06935022 co2 emission

238 IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 12, NO. 1, JANUARY 2015

Energy-Efficient Models of Sustainable Location for aVehicle Inspection Station With Emission Constraints

Guangdong Tian and Yue Liu

Abstract—Transportation facility or automotive service enter-prise location is an interesting and important issue. To improvetransportation efficiency, many researchers have addressed thetraditional facility location allocation (FLA) problem, e.g., theFLA problem with the minimum transportation cost or themaximum obtained profit. However, with the improvement ofsaving energy awareness and the enhancement of environmentalconcerns, energy efficient and low carbon emission should beconsidered as key factors influencing the FLA problem. To handlethis issue via a more practical method, this work proposes asustainable location issue for an automotive service enterprise.That is, by taking the vehicle inspection station as a typical au-tomotive service enterprise and an example, this work presentsnew energy-efficient models of its sustainable location with carbonconstraints. An artificial fish swarm algorithm is proposed tosolve the proposed models. Some numerical examples are given toillustrate the proposed models and testify the effectiveness of thealgorithm.

Note to Practitioners—This work concerns the sustainable man-agement problem for locating an automotive service enterprise.To deal with such an issue, this work proposes a novel sustainableanalysis method. The previous research handles such a problemthrough a methodology based on the traditional economicalmethod, which is not enough without considering sustainablefactors, e.g., energy consumption and carbon emission. The goalof this work is to establish the energy-efficient models for locatingan automotive service enterprise with carbon emission constraint,i.e., to meet the specific requirement of carbon emission assuredlywhile minimizing the transportation energy/fuel consumption ofservice customers. Both theoretical and simulated results demon-strate that the proposed approach is effective and feasible. Suchresults can help decision makers perform better judgments whena practical automotive service enterprise is executed.

Index Terms—Modeling and simulation, optimization algo-rithm, sustainable facility location allocation.

I. INTRODUCTION

S INCE facility location allocation (FLA) problems wereinitialized by Cooper [1] in 1963, there have been many

advances in their solution methods, variants and applications,

Manuscript received June 14, 2014; revised August 28, 2014; acceptedSeptember 21, 2014. Date of publication October 23, 2014; date of currentversion December 31, 2014. This paper was recommended for publication byAssociate Editor Q.-S. Jia and Editor L. Shi upon evaluation of the reviewers’comments. This work was supported in part by Fundamental Research Fundsfor the Central Universities under Grant 2572014BB08, and the PostdoctoralScience Foundation Project of China under Grant 2013M541329.The authors are with the Transportation College, Northeast Forestry Univer-

sity, Harbin 150040, China (e-mail: [email protected]; [email protected]).Color versions of one or more of the figures in this paper are available online

at http://ieeexplore.ieee.org.Digital Object Identifier 10.1109/TASE.2014.2360673

e.g., emergency service systems, telecommunication networks,public services, and transportation facilities. The early workmainly focused on FLA location problems, many differentmodels have been formed [1]–[4], i.e., median models [5], [6],center models [7]–[10], covering models [11], [12], hub loca-tion models [13], and hierarchical location models [14]–[16].Meanwhile, a large number of solution approaches for differentmodels have been proposed, e.g., simulated annealing, GeneticAlgorithm, and Tabu Search [17]–[25].The location of an automotive service enterprise is one of

the important applications of FLA. Some scholars have dis-cussed the related issues under uncertainty. For example, byconsidering the uncertainty of the number of inspection vehi-cles, Tian et al. established a stochastic location model for ve-hicle inspection station to achieve the minimum transportationcost of vehicle inspection customers [26]. They proposed a hy-brid algorithm integrating Genetic Algorithm (GA) and Neuralnetworks (NNs) to solve this vehicle inspection station model[27]. By considering the uncertainty of the number of inspec-tion vehicles and road conditions under different districts of in-spection customers, they proposed the expected value model forvehicle inspection station location problemwith random inspec-tion demand and different transportation cost [28]. Tian et al.established a cost-profit tradeoff model for a vehicle inspec-tion station with stochastic inspection demand [29]. Ling et al.solved the car parking location problem by minimizing the totaltransportation cost of customers based on an iterative algorithm[30]. Wang et al. analyzed the location and pricing of a park-and-ride facility under alternative profit maximization and so-cial cost minimization objectives [31].Based on the above overview, we can see that traditional

models based on traditional evaluation parameters dominate thefield of automotive service enterprise or transportation facilitylocation under deterministic and uncertainty environments. Themain evaluation parameters include the shortest transportationdistance, the minimum transportation cost, and the maximumobtained profit. However, with the rapid development ofeconomy, environmental pollution and energy shortages havebecome two key factors which influence the economic sus-tainable growth. To promote the execution of related worksof energy-saving and emission reduction, we must addressthe sustainable location allocation for an automotive serviceenterprise. To do so, by taking a vehicle inspection stationas a typical automotive service enterprise and an example,this work proposes to establish energy-efficient models of itslocation with carbon constraints for the first time to the bestknowledge of the authors. Namely, this work aims at finding

1545-5955 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

TIAN AND LIU: ENERGY-EFFICIENT MODELS OF SUSTAINABLE LOCATION FOR A VEHICLE INSPECTION STATION WITH EMISSION CONSTRAINTS 239

a new way to determine the location of a vehicle inspectionstation with taking the energy-efficient and carbon emissioninto full account.In Section II, the assumptions and parameters of proposed

models are introduced. In Section III, energy-efficient models ofsustainable location for an automotive service are established.Section IV introduces the algorithm to solve these models. InSection V, some numerical examples are presented to test the ef-fectiveness of the proposed algorithm. Finally, Section VI con-cludes our work and describes some future research issues.

II. PROBLEM STATEMENTS

In order to establish our models conveniently, the followingassumptions and parameters are given in this work.

A. Parameters

1) —The index of vehicle inspection demand region,.

2) —The index of need to establish a vehicle inspection sta-tion, .

3) —The coordinate of the location of the th vehicleinspection demand region.

4) —The coordinate of the location of the th vehicleinspection station. Note that it is a decision variable in thiswork.

5) —Per kilometer energy/fuel consumption from vehicleinspection demand region to inspection station , with itsunit being L/km.

6) —The number of vehicles from vehicle inspection de-mand region to inspection station .

7) —Emission factor, i.e., producing the quantity of carbondioxide to burn a liter of fuel, and its unit is kg/L.

B. Assumptions

This work makes the following assumptions.1) The distribution condition of vehicle inspection customersdemand is not taken into consideration and the center ofeach vehicle inspection demand region is treated as thecoordinate location of the vehicle inspection demand.

2) The inspection capability of a vehicle inspection station isignored, namely, the inspection capability is big enough tomeet requirement of users.

3) The cost per kilometer from the inspection demand cus-tomer region to a vehicle inspection station is constant,that is, the relationship between the transportation cost andtransportation distance is viewed as a linear relationship.

III. ENERGY-EFFICIENT MODELS OF THE LOCATION FOR AVEHICLE INSPECTION STATION

Based on the presented concepts and assumptions, by takinga vehicle inspection station as a typical automotive service en-terprise and an example, we build two energy-efficient modelsof its sustainable location, i.e., energy-efficient model withoutcarbon emission constraint and energy-efficient model withcarbon emission constraint, which are presented next in detail.Note that since there is no vehicle inspection station in manycountries, the scope of these proposed models is suitable for

the country, in which there is a vehicle inspection station atleast, e.g., China.

A. Energy-Efficient Model Without Carbon EmissionConstraint

In the actual location process of a vehicle inspection station,a decision-maker wants to seek the minimum energy consump-tion of vehicle inspection customers. In order to deal with thisissue, we establish an energy-efficient model for a sustainablelocation of a vehicle inspection station. Its objective function is

(1)

(2)

where is the total energy consumption of vehicle inspectioncustomers and it is expressed as is

the number of vehicles from vehicle inspection region to in-spection station , is the amount of energy consumption perkilometer from vehicle inspection demand region to inspectionstation , and is the distance between vehicle inspection de-mand region and inspection station and it is expressed as

. and are the lower andupper bounds of the coordinate , respectively, and are thelower and upper bounds of coordinate . , , and canbe determined by the coordinate of vehicle inspection demandregions.

B. Energy-Efficient Model With Carbon Emission Constraint

Currently, low carbon transportation has become one of thehot topics. Carbon dioxide produced by the combustion of fuelhas a seriously impact on our life environment. To keep trans-portation sustainability, the carbon emission should be takeninto full account in the actual location process of a vehicle in-spection station. In order to satisfy this requirement, a decision-maker must seek the minimum total energy consumption of ve-hicle inspection customers with the given carbon dioxide emis-sion constraint. Finally, the following energy-efficient modelwith the carbon emission constraint for sustainable location of avehicle inspection station is established. The objective functionof this model is written as

(3)

(4)

where is the carbon dioxide emission and it is expressed as, where is carbon dioxide

emission factor. is the given carbon dioxide emission, andis the carbon dioxide emission constraint.

IV. SOLUTION ALGORITHM

Artificial fish swarm algorithm (AFSA) is an effectivemethod to solve industrial optimization problems, e.g., waste-


water management, clustered oversubscription planning, andnuclear fuel cycle [32]–[34]. Thus, we propose to adopt artifi-cial fish swarm algorithm to solve the established models. Itsbasic ideas are presented next.

A. AFSA

According to the swarm intelligence, AFSA is an artificial in-telligent algorithm based upon the simulation of the collectivebehaviors of fish stocks. It simulates the behavior of an arti-ficial fish , and then constructs a swarm of . Everyone of will search its own best of local, pass on informa-tion in its self-organized system and finally achieve the best ofglobal. We should initialize the school of fish randomly. Letthe searching space be -dimensional and there are Fishnumfishes in the colony. The current state vector of an swarm is

, where is the variable tobe optimized. Visual is the visual distance. The occurs onlyin the inner radius of the circle to the length of the field of visionvarious acts. The foodconsistence of in the current positionis represented by , where is the objective func-tion. The distance between the th and the th individualcan be expressed as , and are random fishes.Step means the maximum step size of . isthe degree of congestion factor. In the initial state of the algo-rithm, the iterative number of the variable should be defined asiterative times of searching for food. In addition, Maxgenis set as the limitation of iterative times in this work. Becausethe maximization problem and the minimization problem can betransformed with each other, we take the maximization problemas an example in the following analysis.1) Preying Behavior: This is a basic biological behavior that

tends to the food. Let the current state of an be , then weselect a new state randomly in its visual field. If the foodconsistency of is bigger than that of , it moves a step inthat direction, otherwise, we select a state randomly againand judge whether it satisfies the forward condition. If it cannotbe satisfied after a preset Try_number times, it moves a steprandomly. The step moving follows the following rule:

(5)

where rand () is a random number and ranges 0 from 1.2) Swarming Behavior: Let be the current state of

searching companion in the neighborhood with , if, it means the current position of companion

has higher food consistency and it is not crowded. Thewill move a step towards companion ; otherwise, continuesearching behavior. The following behavior is a mathematic de-scription:

(6)

3) Following Behavior: Let be the current state ofand search its companion whose is the largest inthe neighborhood, if , it means that the currentposition of companion has higher food consistency and itis not crowded. The would move a step towards companion

, otherwise, the preying behavior would be executed.The following is a mathematic description:

(7)4) Bulletin: Bulletin is used to record the optimal result ofand the optimal value of the problem. Each updates its

own state and compares it with that on the bulletin after makingmovements. If its current state of is better, then the valueon the bulletin would be replaced.Based on the above presentation, AFSA has the following

steps.Step 1) Initialize the parameters of artificial fish, i.e., Step

and Visual, the maximum number of explorationTry_number, the maximum number of iterationsMaxgen, and the number of fishes Fishnum. Notethat the carbon emission constraints should bechecked in this Step.

Step 2) Set bulletin board to record the current status of eachfish, and select the optimal value record.

Step 3) Implement preying, swarming and following behav-iors.

Step 4) Update the optimal value on the bulletin board.Step 5) If the Maxgen is accomplished, the optimal result

would be output, otherwise, AFSA returns to Step 2.The above algorithm has been implemented in the MATLAB

programming language (R2010b).

V. CASE STUDY

Considering a location problem of the vehicle inspection sta-tion for Fushun in China, it will be applied to our proposedmodels. This city is divided into five vehicle inspection regions,i.e., Development, Dongzhou, Wanghua, Xinfu, and Shunchengdistricts, as shown in Fig. 1. Their central coordinates of eachdistrict are listed in Table I. In addition, the central coordinateof this city is (41578784.96, 4638592.97). In order to optimizethe location problem conveniently, taking the center coordinateof this city as relative coordinate origin, and the coordinates ofthese five districts are transformed into the following numericalvalues, as shown in Table I. In addition, the number of inspec-tion vehicles and the energy consumption per kilometer of eachdemand region are listed in Table I. The carbon emission factoris set to be 2.7 in this work.The parameters of the AFSA are set next: the number of fish

Fishnum is 50, the visual distance Visual is 1, the Step is 0.1,the maximum number of generationsMaxgen is 100, the degreeof congestion factor is 0.618, and the number of explorationTry_number is 100.


Fig. 1. Schematic diagram of Fushun city.

TABLE ITYPICAL PARAMETERS DESCRIPTION OF ESTABLISHED MODEL

Example 1: A decision-maker hopes to build a vehicle detec-tion station with the minimum energy consumption of vehicleinspection customers. This problem can be translated to solvethe following model:

(8)

(9)

After the algorithm is executed, the following results can beobtained:

The results denote that the location coordinate of the vehicledetection station is ( 2661.18041, 1049.45692), the lowesttotal fuel consumption of vehicle detection customers is

.In addition, when the AFSA is executed, searching process of

optimal coordinates is shown in Fig. 2.Example 2: A decision-maker wishes to build a vehicle

detection station with the minimum energy consumption of

Fig. 2. Searching process of optimal coordinates.

Fig. 3. Searching process of optimal coordinates.

vehicle inspection customers under the following carbon con-straint . This problem can be translated tosolve the following model:

(10)

(11)

After the algorithm is executed, the following results can beobtained:

The results denote that the location coordinate of a vehicledetection station is ( 4748.69585, 362.21457), and the lowesttotal energy consumption of vehicle inspection customers underthis carbon constraint is .Also, when the AFSA is executed, searching process of op-

timal coordinates is shown in Fig. 3.


TABLE IICOMPARISON OF SOLUTIONS OF EXAMPLE 1 WITH DIFFERENT CASES

Moreover, to test the effectiveness of the algorithm, the re-sults of Example 1 with different cases and AFSA parametersare shown inTable II. Their errors are computed and showninTable II, and the relative error is defined as (Actual value-Op-timal value)/Optimal value 100%, where optimal value isthe minimum value of solution results among cases, while ac-tual values are the solution results of the model when the algo-rithm is run in each case. The relative error can be expressed as

.FromTable II, the relative error does not exceed 0.27%. These

indicate that the proposed algorithm is highly satisfactory whenit is used to solve the proposed models.

VI. CONCLUSION

The sustainable location transportation facility or automotiveservice enterprise has a great impact on the execution of relatedworks of energy-saving and emission reduction. To solve theseproblems, this paper proposes to establish energy-efficientmodels of sustainable locations for a vehicle inspection stationfor the first time, i.e., energy-efficient model without carbonemission constraint and energy-efficient model with carbonemission constraint. In addition, an artificial fish swarm algo-rithm is proposed to solve the established models. The resultsreveal that it is feasible and effective when being used to solvethe established models. The results can be used to guide deci-sion makers in making better decisions when a transportationfacility or automotive service enterprise is planed.There are some limitations with the proposed method. For

example, for the region constraint, this work merely considersthe condition of the round shape. However, the region constraintcan more likely to be irregular shapes in reality and the matterthat how to deal with this issue needs to be further discussed. Inaddition, the uncertainty analysis and control for locating an au-tomotive service enterprise needs further discussion [35]–[40].

REFERENCES[1] L. Cooper, “Location-allocation problems,” Oper. Res., vol. 11, pp.

331–344, 1963.[2] B. A. Murtagh and S. R. Niwattisyawong, “Efficient method for the

multi-depot location em dash allocation problem,” J. Oper. Res. Soc.,vol. 33, pp. 629–634, 1982.

[3] M. A. Badri, “Combining the analytic hierarchy process and goal pro-gramming for global facility location-allocation problem,” Int. J.Prod.Econ., vol. 62, pp. 237–248, 1999.

[4] M. Hodey, E. Melachrinoudis, and X. Wu, “Dynamic expansion andlocation of an airport: a multiple objective approach,” Transport. Res.Part A-Pol., vol. 31, pp. 403–417, 1997.

[5] H. K. Rajagopalan, C. Saydam, and J. Xiao, “A multi-period setcovering location model for dynamic redeployment of ambulances,”Comput. Oper. Res., vol. 35, pp. 814–826, 2008.

[6] R. Z. Farahani, N. Asgari, N. Heidari, M. Hosseininia, and M. Goh,“Covering problems in facility location: A review,” Comput. Ind. Eng.,vol. 62, pp. 368–407, 2012.

[7] C. Toregas, R. Swain, C. ReVelle, and L. Bergman, “The location ofemergency service facilities,” Oper. Res., vol. 19, pp. 1363–1373,1971.

[8] R. Church and C. ReVelle, “The maximal covering location problem,”Pap. Reg. Sci. Assoc., vol. 32, pp. 101–118, 1974.

[9] R. Z. Farahani and M. Hekmatfar, Facility Location: Concepts,Models, Algorithms And Case Studies. Heidelberg, Germany:Physica-Verlag, 2009.

[10] S. Nickel and J. Puetro, Location Theory: A Unified Approach.Berlin, Germany: Springer-Verlag, 2005.

[11] M. S. Daskin, Network and Discrete Location: Models, Algorithms,Applications. New York, NY, USA: Wiley, 1995.

[12] S. L. Hakimi, “Optimum locations of switching centers and the abso-lute centers and medians of a graph,” Oper. Res., vol. 12, no. 3, pp.450–459, 1964.

[13] M. E. O’Kelly, “A quadratic integer program for the location of inter-acting hub facilities,” Eur. J. Oper. Res., vol. 32, no. 3, pp. 393–404,1987.

[14] A. Klose and A. Drexl, “Facility location models for distributionsystem design,” Eur. J. Oper. Res., vol. 162, no. 1, pp. 4–29, 2005.

[15] Z. Drezner and H. Hamacher, Facility Location: Applications andTheory. Berlin, Germany: Springer-Verlag, 2002.

[16] G. Sahin and H. Sural, “A review of hierarchical facility locationmodels,” Comput. Oper. Res., vol. 34, pp. 2310–2331, 2007.

[17] A. B. Arabani and R. Z. Farahani, “Facility location dynamics: Anoverview of classifications and applications,” Comput. Ind. Eng., vol.62, pp. 408–420, 2012.

[18] R. E. Kuenne and R. M. Soland, “Exact and approximate solutions tothe multisourceWeber problem,”Math. Program., vol. 3, pp. 193–209,1972.

[19] A. T. Murray and R. L. Church, “Applying simulated annealing to lo-cation-planning models,” J. Heuristics, vol. 2, pp. 31–53, 1996.

[20] A. T. Ernst and M. Krichnamoorthy, “Solution algorithms for the ca-pacitated single allocation hub location problem,” Ann. Oper. Res., vol.86, pp. 141–159, 1999.

[21] D. Gong, M. Gen, W. Xu, and G. Yamazaku, “Hybrid evolutionarymethod for obstacle location-allocation problem,” Int. J. Comput. Ind.Eng., vol. 29, pp. 525–530, 1995.

[22] R. Z. Farahani, M. SteadieSeifi, and N. Asgari, “Multiple criteria fa-cility location problems: A survey,” Appl. Math. Model, vol. 34, pp.1689–1709, 2010.

[23] K. J. Wang, B. Makond, and S. Y. Liu, “Location and allocation deci-sions in a two-echelon supply chain with stochastic demand-A genetic-algorithm based solution,” Expert Sys. Appl., vol. 38, pp. 6125–6131,2011.

[24] Z. Q. Lu and N. Bostel, “A facility location model for logistics sys-tems including reverse flows: The case of remanufacturing activities,”Comput. Oper. Res., vol. 34, pp. 299–323, 2007.

[25] M. A. Arostegui, Jr., S. N. Kadipasaoglu, and B. M. Khumawala, “Anempirical comparison of tabu search, simulated annealing, genetic al-gorithms for facilities location problems,” Int. J. Prod. Econ., vol. 103,pp. 742–754, 2006.

[26] G. D. Tian, Y. M. Liu, G. Xu, J. Y. Dai, and H. G. Ren, “Analysis ofnetwork layout optimization problem for vehicle inspection station inrandom environment,” J. Jilin Univ. (Eng. Technol. Edit.), vol. 41, pp.122–125, 2011.

[27] T. G. Qiang, G. D. Tian, J. W. Chu, Z. K. Li, and D. H. Xiong, “Loca-tion analysis of vehicle inspection station based on GA-NN,” Adv. Inf.Sci. Serv. Sci., vol. 5, no. 4, pp. 622–629, 2013.

[28] G. D. Tian, Y. M. Liu, J. Xie, Z. J. Lai, and X. N. Cao, “Expectedvalue model for vehicle inspection station location problem withrandom inspection demand and different transportation cost,” Appl.Mech. Mater., vol. 97–98, pp. 542–545, 2011.

[29] G. D. Tian, M. C. Zhou, J. W. Chu, T. G. Qiang, and H. S. Hu, “Sto-chastic cost-profit trade-off model for locating an automotive serviceenterprise,” IEEE Trans. Autom. Sci. Eng., 2014, 10.1109/TASE.2013.2297623, to be published.

[30] L. Ling, W. Li, W. Wang, and G. Hu, “Multi-area distributing and iter-ative optimal location algorithm for parking areas,” China Civil Eng.J., vol. 36, no. 7, pp. 18–21, 2003.

[31] J. Y. T. Wang, H. Yang, and R. Lindsey, “Locating and pricing park-and-ride facilities in a linear monocentric city with deterministic modechoice,” Transp. Res., vol. 38, no. 8, pp. 709–731, 2004.


[32] K. M. Wagialla, M. M. El-Halwagi, and J. M. Ponce-Ortega, “An inte-grated approach to the optimization of in-plant wastewater interceptionwith mass and property constraints,” Clean. Techn. Environ. Policy,vol. 14, no. 2, pp. 257–265, 2012.

[33] Z. Huang and Y. Chen, “An improved artificial fish swarm algorithmbased on hybrid behavior selection,” Int. J. Contr. Autom., vol. 6, pp.103–116, 2013.

[34] M. Neshat, G. Sepidnam, M. Sargolzaei, M. Sargolzaei, and A. N.Toosi, “Artificial fish swarm algorithm: A survey of the state-of-the-art,hybridization, combinatorial and indicative applications,” Artif. Intell.Rev., 2012, 10.1007/s10462-012-9342-2.

[35] G. D. Tian, M. C. Zhou, J. W. Chu, and Y. M. Liu, “Probability eval-uation models of product disassembly cost subject to random removaltime and different removal labor cost,” IEEE Trans. Autom. Sci. Eng.,vol. 9, no. 2, pp. 288–295, Apr. 2012.

[36] Y. F. Fang, F. Chu, and M. C. Zhou, “Optimal lane reservation intransportation network,” IEEE Trans. Intell. Transp., vol. 13, no. 2, pp.482–491, 2012.

[37] W. Sun, H. Gao, and B. Yao, “Adaptive robust vibration control offull-car active suspensions with electro-hydraulic actuators,” IEEETrans. Contr. Syst. T., vol. 21, no. 6, pp. 2417–2422, 2013.

[38] W. Sun, H. Gao, and O. Kaynak, “Finite frequency control forvehicle active suspension systems,” IEEE Trans. Contr. Syst. T., vol.19, no. 2, pp. 416–422, 2011.

[39] H. Ke and B. Liu, “Project scheduling problem with stochastic activityduration times,” Appl. Math. Comp., vol. 168, pp. 342–353.

[40] H. Ke and B. Liu, “Project scheduling problem with mixed uncer-tainty of randomness and fuzziness,” Eur. J. Oper. Res., vol. 183, pp.135–147, 2007.

Guangdong Tian received the B.S. degree invehicle engineering from Shandong University ofTechnology, Zibo, China, in 2007, and the M.S. andPh.D. degrees in automobile application engineeringfrom Jilin University, Changchun, China, in 2009and 2012, respectively.He is currently an Associate Professor with

the Transportation College of Northeast ForestryUniversity, Harbin, China, and Founding Memberof Sustainable Production and Service Automationof the IEEE Robotics and Automation Society. He

was invited to organize the International Conference of Numerical Analysisand Applied Mathematics 2012 (ICNAAM 2012) and gave a session proposalentitled “Recent Advances on the Uncertain Theory, Analysis and Their Appli-cation.” His research focuses on remanufacturing, recycling and reuse of usedproducts, intelligent detection and repair of automotive, decision making andintelligent optimization, facility location, and prediction and assessment foreconomic and environment. He has published over 40 journal and conferenceproceedings papers in the above research areas.Dr. Tian has published papers in the IEEE TRANSACTIONS ON AUTOMATION

SCIENCE AND ENGINEERING, Computer and Chemical Engineering, and Com-puters and Industrial Engineering, etc. In addition, he serves as a Reviewerfor more than ten international journals and conferences including the IEEETRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, the IEEETRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, PART A , the AsianJournal of Control, etc. He is listed in Marquis Who’s Who in the World, 30thEdition, 2013.

Yue Liu received the B.S. degree in vehicle engi-neering from Northeast Forestry University, Harbin,China, in 2013. She is currently a graduate of Trans-portation College, Northeast Forestry University,Harbin, China.Her research focuses on remanufacturing, recy-

cling and reuse of used products, decision makingand intelligent optimization, facility location.

06935022 co2 emission

Documents