 11988821199031198861198991198892 (119892119887119890119904119905119895119894 minus 119892119895119894 (119905))

(13)

119892119895119894 (119905 + 1) = 119892119895119894 (119905) + V119894 (119905 + 1) (14)In formula (13) 1198880 is the weight coefficient whose value

set in the range of [09 12] and enables the PSO algorithmto have good convergence based on relevant literature [46]therefore we set 1198880 = 1 1198881 is the manufacturerrsquos learning coef-ficient which reflects its self-learning ability from its history1198882 is the manufacturerrsquos swarm cognition coefficient whichindicates the degree of collaboration between manufacturersand the degree of themanufacturerrsquos acceptance of the grouprsquoscommon information [47] and 1199031198861198991198891 1199031198861198991198892 are randomnumbers drawn from a uniform distribution in the range(01)

Step 6 Return to Step 2 until all manufacturers have beenlooped through

4 Simulation Scenarios Design

We collected information about manufacturers and annualsales of split air conditioners from Taobao and JD andreferred to the ratio between the number of manufacturersand their annual sales We assumed that there are 10 man-ufacturers and 10000 consumers in the market and theirdecision-making processes and interactions are shown inFigure 2 In the initial period of simulation manufacturerssell their products with different prices and green degreesThe consumers calculate the utilities of all products thatmeet their purchasing power and make purchase decisionsbased on the principle of maximum utility After consumerspurchase products manufacturers are able to obtain theirown competitive performance indicators such as sales andprofit and they seek out the ldquobenchmark manufacturerrdquo bycomparing the profit of other manufacturers Subsequentlythey adjust the productsrsquo green degree and prices in thenext business cycle based on the PSO algorithm and otherinfluencing factors

Therefore we keep the basic parameters of the modelunchanged and change 119868119899119888119900119898119890119895 in three different scenariosas well as the parameter 119868119899119888119900119898119890119895 values in three ranges119873(400 75)119873(500 75) and119873(600 75) correspondinglyThevariances of consumer income levels are kept constant andthemeans of consumer income levels are different in the threescenarios

To reduce randomness in the simulations and improvethe statistical stability and validity of simulation results eachsimulation scenario is ran 10 times and the total numberof ticks per simulation scenario 119879max is set to 150 theaverage results of which is then statistically analyzed Theinitial value-setting of the basic model parameters is shownin Table 1

To analyze the manufacturersrsquo decision-making concern-ing their productsrsquo green degree in a more detailed manner

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

6 Complexity

Moreover we set the parameter 119901119888119895119894 to indicate whetherconsumer 119895 purchases the product 119894 and if the answer is yes119901119888119895119894 = 1 otherwise 119901119888119895119894 = 0 then the sales volume of product119894 can be calculated as in formula (11) below where 119902119894 is thesales volume of product 119894

119902119894 =119873

sum119895=1

119901119888119895119894 (11)

Thus the profit of the manufacturer is shown in formula(12) below

120587119894 = (119901119895119894 minus 119888119895119894 ) 119902119894 = (119901119895119894 minus 119888119895119894 )119873

sum119895=1

119901119888119895119894 (12)

332Manufacturersrsquo Decision-Making Manufacturers adjustthe green degrees of their products based on their profits Tobetter analyze the evolution of manufacturersrsquo product greendegree decision-making this study assumes that there are notechnical barriers and production performance limitationsfor manufacturers indicating manufacturers can adjust thegreen degree of their products based on their profits withouthaving to bear additional costs from production conversionMoreover the algorithm PSO [15] is applied to simulate themanufacturersrsquo productsrsquo green degree adjustment behaviorso that the mutual imitation and learning among manu-facturers can be better considered A manufacturer learnsand imitates from the ldquobenchmark manufacturerrdquo with thebest profit in the market Manufacturer 119894 determines thesubsequent green degree of its products based on its cur-rent product green degree (119892119895119894 (119905)) the historical productgreen degree (119901119887119890119904119905119895119894 ) that resulted in its best profit thusfar and the current product green degree of the ldquobench-mark manufacturerrdquo in the market (119892119887119890119904119905119895119894 ) The productgreen degree of manufacturer 119894 is thus updated by thefollowing

Step 1 Initialize the manufacturer parameters that is set-ting the number of manufacturers 119872 and the initialproduct green degree 119892119895119894 (119905) randomly generating velocityV119894(119905)Step 2 Calculate each manufacturerrsquos current profit 120587119894(119905)under its current product green degree 119892119895119894 (119905)Step 3 If the manufacturerrsquos current profit 120587119894(119905) is largerthan the manufacturerrsquos previous best profit 120587119894119901 then the120587119894119901 is replaced with the 120587119894(119905) and the product green degreecorresponding to the previous best profit 119901119887119890119904119905119895119894 is updated tothe current green degree 119892119895119894 (119905)Step 4 It is assumed that the manufacturer can learn fromthe ldquobenchmarkmanufacturerrdquoTheprofit of the ldquobenchmarkmanufacturerrdquo in the current market is recorded as 120587119887119890119904119905which can be shown as 120587119887119890119904119905 = max(120587119894(119905)) and the productgreen degree of the ldquobenchmark manufacturerrdquo is updated to119892119887119890119904119905119895119894

Step 5 Thevelocity and the product greendegree are updatedas shown in formulas (13) and (14) below

V119894 (119905 + 1) = 1198880V119894 (119905) + 11988811199031198861198991198891 (119901119887119890119904119905119895119894 minus 119892119895119894 (119905))+ 11988821199031198861198991198892 (119892119887119890119904119905119895119894 minus 119892119895119894 (119905))

(13)

119892119895119894 (119905 + 1) = 119892119895119894 (119905) + V119894 (119905 + 1) (14)In formula (13) 1198880 is the weight coefficient whose value

set in the range of [09 12] and enables the PSO algorithmto have good convergence based on relevant literature [46]therefore we set 1198880 = 1 1198881 is the manufacturerrsquos learning coef-ficient which reflects its self-learning ability from its history1198882 is the manufacturerrsquos swarm cognition coefficient whichindicates the degree of collaboration between manufacturersand the degree of themanufacturerrsquos acceptance of the grouprsquoscommon information [47] and 1199031198861198991198891 1199031198861198991198892 are randomnumbers drawn from a uniform distribution in the range(01)

Step 6 Return to Step 2 until all manufacturers have beenlooped through

4 Simulation Scenarios Design

We collected information about manufacturers and annualsales of split air conditioners from Taobao and JD andreferred to the ratio between the number of manufacturersand their annual sales We assumed that there are 10 man-ufacturers and 10000 consumers in the market and theirdecision-making processes and interactions are shown inFigure 2 In the initial period of simulation manufacturerssell their products with different prices and green degreesThe consumers calculate the utilities of all products thatmeet their purchasing power and make purchase decisionsbased on the principle of maximum utility After consumerspurchase products manufacturers are able to obtain theirown competitive performance indicators such as sales andprofit and they seek out the ldquobenchmark manufacturerrdquo bycomparing the profit of other manufacturers Subsequentlythey adjust the productsrsquo green degree and prices in thenext business cycle based on the PSO algorithm and otherinfluencing factors

Therefore we keep the basic parameters of the modelunchanged and change 119868119899119888119900119898119890119895 in three different scenariosas well as the parameter 119868119899119888119900119898119890119895 values in three ranges119873(400 75)119873(500 75) and119873(600 75) correspondinglyThevariances of consumer income levels are kept constant andthemeans of consumer income levels are different in the threescenarios

To reduce randomness in the simulations and improvethe statistical stability and validity of simulation results eachsimulation scenario is ran 10 times and the total numberof ticks per simulation scenario 119879max is set to 150 theaverage results of which is then statistically analyzed Theinitial value-setting of the basic model parameters is shownin Table 1

To analyze the manufacturersrsquo decision-making concern-ing their productsrsquo green degree in a more detailed manner

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 7

Start

Store the consumer

purchasing data

Manufacturer adjusts its product green degree - Particle

Swarm Optimization algorithm

Update the product green degree production cost and

product price

Store the manufacturerrsquo s

data

Products are affordable

Consumer refuses to purchase

Y

Y

N

Load consumer agents assign their characteristics

Load manufacturer agents assign their initial product

green degree

Tick=1

Consumer calculates the product utility

Consumer purchases product that best maximizes his utility

Manufacturer calculates its product sales volume and

profit

Tick gt TmaxTick=Tick+1N

Finish

Figure 2 Agent interaction flow chart

Table 1 The initial value-setting of the basic parameters of the model

Parameters Ranges of values Distributions ReferencesDescriptions119873 10000 Constant Refer to the information from Taobao and JD119872 10 Constant Refer to the information from Taobao and JD120582119895 U(120) Uniform distribution Zhang and Zhang (2007) [39]120573119895 U(0406) Uniform distribution Zhang and Zhang (2007) [39]

119898 0067 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

119886 -1354 Constant Eppstein et al (2011) [45]Adepetu et al (2016) [44]

120579119895 U(0001) Uniform distribution Zhang and Zhang (2007) [39]119875119903 02 Constant McCoy and Lyons (2014) [48]119870 4 Constant McCoy and Lyons (2014) [48]119888119894 5 Constant Refer to the manufacturersrsquo interview results119903119894 002 Constant Liu et al (2012) [16]119900119894 01 Constant Refer to the manufacturersrsquo interview results1198880 1 Constant Shi and Eberhart (1998) [46]1198881 2 Constant Kennedy (1997) [47]1198882 2 Constant Kennedy (1997) [47]

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

8 Complexity

Table 2 Descriptions of manufacturersrsquo categories

Parameters Descriptions

119872ℎ Manufacturers whose initial product green degree is above 50 called the manufacturer with high initial green degree(HIGD)

1198981 Number of119872ℎ1198981 lt 119872ℎ119872119898 Manufacturers whose initial product green degree is within the range of [30 50) called the manufacturer with medium

initial green degree (MIGD)1198982 Number of 1198721198981198982 lt 119872119872119897 Manufacturers whose initial product green degree is within the range of [15 30) called the manufacturer with low initial

green degree (LIGD)1198983 Number of 119872119897 1198993 lt 119872

Table 3 Descriptions of indicators

Indicators Denotations DescriptionsGreen degree of HIGD sum119872ℎ 1198921198941198981 The average product green degree of manufacturers with HIGDGreen degree of MIGD sum119872119898 1198921198941198982 The average product green degree of manufacturers with MIGDGreen degree of LIGD sum119872119897 1198921198941198983 The average product green degree of manufacturers with LIGDManufacturersrsquo green supply sum119872 119892119894119872 The average green degree of products made by all manufacturers in the marketConsumersrsquo green demand sum119872 119892119894 sdot 119902119894sum119872 119902119894 The average green degree of products that are actually purchased by consumersSales of HIGD sum119872ℎ 1198921198941198981 The average product sales volume of manufacturers with HIGDSales of MIGD sum119872119898 1199021198941198982 The average product sales volume of manufacturers with MIGDSales of LIGD sum119872119897 1199021198941198983 The average product sales volume of manufacturers with LIGDHighest market share 119902119894maxsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the largestLowest market share 119902119894minsum119872 119902119894 Themarket share of the manufacturer whose sales volume is the smallestProfit of HIGD sum119872ℎ 1205871198941198981 The average profit of manufacturers with HIGDProfit of MIGD sum119872119898 1205871198941198982 The average profit of manufacturers with MIGDProfit of LIGD sum119872119897 1205871198941198983 The average profit of manufacturers with LIGDProfit of market sum119872 120587119894119872 The average profit of all manufacturers in the market

this study divided the manufacturers into three categoriesbased on the initial green degree of their products (similar toChinarsquos split air conditionersrsquo green degree as described in theintroduction) Classifying manufacturers into different typesbased on their initial product green degree helps us analyzethe macroevolution trends of the manufacturing industryrsquosgreen degree the competition performance of different greendegrees of products and the consumersrsquo influences underdifferent scenarios The three categories of manufacturers aredenoted and described in Table 2 where119872 = 1198981 + 1198982 + 1198983

The included indicators micro and macro are designedto reveal consumersrsquo consumption choices manufacturersrsquogreen decision and market competition performance andevolutionary trend The green degree decision product salesand manufacturersrsquo profits are analyzed in different scenariosof consumer income All the indicators are described inTable 3 and all the indicators are studied in Section 5

5 Results Analysis

51 Green Degree Decisions of Manufacturers The greendegree evolutions of the three manufacturer categories are

shown in Figure 3 The consumersrsquo income is normallydistributed in N(400 75) N(500 75) and N(600 75) whichare shown in Figures 3(a) 3(b) and 3(c) respectively

During market competition the green degree of HIGDgradually reduces to a stable level while the green degreeof LIGD gradually rises to a stable level and the differencesamong manufacturers narrow to a certain level Howeverthe green degrees of HIGD MIGD and LIGD remain atthe highest middle and lowest levels respectively Moreoverthe manufacturersrsquo green degrees increase as the consumerincome levels rise for example the green degree curve ofHIGD is higher when consumer income level is at 600 ascompared to those of other income levels

In analyzing the green degree evolution we make threeobservations First through imitation and learning amongmanufacturers the manufacturing industry emerges show-ing a ldquoconvergencerdquo effect signifying smaller differencesamong manufacturers Second the differences always exist inthe manufacturing industry since consumersrsquo demands arediverse and manufacturers also rely on their own experi-ences to make decisions to attract more consumers Thirdthrough market competition manufacturers win their ownconsumers and the manufacturing industry enters a stable

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 9

0 30 60 90 120 150

20

25

30

35

40

45

50

55G

reen

deg

ree

Tick

Green degree of LIGD Green degree of MIGD Green degree of HIGD

(a) Mean consumer income=400

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

Green degree of LIGD Green degree of MIGD Green degree of HIGD

Tick20

25

30

35

40

45

50

55

Gre

en d

egre

e

(c) Mean consumer income=600

Figure 3 Green degree evolutions of the three manufacturer categories under different scenarios

state Therefore manufacturersrsquo green degree decisions keepsteady at a certain level and manufacturers produce productsthat fit their own consumers

In Figure 4 we illustrate the indicators of ldquomanufac-turersrsquo green supplyrdquo and ldquoconsumersrsquo green demandrdquo Theldquomanufacturersrsquo green supplyrdquo reflects the green degree ofproducts that are supplied bymanufacturersThe ldquoconsumersrsquogreen demandrdquo reflects the green degree of products that arepurchased by consumers

First by comparing Figures 4(a) 4(b) and 4(c) weobserve that at the initial stage of the system (Tick=0) con-sumersrsquo green demand is higher along with higher consumerincome Second consumersrsquo green demand shows an upwardtrend in all scenarios Third the difference between thetwo indicators narrows with market competition while themanufacturersrsquo green supply is higher than the consumersrsquogreen demand at the final stage of the systemrsquos operationWhen consumers have high income levels (mean consumerincome = 600) manufacturersrsquo green supply matches wellwith the consumersrsquo green demand

Thus it can be seen that during market competitionthe manufacturers may overestimate the consumersrsquo greendegree demand making the green degree of supplied prod-ucts higher than that of products purchased by consumersHowever manufacturers make decisions by imitating othermanufacturers and by learning from their experiences andthe supplied productsrsquo green degree gets adjusted to bettermatch the demandrsquos green degree so that the differencebetween the two indicators narrows

52 Sales of Manufacturers Figures 5(a) 5(b) and 5(c) showthe sales evolutions of the three manufacturer categoriesConsumer incomes are normally distributed in N(400 75)

N(500 75) and N(600 75) respectively As can be seenin Figure 5 when the consumer income level is higherthey display higher expectations of green degree whichlead to increasing product sales for MIGD and HIGD inthe three scenarios Higher consumer income favors greenproduction consequently manufacturers with higher greendegrees become more competitive

In the initial stage of product competition (Tick 0-30)combined with the analysis on Figure 3 we observe that eventhough HIGDs reduce the green degree of their productsthey still maintain the competitiveness of their productsrsquogreen degree and simultaneously their product prices alsobecome increasingly competitive leading to an increase intheir sales as shown in Figure 5

During competition we can see from Figure 3 that LIGDskeep improving the green degree of their products such thatthe difference in their productsrsquo green degree against that ofHIGDs and MIGDs continuously decreases thus increasingthe competitiveness of LIGDs while simultaneously main-taining the attractiveness of their productsrsquo prices As a resultwe can see that LIGDs can achieve the highest sales inFigure 5(a) and their sales continue to rise as shown inFigures5(b) and 5(c)

MIGDs change their productsrsquo green degree slightly asshown in Figure 3 but HIGDs and LIGDs become increas-ingly competitive Moreover consumers constantly evaluateproducts in the market and make their best purchasingdecisions leading to some of MIGDrsquos market share beingseized by LIGD and HIGD therefore the product sales ofMIGDs show a downward trend in Figure 5 The sales ofMIGDs become steady when the green degrees of MIGDsenter a stable state (after tick 30)

The product market concentration is signified by indi-cators ldquohighest market sharerdquo and ldquolowest market sharerdquo

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

10 Complexity

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

26

28

30

32

34

36

38G

reen

Deg

ree

(a) Mean consumer income=400

0 30 60 90 120 150

Tick

Manufacturers green supply Consumers green demand

32

34

36

38

Gre

en D

egre

e

(b) Mean consumer income=500

0 30 60 90 120 150

38

40

42

44

Manufacturers green supply Consumers green demand

Tick

Gre

en D

egre

e

(c) Mean consumer income=600

Figure 4 Green degree of manufacturersrsquo supply and consumersrsquo demand under different scenarios

0 30 60 90 120 150

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

0

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

(a) Mean consumer income=400

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sale

s vol

ume

Tick

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

(b) Mean consumer income=500

0 30 60 90 120 1500

250

500

750

1000

1250

1500

1750

2000

2250

Sales volume of LIGD Sales volume of MIGD Sales volume of HIGD

Sale

s vol

ume

Tick

(c) Means of consumer income=600

Figure 5 Sales evolution of three manufacturer categories under different scenarios

The manufacturersrsquo competition results (whether the marketis evenly distributed or dominated) can be analyzed underdifferent scenarios The consumer incomes are normallydistributed in N(400 75) N(500 75) and N(600 75) asshown in Figures 6(a) 6(b) and 6(c) respectively

The product concentration degree reflects the differencein competitiveness among manufacturers in the market Ascan be seen in Figure 6 when the mean consumer income is400 the highest market share increases to more than 40while the lowest market share is minimal indicating thatthe market is gradually dominated by the manufacturer withhigher sales whose competitive advantage is constantly beingenhanced When the mean consumer income is 500 thehighestmarket share gradually stabilizes at around 25while

the lowestmarket share gradually reduces to about 2Whenthe mean consumer income is 600 the highest market sharegradually stabilizes at around 18 while the lowest marketshare gradually increases to about 4

Therefore it can be concluded that it is much harder formanufacturers to gain monopolistic advantage in a marketwhere consumers have higher income As the purchasingpower of consumers constantly increases the sales of man-ufacturers become more easily balanced such that the marketshare is evenly distributed instead of being a monopoly

53 Profits of Manufacturers The profit evolutions of thethree categories of manufacturers are shown in Figure 7 theconsumers income are normally distributed in N(400 75)

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 11

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40M

arke

t con

cent

ratio

n de

gree

(a) Mean consumer income=400

0 30 60 90 120 150Tick

Highest market share Lowest market share

0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(b) Mean consumer income=500

0 30 60 90 120 150

Highest market share Lowest market share

Tick0

5

10

15

20

25

30

35

40

Mar

ket c

once

ntra

tion

degr

ee

(c) Means of consumer income=600

Figure 6 Highest and lowest market shares under different scenarios

0 30 60 90 120 150

Tick

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(a) Mean consumer income=400

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(b) Mean consumer income=500

0 30 60 90 120 150

Profit of LIGD Profit of MIGD Profit of HIGD Market profit

Tick0

600

1200

1800

2400

3000

3600

4200

Aver

age p

rofit

(c) Mean consumer income=600

Figure 7 Profit evolution of manufacturers under different scenarios

N(500 75) and N(600 75) as shown in Figures 7(a) 7(b)and 7(c) respectively

Themanufacturersrsquo profits are influenced by the productsrsquosales and green degrees From Figures 3 5 and 7 weobserve that the profits evolution of the three categories ofmanufacturers is similar to the sales evolution in Figure 5under the same scenario However the difference in profitsis not the same as the difference in sales among the threecategories

As can be seen from Figure 7(b) similar to the productsales of the three categories the profit of HIGD initially rises

and then decreases The profit of MIGD initially decreasesand then stabilizes while the profit of LIGD initially risesand then stabilizes During market competition the profit ofMIGD is slightly higher than that of LIGD and the profitsof both are higher than the average market profit indicatingthat although the sales of MIGD are lower than that of LIGDMIGD can still achieve higher profits due to the higherproduct price As can be seen from Figure 7(c) even if theproduct sale of LIGD exceeds that of HIGD the profit ofLIGD is far lower than that of HIGD and MIGD Moreoverwhile the product sale of MIGD is much higher than that

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

12 Complexity

of HIGD the profit of HIGD can surpass that of MIGDin some periods of the market competition Furthermorea comparison of Figures 7(a) 7(b) and 7(c) shows that asthe income and purchasing power of consumers increasethe profits of MIGD and HIGD also increase The profitof LIGD is the lowest when the mean consumer incomeis 600

The above analysis indicates that when consumer incomeis low the profit of manufacturers is mainly obtained fromthe product sales With an increase in consumer incomethe profit of manufacturers gradually gets influenced by theproductsrsquo green degree The higher the green degree of aproduct the higher the unit profit brought by the productthus contributing a larger profit margin for the manufacturerIn summary the income of consumers is one of the mostsignificant factors influencing the profits of different cat-egories of manufacturers Therefore manufacturers shouldtakemeasures to get information about the actual income andgreen preferences of consumers in the market before makingproduct green degree decisions

6 Conclusions

To help manufacturers make reasonable green decisions andgain competitive advantage it is important to clarify theevolution of productsrsquo green degrees and its impact on thecompetitive performance of manufacturers under differentconsumer income scenarios In this light this study builtan agent-based model considered complex characteristics ofagents and used the swarm intelligence algorithm to describethe decision-making behavior of manufacturers concerningthe green degree of their products By fully considering theinteractions among consumers andmanufacturers this studypresented a visual analysis of the evolution of manufactur-ersrsquo decision-making on their productsrsquo green degree basedon different categories of manufacturers This study alsoanalyzed the macrobehaviors of manufacturers consumersand products based onmulti-dimensional performance indi-cators and revealed the micromechanisms affecting themacroemergence of markets

The results show that first through imitation and learningamong manufacturers the manufacturing industry emergesshowing a ldquoconvergencerdquo effect such that the differences inmanufacturersrsquo productsrsquo green degrees become smaller butalways exist Moreover the manufacturers may overestimatethe consumersrsquo green degree demand making the greendegree of supplied products higher than that of productspurchased by consumers but this gradually gets correctedthrough the mechanisms of market competition In addi-tion as consumer income increases it becomes easier formanufacturers to adapt to the marketrsquos supply and demandas impacted by the productsrsquo green degrees and it becomesunfavorable for them to form a monopoly in the marketFurthermore when consumer income is low the profitsof manufacturers mainly come from the sales howeverwith an increase in consumer income the profits of man-ufacturers are gradually influenced by the productsrsquo greendegree

Data Availability

Data used in the current study is available to other researchersin order to advance the science and profession Data gen-erated or analyzed during the study are available from thecorresponding author by request

Conflicts of Interest

Nopotential conflicts of interestwere reported by the authors

Acknowledgments

This work was supported by the National Natural ScienceFoundation of China (Nos 71671078 71501084) The KeyProject of Philosophy and Social Science Research in Col-leges and Universities in Jiangsu Province (2018SJZDI052)sponsored by Qing Lan Project of Jiangsu Province YouthBackbone Teacher Training Project of Jiangsu UniversityThe National Social Science Fund of China (18VSJ03817AGL010) Humanities and Social Sciences by the Ministryof Education of China (16YJC630133)

References

[1] P Nimse A Vijayan A Kumar and C Varadarajan ldquoAreview of green product databasesrdquo Environmental Progress ampSustainable Energy vol 26 no 2 pp 131ndash137 2007

[2] W Zhu and Y He ldquoGreen product design in supply chainsunder competitionrdquo European Journal of Operational Researchvol 258 no 1 pp 165ndash180 2017

[3] Y-S Chen ldquoThe positive effect of green intellectual capital oncompetitive advantages of firmsrdquo Journal of Business Ethics vol77 no 3 pp 271ndash286 2008

[4] Y Tian K Govindan and Q Zhu ldquoA system dynamics modelbased on evolutionary game theory for green supply chainmanagement diffusion among Chinese manufacturersrdquo Journalof Cleaner Production vol 80 pp 96ndash105 2014

[5] Y Yu X Han andG Hu ldquoOptimal production for manufactur-ers considering consumer environmental awareness and greensubsidiesrdquo International Journal of Production Economics vol182 pp 397ndash408 2016

[6] AHuttel F ZiesemerMPeyer and I Balderjahn ldquoTopurchaseor not Why consumers make economically (non-)sustainableconsumption choicesrdquo Journal of Cleaner Production vol 174pp 827ndash836 2018

[7] S Yang and D Zhao ldquoDo subsidies work better in low-incomethan in high-income families Survey on domestic energy-efficient and renewable energy equipment purchase in ChinardquoJournal of Cleaner Production vol 108 pp 841ndash851 2015

[8] E Vasileiou and N Georgantzıs ldquoAn experiment on energy-saving competition with socially responsible consumers open-ing the black boxrdquo Journal of Behavioral and ExperimentalEconomics vol 58 pp 1ndash10 2015

[9] C Park H Lee J Jun and T Lee ldquoTwo-sided effectsof customer participation roles of relationships and social-interaction values in social servicesrdquo Service Business vol 12no 3 pp 621ndash640 2018

[10] E Moser A Seidl and G Feichtinger ldquoHistory-dependencein production-pollution-trade-off models a multi-stage

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Complexity 13

approachrdquo Annals of Operations Research vol 222 no 1 pp457ndash481 2014

[11] Y Teng X Li P Wu and X Wang ldquoUsing Cooperative GameTheory to Determine Profit Distribution in IPD ProjectsrdquoInternational Journal of Construction Management vol 19 no1 pp 32ndash45 2019

[12] Z Li X Lv H Zhu and Z Sheng ldquoAnalysis of complexityof unsafe behavior in construction teams and a multiagentsimulationrdquo Complexity vol 2018 Article ID 6568719 15 pages2018

[13] Q Meng Z Li H Liu and J Chen ldquoAgent-based simulation ofcompetitive performance for supply chains based on combinedcontractsrdquo International Journal of Production Economics vol193 pp 663ndash676 2017

[14] BWuWHuang andP Liu ldquoCarbon reduction strategies basedon an NW small-world network with a progressive carbon taxrdquoSustainability vol 9 no 10 p 1747 2017

[15] J Kennedy ldquoParticle swarm optimizationrdquo in Encyclopedia ofMachine Learning pp 760ndash766 2010

[16] Z Liu T D Anderson and J M Cruz ldquoConsumer environ-mental awareness and competition in two-stage supply chainsrdquoEuropean Journal of Operational Research vol 218 no 3 pp602ndash613 2012

[17] S Swami and J Shah ldquoChannel coordination in green supplychainmanagementrdquo Journal of the Operational Research Societyvol 64 no 3 pp 336ndash351 2013

[18] I Nouira Y Frein and A B Hadj-Alouane ldquoOptimizationof manufacturing systems under environmental considerationsfor a greenness-dependent demandrdquo International Journal ofProduction Economics vol 150 pp 188ndash198 2014

[19] C-T Zhang H-X Wang and M-L Ren ldquoResearch on pricingand coordination strategy of green supply chain under hybridproduction moderdquoComputers amp Industrial Engineering vol 72pp 24ndash31 2014

[20] S Xi and C Lee ldquoA game theoretic approach for the optimalinvestment decisions of green innovation in a manufacturer-retailer supply chainrdquo International Journal of Industrial Engi-neering vol 22 no 1 2015

[21] M A Ulku and J Hsuan ldquoTowards sustainable consumptionand production Competitive pricing of modular products forgreen consumersrdquo Journal of Cleaner Production vol 142 pp4230ndash4242 2017

[22] A Yenipazarli and A J Vakharia ldquoGreen greener or brownchoosing the right color of the productrdquo Annals of OperationsResearch vol 250 no 2 pp 537ndash567 2017

[23] S A Raza S Rathinam M Turiac and L Kerbache ldquoAnintegrated revenue management framework for a firmrsquos green-ing pricing and inventory decisionsrdquo International Journal ofProduction Economics vol 195 pp 373ndash390 2018

[24] D Ghosh and J Shah ldquoA comparative analysis of greeningpolicies across supply chain structuresrdquo International Journal ofProduction Economics vol 135 no 2 pp 568ndash583 2012

[25] C-T Zhang and L-P Liu ldquoResearch on coordination mecha-nism in three-level green supply chain under non-cooperativegamerdquoApplied Mathematical Modelling vol 37 no 5 pp 3369ndash3379 2013

[26] D Ghosh and J Shah ldquoSupply chain analysis under green sensi-tive consumer demand and cost sharing contractrdquo InternationalJournal of Production Economics vol 164 pp 319ndash329 2015

[27] Y Huang KWang T Zhang and C Pang ldquoGreen supply chaincoordination with greenhouse gases emissions management a

game-theoretic approachrdquo Journal of Cleaner Production vol112 pp 2004ndash2014 2016

[28] P Liu and S Yi ldquoPricing policies of green supply chainconsidering targeted advertising and product green degree inthe Big Data environmentrdquo Journal of Cleaner Production vol164 pp 1614ndash1622 2017

[29] D Yang and T Xiao ldquoPricing and green level decisions ofa green supply chain with governmental interventions underfuzzy uncertaintiesrdquo Journal of Cleaner Production vol 149 pp1174ndash1187 2017

[30] S R Madani and M Rasti-Barzoki ldquoSustainable supply chainmanagement with pricing greening and governmental tariffsdetermining strategies a game-theoretic approachrdquo Computersamp Industrial Engineering vol 105 pp 287ndash298 2017

[31] W Xing J Zou and T-L Liu ldquoIntegrated or decentralized ananalysis of channel structure for green productsrdquo Computers ampIndustrial Engineering vol 112 pp 20ndash34 2017

[32] H Song and X Gao ldquoGreen supply chain game model andanalysis under revenue-sharing contractrdquo Journal of CleanerProduction vol 170 pp 183ndash192 2018

[33] Z Hong W Dai H Luh and C Yang ldquoOptimal configurationof a green product supply chain with guaranteed service timeand emission constraintsrdquo European Journal of OperationalResearch vol 266 no 2 pp 663ndash677 2018

[34] X Li G Q Shen P Wu and T Yue ldquoIntegrating BuildingInformation Modeling and Prefabrication Housing Produc-tionrdquo Automation in Construction vol 100 pp 46ndash60 2019

[35] R S Achrol and P Kotler ldquoMarketing in the network economyrdquoJournal of Marketing vol 63 pp 146ndash163 1999

[36] F Maroofi ldquoNetwork structure and network effects and con-sumer interaction in mobile telecommunications among stu-dentsrdquo African Journal of Marketing Management vol 4 no 2pp 55ndash64 2012

[37] M E J Newman ldquoModels of the small worldrdquo Journal ofStatistical Physics vol 101 no 3-4 pp 819ndash841 2000

[38] D J Watts and S H Strogatz ldquoCollective dynamics of rsquosmall-worldrsquo networksrdquo Nature vol 393 no 6684 pp 440ndash442 1998

[39] T Zhang and D Zhang ldquoAgent-based simulation of consumerpurchase decision-making and the decoy effectrdquo Journal ofBusiness Research vol 60 no 8 pp 912ndash922 2007

[40] B-D Kim R C Blattberg and P E Rossi ldquoModeling thedistribution of price sensitivity and implications for optimalretail pricingrdquo Journal of Business and Economic Statistics vol13 no 3 pp 291ndash303 1995

[41] S Patel and A Schlijper ldquoModels of consumer behaviourrdquo inProceedings of the 49th European Study Group with Industry2004

[42] H V Dang and M Lin ldquoHerd mentality in the stock marketOn the role of idiosyncratic participants with heterogeneousinformationrdquo International Review of Financial Analysis vol 48pp 247ndash260 2016

[43] S-K Chang ldquoHerd behavior bubbles and social interactions infinancial marketsrdquo Studies in Nonlinear Dynamics amp Economet-rics vol 18 no 1 pp 89ndash101 2014

[44] A Adepetu S Keshav and V Arya ldquoAn agent-based electricvehicle ecosystem model San Francisco case studyrdquo TransportPolicy vol 46 pp 109ndash122 2016

[45] M J Eppstein D K Grover J S Marshall and D M RizzoldquoAn agent-based model to study market penetration of plug-inhybrid electric vehiclesrdquo Energy Policy vol 39 no 6 pp 3789ndash3802 2011

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

14 Complexity

[46] Y Shi and R Eberhart ldquoA modified particle swarm opti-mizerrdquo in Proceedings of the IEEE International Conferenceon Evolutionary Computation and IEEE World Congress onComputational Intelligence (Cat No 98TH8360) pp 69ndash73IEEE Anchorage Alaska USA 1998

[47] J Kennedy ldquoThe particle swarm social adaptation of knowl-edgerdquo inProceedings of the 997 IEEE International Conference onEvolutionary Computation (ICEC rsquo97) pp 303ndash308 IEEE 1997

[48] D McCoy and S Lyons ldquoConsumer preferences and theinfluence of networks in electric vehicle diffusion an agent-based microsimulation in Irelandrdquo Energy Research and SocialScience vol 3 pp 89ndash101 2014

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom

Hindawiwwwhindawicom Volume 2018

MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Mathematical Problems in Engineering

Applied MathematicsJournal of

Hindawiwwwhindawicom Volume 2018

Probability and StatisticsHindawiwwwhindawicom Volume 2018

Journal of

Hindawiwwwhindawicom Volume 2018

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawiwwwhindawicom Volume 2018

OptimizationJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Engineering Mathematics

International Journal of

Hindawiwwwhindawicom Volume 2018

Operations ResearchAdvances in

Journal of

Hindawiwwwhindawicom Volume 2018

Function SpacesAbstract and Applied AnalysisHindawiwwwhindawicom Volume 2018

International Journal of Mathematics and Mathematical Sciences

Hindawiwwwhindawicom Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Hindawiwwwhindawicom Volume 2018Volume 2018

Numerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisNumerical AnalysisAdvances inAdvances in Discrete Dynamics in

Nature and SocietyHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Dierential EquationsInternational Journal of

Volume 2018

Hindawiwwwhindawicom Volume 2018

Decision SciencesAdvances in

Hindawiwwwhindawicom Volume 2018

AnalysisInternational Journal of

Hindawiwwwhindawicom Volume 2018

Stochastic AnalysisInternational Journal of

Submit your manuscripts atwwwhindawicom