Research ArticleA Hybrid MCDM Model for Improving GIS-BasedSolar Farms Site Selection

Chao-Rong Chen1 Chi-Chen Huang1 and Hung-Jia Tsuei23

1 Department of Electrical Engineering National Taipei University of Technology Taipei 10608 Taiwan2Department of Electronic Engineering National Taipei University of Technology Taipei 10608 Taiwan3Graduate Institute of Networking and Multimedia National Taiwan University Taipei 10617 Taiwan

Correspondence should be addressed to Hung-Jia Tsuei taiwanseogmailcom

Received 16 May 2014 Revised 16 July 2014 Accepted 16 July 2014 Published 20 August 2014

Academic Editor Ching-Song Jwo

Copyright copy 2014 Chao-Rong Chen et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The purpose of this research is to establish a decision model for improving the performance of solar farms To investigate theinterdependent interrelationship and influential weights among criteria for solar farms site selection a hybrid MCDM modelincluding decision-making trial and evaluation laboratory (DEMATEL) and DEMATEL-based analytic network process (DANP)based on geographical information systems (GIS) is utilized The empirical results display that there are interdependence andself-effect relationships among criteria via DEMATEL technique According to the influential network relation map (INRM) thedimension that administrators of solar energy industry should improve first when enhancing the performance of solar farms isorography In the ten criteria solar radiation is the most important criterion impacting solar farms site selection followed byaverage temperature and distance to villages

1 Introduction

Among the different types of renewable energy resourcesphotovoltaic (PV) solar energy is by far the largest exploitableresource for offering more energy in 1 hour to the earth thanall of the energy consumed by humans in a whole year [1]In addition the PV solar market increased rapidly becauserevenues created from the industry will achieve over US$100billion before 2020 [2] However PV solar energy has notachieved adequate maturity so great efforts are being maderegarding lower manufacturing costs and higher efficiencies[3 4] Solar farms site selection becomes one of the mostimportant issues for administrators of solar energy industryto maximize the performance of solar farms

Previous studies regarding solar farms locations focusedon considering simply what factors would influence solarfarms [5 6] conveying what factors influence solar farmsand whether the impacts were positive or negative Thesediscoveries for building a decision model of solar farms siteselection have little contribution to it In addition although

the interrelationship and influential weights among criteriaare extremely useful for the administrators to simultaneouslyconsider interdependent multicriteria in real world litera-tures concerning these problems are inadequate Thereforethe purpose of this research is to construct a decision modelfor enhancing the performance of solar farms The specificimprovement process and influential weights of solar farmssite will be explored by a hybrid multiple criteria decisionmaking (MCDM) model based on geographical informationsystems (GIS)

In order to provide the administrators with the solutionto these issues a hybrid MCDMmodel combining decision-making trial and evaluation laboratory (DEMATEL) withDEMATEL-based analytical network process (DANP) isemployed The criteria of solar farms site are identifiedthrough GIS According to the survey of experts this paperadopts DEMATEL technique to probe into the interde-pendent decision making for constructing the influentialnetwork relation map (INRM) The strategies for improvingthe performance of solar farms can thus be obtained through

Hindawi Publishing CorporationInternational Journal of PhotoenergyVolume 2014 Article ID 925370 9 pageshttpdxdoiorg1011552014925370

2 International Journal of Photoenergy

the influence values of criteria in INRM Subsequently tosolve the problems with interdependent criteria the influ-ential weights of solar farms site can be received via DANPderived from the basic concept of analytical network process(ANP) proposed by Saaty [7]

2 Calculation Models and Methods

21 Criteria for Solar Farms Site Selection Industrial siteselection is one of the fundamental decisions in the start-upprocess expansion or relocation GIS are adopted combiningwith other systems and methods such as systems for decisionmaking (DSS) and the method for MCDM Synergistic effectis created in conjunction with these tools which contribute tothe efficiency and quality of spatial analysis for industrial siteselection [8] Chen and Pang [9] proposed a fuzzy analyticnetwork process (FANP) to investigate critical characteris-tics of successful PV solar energy industry and examinedsuitable forms of organization for knowledge management(KM) in order to distribute existing knowledge as wellas to create new knowledge Carrion et al [10] used anenvironmental decision-support system (EDSS) for choosingoptimal sites for grid-connected PV power plantsThe systemcombined multicriteria analysis and the analytic hierarchyprocess (AHP) with GIS technology and took into accountenvironment orography location and climate factors in themeantime Moreover the combination of GIS and MCDMmethods is used to acquire the assessment of the optimalplacement of PV solar power plants The exceptional toolcan also be employed to analyze an extensive cartographicand alphanumeric database for simplifying problems to solveand promote the use of multiple criteria [11] Uyan [12]indicated that solar energy investments have been expandedquickly in recent years It is a crucial issue for huge solarfarms investments to select location for the considerationof terrain local weathering factors proximity to high trans-mission capacity lines agricultural facilities and environ-mental conservation Multiple criteria evaluation methodsare often utilized for various site selection researches Thestudy determined proper site selection for solar farms byadopting GIS and AHP Sanchez-Lozano et al [13] revealedthat great position with high percentages of potential solarradiation can host electricity generation plants via PV solarfarms The best plots suitable for installing PV solar farmsare identified by employing GIS and classified accordingto multiple evaluation aspects by means of a multicriteriamodel

By literature review criteria affecting solar farms siteselection for improving the performance of solar energyindustry are arranged as follows Solar farms site includesfour dimensions environment (119863

1) orography (119863

2) location

(1198633) and climatology (119863

4) To be specific environment is

affected by agrological capacity (1198621) orography is influenced

by slope (1198622) orientation (119862

3) and area (119862

4) location is

affected by distance to roads (1198625) distance to power lines

(1198626) distance to villages (119862

7) and distance to substations

(1198628) climatology is influenced by solar radiation (119862

9) and

average temperature (11986210)

22 DEMATEL Method The DEMATEL method is utilizedto probe into the problems of interdependent criteria forestablishing the INRM [14 15] This method has been practi-cally used in decisionmaking problems of various fields suchas vendor selection for recycledmaterial and the organic lightemitting diode technology selection [16 17]

Experts with specialty of technology for PV solar energyand experience from solar energy industry including officialsof Bureau of Energy scholars of energy engineering andmanagers of solar industry are invited to help carry out thisresearch Information required for sufficient evaluation ofsolar farms site is collected by utilizing interviews and fillingsuitable questionnaires In the questionnaires a scale of 0 12 3 and 4 presents the degree from ldquono influencerdquo to ldquoveryhigh influencerdquo In addition the confidence level for expertsis tested (97638) in this real case of fifteen experts by thisresearch

The technique is introduced as follows in the first placethe influence matrix is received by scoresThe related expertsare asked to indicate the degrees of influence among criteriathat is to point out how much the criteria affect each otherThe influence matrix A can thus be obtained Secondly thenormalized influence matrix G can be calculated by utilizing(1) and (2) to normalize A

G = 119898 sdot A (1)

119898 = min

1

max119894sum119899

119895=1

10038161003816100381610038161003816119886119894119895

10038161003816100381610038161003816

1

max119895sum119899

119894=1

10038161003816100381610038161003816119886119894119895

10038161003816100381610038161003816

(2)

Thirdly the total influence matrix T can be acquired via theformula T = G + G2 + G3 + sdot sdot sdot + G119902 = G(I minus G)

minus1 whenlimℎrarrinfin

Gℎ = [0]119899times119899

where I denotes the identity matrixThe fourth step is to use (3) and (4) to construct the INRMthrough vectors r and d (the sums of the rows and columnsseparately within the total influence matrix T = [119905

119894119895]119899times119899

)

r = [119903119894]119899times1

= [

[

119899

sum

119895=1

119905119894119895]

]119899times1

(3)

d = [119889119895]119899times1

= [

119899

sum

119894=1

119905119894119895]

1015840

1times119899

(4)

where the superscript 1015840 stands for transpose If 119903119894represents

the row sum of the 119894th row in matrix T then 119903119894reveals the

sum of direct and indirect influences of criterion 119894 on theother criteria Also if 119904

119895stands for the column sum of the

119895th column of matrix T then 119904119895displays the sum of direct

and indirect effects that criterion 119895 has obtained from theother criteria Furthermore when 119894 = 119895 the sum of therow and column aggregates (119903

119894+ 119889119894) which presents the

giving and received degree of influences that is (119903119894+ 119889119894)

shows the intensity of the important role that the 119894th criterionplays in the problem When (119903

119894minus 119889119894) is positive the 119894th

criterion influences other criteria On the contrary if (119903119894minus119889119894)

is negative other criteria influence the 119894th criterion (ie 119894thcriterion receives influence from other criteria) Therefore

International Journal of Photoenergy 3

the interdependent interrelationship among criteria can beestablished by the INRM [18 19]

23 DANP Method Comparing with [11ndash13] the advantageof the proposed model is to improve the equal weightingassumption of the traditionalmethod for solving complicatedproblems in influential weights [16] ANP was developedby Saaty [7] to reduce the limitations of the AHP and todetermine nonlinear and complicated network relationshipsTherefore after investigating the influential relationship thisstudy integrates DEMATEL with basic concept of ANP(called DEMATEL-based ANP DANP) to obtain the influ-ential weights of criteria [20] The DANP includes foursteps First build the influence network structure based onDEMATEL Second find the unweighted supermatrix Thetotal influence matrix T presented in (5) is derived fromDEMATEL

T119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T11119888

sdot sdot sdot T1119895119888

sdot sdot sdot T1119899119888

T1198941119888

sdot sdot sdot T119894119895119888

sdot sdot sdot T119894119899119888

T1198991119888

sdot sdot sdot T119899119895119888

sdot sdot sdot T119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(5)

Normalize each level of T119888by using the total degree of

influence to obtain T120572119888via

T120572119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T12057211119888

sdot sdot sdot T1205721119895119888

sdot sdot sdot T1205721119899119888

T1205721198941119888

sdot sdot sdot T120572119894119895119888

sdot sdot sdot T120572119894119899119888

T1205721198991119888

sdot sdot sdot T120572119899119895119888

sdot sdot sdot T120572119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(6)

whereT12057211119888

can be calculated through (7) and (8) by the sameway T120572119899119899

119888can be obtained

11988911

119894=

1198981

sum

119895=1

11990511

119888119894119895 119894 = 1 2 119898

1 (7)

T12057211119888

=

[[[[[[[[[[[[[[[

[

11990511

11986211

11988911

1

sdot sdot sdot11990511

1198621119895

11988911

1

sdot sdot sdot11990511

11986211198981

11988911

1

11990511

1198621198941

11988911

119894

sdot sdot sdot11990511

119862119894119895

11988911

119894

sdot sdot sdot11990511

1198621198941198981

11988911

119894

11990511

11986211989811

119889111198981

sdot sdot sdot11990511

1198621198981119895

119889111198981

sdot sdot sdot11990511

11986211989811198981

119889111198981

]]]]]]]]]]]]]]]

]

=

[[[[[[[

[

11990512057211

11986211 sdot sdot sdot 119905

12057211

1198621119895 sdot sdot sdot 119905

12057211

11986211198981

11990512057211

1198621198941 sdot sdot sdot 119905

12057211

119862119894119895 sdot sdot sdot 119905

12057211

1198621198941198981

11990512057211

11986211989811

sdot sdot sdot 11990512057211

1198621198981119895 sdot sdot sdot 119905

12057211

11986211989811198981

]]]]]]]

]

(8)

The unweighted supermatrix can be acquired by adoptingthe interdependent relationship in group to array T120572

119888via

W = (T120572119888)1015840

=

1198631

119863119895

119863119899

11988811

11988812

11988811198981

1198881198951

1198881198952

119888119895119898119895

1198881198991

1198881198992

119888119899119898119899

1198631

119863119894

119863119899

1198881111988811198981 119888

1198941119888119894119898119894 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

W11 sdot sdot sdot W1198941 sdot sdot sdot W1198991

W1119895 sdot sdot sdot W119894119895 sdot sdot sdot W119899119895

W1119899 sdot sdot sdot W119894119899 sdot sdot sdot W119899119899

]]]]]]]]]]]]]]]]]]]]]]]

]

(9)

where W11 is exhibited by (10) and W119899119899 can be obtained inthe same way A blank space or 0 in the matrix stands for

4 International Journal of Photoenergy

independence of the group of criteria or a single criterion inrelation to other criteria Consider

W11 = (T11)1015840

=

11988811

sdot sdot sdot 1198881119894

sdot sdot sdot 11988811198981

11988811

1198881119895

11988811198981

[[[[[[[[[

[

11990512057211

11988811sdot sdot sdot 11990512057211

1198881198941sdot sdot sdot 119905

12057211

11988811989811

11990512057211

1198881119895sdot sdot sdot 11990512057211

119888119894119895sdot sdot sdot 119905

12057211

1198881198981119895

11990512057211

11988811198981

11990512057211

1198881198941198981

sdot sdot sdot 11990512057211

11988811989811198981

]]]]]]]]]

]

(10)

The third step is to calculate the weighted supermatrixThe total influence matrix of dimensions T

119863is acquired by

(12) Utilize the total degree of influence to normalize eachlevel of T

119863through (13) to receive T120572

119863

119889119894=

119899

sum

119895=1

119905119894119895

119863 119894 = 1 2 119899 (11)

T119863

=

[[[[[[[[

[

11990511

119863sdot sdot sdot 1199051119895

119863sdot sdot sdot 1199051119899

119863

1199051198941

119863sdot sdot sdot 119905119894119895

119863sdot sdot sdot 119905119894119899

119863

1199051198991

119863sdot sdot sdot 119905119899119895

119863sdot sdot sdot 119905119899119899

119863

]]]]]]]]

]

(12)

T120572119863

=

[[[[[[[[[[[[[[[

[

11990511

119863

1198891

sdot sdot sdot1199051119895

119863

1198891

sdot sdot sdot1199051119899

119863

1198891

1199051198941

119863

119889119894

sdot sdot sdot119905119894119895

119863

119889119894

sdot sdot sdot119905119894119899

119863

119889119894

1199051198991

119863

119889119899

sdot sdot sdot119905119899119895

119863

119889119899

sdot sdot sdot119905119899119899

119863

119889119899

]]]]]]]]]]]]]]]

]

=

[[[[[[[[

[

11990512057211

119863sdot sdot sdot 1199051205721119895

119863sdot sdot sdot 1199051205721119899

119863

1199051205721198941

119863sdot sdot sdot 119905120572119894119895

119863sdot sdot sdot 119905120572119894119899

119863

1199051205721198991

119863sdot sdot sdot 119905120572119899119895

119863sdot sdot sdot 119905120572119899119899

119863

]]]]]]]]

]

(13)

The weighted supermatrix W120572 can be obtained by nor-malizing T120572

119863into the unweighted supermatrix W displayed

in

W120572 = T120572119863W

=

[[[[[[[[

[

11990512057211

119863timesW11 sdot sdot sdot 119905

1205721198941

119863timesW1198941 sdot sdot sdot 119905

1205721198991

119863timesW1198991

1199051205721119895

119863timesW1119895 sdot sdot sdot 119905

120572119894119895

119863timesW119894119895 sdot sdot sdot 119905

120572119899119895

119863timesW119899119895

1199051205721119899

119863timesW1119899 sdot sdot sdot 119905

120572119894119899

119863timesW119894119899 sdot sdot sdot 119905

120572119899119899

119863timesW119899119899

]]]]]]]]

]

(14)

Fourthly receive the influential weights of DANP Theweighted supermatrixW120572 ismultiplied by itself enough timesto calculate the limit supermatrix according to the conceptof Markov Chain The influential weight of criteria can thusbe calculated by lim

119911rarrinfin(W120572)119911 The influential weights of

DANP are received by the limit supermatrix application W120572with power 119911 a large enough integer until the supermatrixW120572 has converged and turns a long-term stable supermatrixto acquire the global priority vectors

3 Results and Discussion

31 Establishing the INRM for Comprehending the Interre-lationship The DEMATEL technique is used to investigatethe problems of interdependence and feedback among tencriteria from a literature review Thereafter the study con-struction of the influence network is displayed as Figure 1The influencematrixA is displayed in the beginning (Table 1)Second the normalized influence matrix G exhibited inTable 2 can be obtained using (1) Thirdly the total influencematrixT shown in Table 3 is calculated via (3) The INRM ofinfluential interrelationship for solar farms site is eventuallyconstructed by the vector r and vector d (Table 4) derivedfrom the total influence matrix T shown in Figure 2

32 Calculating Influential Weights of Criteria for DecisionMaking DANP is utilized by this study to obtain the level ofinfluential weights of ten criteria for solar farms site selectionshown in Tables 5 6 and 7 based on the constructionof the influence network from DEMATEL The empiricalfindings present that experts pay more attention to solarradiation (119862

9) temperature (119862

10) and distance to villages

(1198627) however less on orientation (119862

3) and distance to power

lines (1198626) The outcomes reveal that the level of influential

weights is much higher in solar radiation temperature anddistance to villages More specifically solar radiation gets thehighest influential weight of 012 followed by temperature(0116) and distance to villages (0103) Moreover the level ofinfluential weight of orientation and distance to power linesis relatively lower averaging 009

When comparing criteria within dimension the influen-tial weight of agrological capacity is 0097 in the dimension ofenvironment (119863

1) Experts think area is the most important

criterion in the dimension of orography (1198632) As for location

(1198633) the influential weight of distance to villages is the

highest Solar radiation is regarded by experts as the mostimportant criterion in the dimension of climatology (119863

4)

Received results present that solar radiation (ranked top one)is the last criterion which can be neglected when improvingthe performance of solar farms Experts are much concernedwith dimension of climatology (119863

4) in that the mean (0118)

of its criteria is much higher than others in the standpoint ofdimensions

In addition this study collects comparative data fromthree different regions of China for empirical analysis by thepurposed model to reveal the improving of the performanceof solar farms site because China has one of the biggest

International Journal of Photoenergy 5

Table 1 The initial influence matrix A

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0000 2000 0467 2000 3000 2600 2533 2067 3467 34671198622

2000 0000 2600 2533 2533 2067 3000 2067 3533 30671198623

2000 2600 0000 2067 2533 2067 2533 2533 3533 35331198624

2467 2533 2533 0000 2533 2533 3000 3000 3533 26001198625

2533 2533 2533 3000 0000 2533 3000 2533 3533 30671198626

2600 2067 2533 2533 2533 0000 3000 3000 2533 25331198627

3000 2533 2067 3000 3000 2533 0000 2533 3000 35331198628

2067 2067 2067 2067 2533 2533 2533 0000 2533 25331198629

3000 3067 3067 3533 2533 2533 2533 2533 0000 400011986210

3467 3533 3533 3067 2533 2067 3067 2067 4000 0000Note Consider (1(119899(119899 minus 1)))sum119899

119894=1sum119901

119895=1(|119905119906

119894119895minus 119905119906minus1

119894119895|119905119906

119894119895) times 100 = 2362 lt 5 that is confidence is 97368 where 119906 = 15 denotes the number of experts

and 119905119906119894119895is the average influence of 119894 criterion on 119895 n denotes the number of criteria here 119899 = 10 and 119899 times 119899matrix

Table 2 The normalized direct-influence matrix G

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 1357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 1336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 1385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 1387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 1326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 1408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 1306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 1504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 1494

Table 3 The total influence matrix T

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 0357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 0336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 0385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 0387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 0326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 0408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 0306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 0504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 0494

Table 4 The sums of giving and received influences

Dimensionscriteria 119903119894(effects) 119889

119894(received influences) 119903

119894+ 119889119894(centrality) 119903

119894minus 119889119894(causality)

Environment (1198631) 4135 4403 8538 minus0268

Agrological capacity (1198621) 4135 4403 8538 minus0268

Orography (1198632) 13515 12994 26509 0521

Slope (1198622) 4441 4369 8810 0072

Orientation (1198623) 4436 4103 8539 0334

Area (1198624) 4638 4523 9161 0115

Location (1198633) 17839 17427 35267 0412

Distance to roads (1198625) 4746 4455 9202 0291

Distance to power lines (1198626) 4381 4063 8444 0319

Distance to villages (1198627) 4731 4709 9440 0023

Distance to substations (1198628) 3981 4201 8181 minus0220

Climatology (1198634) 10094 10758 20852 minus0664

Solar radiation (1198629) 5001 5486 10487 minus0484

Average temperature (11986210) 5092 5273 10365 minus0180

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

2 International Journal of Photoenergy

the influence values of criteria in INRM Subsequently tosolve the problems with interdependent criteria the influ-ential weights of solar farms site can be received via DANPderived from the basic concept of analytical network process(ANP) proposed by Saaty [7]

2 Calculation Models and Methods

21 Criteria for Solar Farms Site Selection Industrial siteselection is one of the fundamental decisions in the start-upprocess expansion or relocation GIS are adopted combiningwith other systems and methods such as systems for decisionmaking (DSS) and the method for MCDM Synergistic effectis created in conjunction with these tools which contribute tothe efficiency and quality of spatial analysis for industrial siteselection [8] Chen and Pang [9] proposed a fuzzy analyticnetwork process (FANP) to investigate critical characteris-tics of successful PV solar energy industry and examinedsuitable forms of organization for knowledge management(KM) in order to distribute existing knowledge as wellas to create new knowledge Carrion et al [10] used anenvironmental decision-support system (EDSS) for choosingoptimal sites for grid-connected PV power plantsThe systemcombined multicriteria analysis and the analytic hierarchyprocess (AHP) with GIS technology and took into accountenvironment orography location and climate factors in themeantime Moreover the combination of GIS and MCDMmethods is used to acquire the assessment of the optimalplacement of PV solar power plants The exceptional toolcan also be employed to analyze an extensive cartographicand alphanumeric database for simplifying problems to solveand promote the use of multiple criteria [11] Uyan [12]indicated that solar energy investments have been expandedquickly in recent years It is a crucial issue for huge solarfarms investments to select location for the considerationof terrain local weathering factors proximity to high trans-mission capacity lines agricultural facilities and environ-mental conservation Multiple criteria evaluation methodsare often utilized for various site selection researches Thestudy determined proper site selection for solar farms byadopting GIS and AHP Sanchez-Lozano et al [13] revealedthat great position with high percentages of potential solarradiation can host electricity generation plants via PV solarfarms The best plots suitable for installing PV solar farmsare identified by employing GIS and classified accordingto multiple evaluation aspects by means of a multicriteriamodel

By literature review criteria affecting solar farms siteselection for improving the performance of solar energyindustry are arranged as follows Solar farms site includesfour dimensions environment (119863

1) orography (119863

2) location

(1198633) and climatology (119863

4) To be specific environment is

affected by agrological capacity (1198621) orography is influenced

by slope (1198622) orientation (119862

3) and area (119862

4) location is

affected by distance to roads (1198625) distance to power lines

(1198626) distance to villages (119862

7) and distance to substations

(1198628) climatology is influenced by solar radiation (119862

9) and

average temperature (11986210)

22 DEMATEL Method The DEMATEL method is utilizedto probe into the problems of interdependent criteria forestablishing the INRM [14 15] This method has been practi-cally used in decisionmaking problems of various fields suchas vendor selection for recycledmaterial and the organic lightemitting diode technology selection [16 17]

Experts with specialty of technology for PV solar energyand experience from solar energy industry including officialsof Bureau of Energy scholars of energy engineering andmanagers of solar industry are invited to help carry out thisresearch Information required for sufficient evaluation ofsolar farms site is collected by utilizing interviews and fillingsuitable questionnaires In the questionnaires a scale of 0 12 3 and 4 presents the degree from ldquono influencerdquo to ldquoveryhigh influencerdquo In addition the confidence level for expertsis tested (97638) in this real case of fifteen experts by thisresearch

The technique is introduced as follows in the first placethe influence matrix is received by scoresThe related expertsare asked to indicate the degrees of influence among criteriathat is to point out how much the criteria affect each otherThe influence matrix A can thus be obtained Secondly thenormalized influence matrix G can be calculated by utilizing(1) and (2) to normalize A

G = 119898 sdot A (1)

119898 = min

1

max119894sum119899

119895=1

10038161003816100381610038161003816119886119894119895

10038161003816100381610038161003816

1

max119895sum119899

119894=1

10038161003816100381610038161003816119886119894119895

10038161003816100381610038161003816

(2)

Thirdly the total influence matrix T can be acquired via theformula T = G + G2 + G3 + sdot sdot sdot + G119902 = G(I minus G)

minus1 whenlimℎrarrinfin

Gℎ = [0]119899times119899

where I denotes the identity matrixThe fourth step is to use (3) and (4) to construct the INRMthrough vectors r and d (the sums of the rows and columnsseparately within the total influence matrix T = [119905

119894119895]119899times119899

)

r = [119903119894]119899times1

= [

[

119899

sum

119895=1

119905119894119895]

]119899times1

(3)

d = [119889119895]119899times1

= [

119899

sum

119894=1

119905119894119895]

1015840

1times119899

(4)

where the superscript 1015840 stands for transpose If 119903119894represents

the row sum of the 119894th row in matrix T then 119903119894reveals the

sum of direct and indirect influences of criterion 119894 on theother criteria Also if 119904

119895stands for the column sum of the

119895th column of matrix T then 119904119895displays the sum of direct

and indirect effects that criterion 119895 has obtained from theother criteria Furthermore when 119894 = 119895 the sum of therow and column aggregates (119903

119894+ 119889119894) which presents the

giving and received degree of influences that is (119903119894+ 119889119894)

shows the intensity of the important role that the 119894th criterionplays in the problem When (119903

119894minus 119889119894) is positive the 119894th

criterion influences other criteria On the contrary if (119903119894minus119889119894)

is negative other criteria influence the 119894th criterion (ie 119894thcriterion receives influence from other criteria) Therefore

International Journal of Photoenergy 3

the interdependent interrelationship among criteria can beestablished by the INRM [18 19]

23 DANP Method Comparing with [11ndash13] the advantageof the proposed model is to improve the equal weightingassumption of the traditionalmethod for solving complicatedproblems in influential weights [16] ANP was developedby Saaty [7] to reduce the limitations of the AHP and todetermine nonlinear and complicated network relationshipsTherefore after investigating the influential relationship thisstudy integrates DEMATEL with basic concept of ANP(called DEMATEL-based ANP DANP) to obtain the influ-ential weights of criteria [20] The DANP includes foursteps First build the influence network structure based onDEMATEL Second find the unweighted supermatrix Thetotal influence matrix T presented in (5) is derived fromDEMATEL

T119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T11119888

sdot sdot sdot T1119895119888

sdot sdot sdot T1119899119888

T1198941119888

sdot sdot sdot T119894119895119888

sdot sdot sdot T119894119899119888

T1198991119888

sdot sdot sdot T119899119895119888

sdot sdot sdot T119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(5)

Normalize each level of T119888by using the total degree of

influence to obtain T120572119888via

T120572119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T12057211119888

sdot sdot sdot T1205721119895119888

sdot sdot sdot T1205721119899119888

T1205721198941119888

sdot sdot sdot T120572119894119895119888

sdot sdot sdot T120572119894119899119888

T1205721198991119888

sdot sdot sdot T120572119899119895119888

sdot sdot sdot T120572119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(6)

whereT12057211119888

can be calculated through (7) and (8) by the sameway T120572119899119899

119888can be obtained

11988911

119894=

1198981

sum

119895=1

11990511

119888119894119895 119894 = 1 2 119898

1 (7)

T12057211119888

=

[[[[[[[[[[[[[[[

[

11990511

11986211

11988911

1

sdot sdot sdot11990511

1198621119895

11988911

1

sdot sdot sdot11990511

11986211198981

11988911

1

11990511

1198621198941

11988911

119894

sdot sdot sdot11990511

119862119894119895

11988911

119894

sdot sdot sdot11990511

1198621198941198981

11988911

119894

11990511

11986211989811

119889111198981

sdot sdot sdot11990511

1198621198981119895

119889111198981

sdot sdot sdot11990511

11986211989811198981

119889111198981

]]]]]]]]]]]]]]]

]

=

[[[[[[[

[

11990512057211

11986211 sdot sdot sdot 119905

12057211

1198621119895 sdot sdot sdot 119905

12057211

11986211198981

11990512057211

1198621198941 sdot sdot sdot 119905

12057211

119862119894119895 sdot sdot sdot 119905

12057211

1198621198941198981

11990512057211

11986211989811

sdot sdot sdot 11990512057211

1198621198981119895 sdot sdot sdot 119905

12057211

11986211989811198981

]]]]]]]

]

(8)

The unweighted supermatrix can be acquired by adoptingthe interdependent relationship in group to array T120572

119888via

W = (T120572119888)1015840

=

1198631

119863119895

119863119899

11988811

11988812

11988811198981

1198881198951

1198881198952

119888119895119898119895

1198881198991

1198881198992

119888119899119898119899

1198631

119863119894

119863119899

1198881111988811198981 119888

1198941119888119894119898119894 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

W11 sdot sdot sdot W1198941 sdot sdot sdot W1198991

W1119895 sdot sdot sdot W119894119895 sdot sdot sdot W119899119895

W1119899 sdot sdot sdot W119894119899 sdot sdot sdot W119899119899

]]]]]]]]]]]]]]]]]]]]]]]

]

(9)

where W11 is exhibited by (10) and W119899119899 can be obtained inthe same way A blank space or 0 in the matrix stands for

4 International Journal of Photoenergy

independence of the group of criteria or a single criterion inrelation to other criteria Consider

W11 = (T11)1015840

=

11988811

sdot sdot sdot 1198881119894

sdot sdot sdot 11988811198981

11988811

1198881119895

11988811198981

[[[[[[[[[

[

11990512057211

11988811sdot sdot sdot 11990512057211

1198881198941sdot sdot sdot 119905

12057211

11988811989811

11990512057211

1198881119895sdot sdot sdot 11990512057211

119888119894119895sdot sdot sdot 119905

12057211

1198881198981119895

11990512057211

11988811198981

11990512057211

1198881198941198981

sdot sdot sdot 11990512057211

11988811989811198981

]]]]]]]]]

]

(10)

The third step is to calculate the weighted supermatrixThe total influence matrix of dimensions T

119863is acquired by

(12) Utilize the total degree of influence to normalize eachlevel of T

119863through (13) to receive T120572

119863

119889119894=

119899

sum

119895=1

119905119894119895

119863 119894 = 1 2 119899 (11)

T119863

=

[[[[[[[[

[

11990511

119863sdot sdot sdot 1199051119895

119863sdot sdot sdot 1199051119899

119863

1199051198941

119863sdot sdot sdot 119905119894119895

119863sdot sdot sdot 119905119894119899

119863

1199051198991

119863sdot sdot sdot 119905119899119895

119863sdot sdot sdot 119905119899119899

119863

]]]]]]]]

]

(12)

T120572119863

=

[[[[[[[[[[[[[[[

[

11990511

119863

1198891

sdot sdot sdot1199051119895

119863

1198891

sdot sdot sdot1199051119899

119863

1198891

1199051198941

119863

119889119894

sdot sdot sdot119905119894119895

119863

119889119894

sdot sdot sdot119905119894119899

119863

119889119894

1199051198991

119863

119889119899

sdot sdot sdot119905119899119895

119863

119889119899

sdot sdot sdot119905119899119899

119863

119889119899

]]]]]]]]]]]]]]]

]

=

[[[[[[[[

[

11990512057211

119863sdot sdot sdot 1199051205721119895

119863sdot sdot sdot 1199051205721119899

119863

1199051205721198941

119863sdot sdot sdot 119905120572119894119895

119863sdot sdot sdot 119905120572119894119899

119863

1199051205721198991

119863sdot sdot sdot 119905120572119899119895

119863sdot sdot sdot 119905120572119899119899

119863

]]]]]]]]

]

(13)

The weighted supermatrix W120572 can be obtained by nor-malizing T120572

119863into the unweighted supermatrix W displayed

in

W120572 = T120572119863W

=

[[[[[[[[

[

11990512057211

119863timesW11 sdot sdot sdot 119905

1205721198941

119863timesW1198941 sdot sdot sdot 119905

1205721198991

119863timesW1198991

1199051205721119895

119863timesW1119895 sdot sdot sdot 119905

120572119894119895

119863timesW119894119895 sdot sdot sdot 119905

120572119899119895

119863timesW119899119895

1199051205721119899

119863timesW1119899 sdot sdot sdot 119905

120572119894119899

119863timesW119894119899 sdot sdot sdot 119905

120572119899119899

119863timesW119899119899

]]]]]]]]

]

(14)

Fourthly receive the influential weights of DANP Theweighted supermatrixW120572 ismultiplied by itself enough timesto calculate the limit supermatrix according to the conceptof Markov Chain The influential weight of criteria can thusbe calculated by lim

119911rarrinfin(W120572)119911 The influential weights of

DANP are received by the limit supermatrix application W120572with power 119911 a large enough integer until the supermatrixW120572 has converged and turns a long-term stable supermatrixto acquire the global priority vectors

3 Results and Discussion

31 Establishing the INRM for Comprehending the Interre-lationship The DEMATEL technique is used to investigatethe problems of interdependence and feedback among tencriteria from a literature review Thereafter the study con-struction of the influence network is displayed as Figure 1The influencematrixA is displayed in the beginning (Table 1)Second the normalized influence matrix G exhibited inTable 2 can be obtained using (1) Thirdly the total influencematrixT shown in Table 3 is calculated via (3) The INRM ofinfluential interrelationship for solar farms site is eventuallyconstructed by the vector r and vector d (Table 4) derivedfrom the total influence matrix T shown in Figure 2

32 Calculating Influential Weights of Criteria for DecisionMaking DANP is utilized by this study to obtain the level ofinfluential weights of ten criteria for solar farms site selectionshown in Tables 5 6 and 7 based on the constructionof the influence network from DEMATEL The empiricalfindings present that experts pay more attention to solarradiation (119862

9) temperature (119862

10) and distance to villages

(1198627) however less on orientation (119862

3) and distance to power

lines (1198626) The outcomes reveal that the level of influential

weights is much higher in solar radiation temperature anddistance to villages More specifically solar radiation gets thehighest influential weight of 012 followed by temperature(0116) and distance to villages (0103) Moreover the level ofinfluential weight of orientation and distance to power linesis relatively lower averaging 009

When comparing criteria within dimension the influen-tial weight of agrological capacity is 0097 in the dimension ofenvironment (119863

1) Experts think area is the most important

criterion in the dimension of orography (1198632) As for location

(1198633) the influential weight of distance to villages is the

highest Solar radiation is regarded by experts as the mostimportant criterion in the dimension of climatology (119863

4)

Received results present that solar radiation (ranked top one)is the last criterion which can be neglected when improvingthe performance of solar farms Experts are much concernedwith dimension of climatology (119863

4) in that the mean (0118)

of its criteria is much higher than others in the standpoint ofdimensions

In addition this study collects comparative data fromthree different regions of China for empirical analysis by thepurposed model to reveal the improving of the performanceof solar farms site because China has one of the biggest

International Journal of Photoenergy 5

Table 1 The initial influence matrix A

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0000 2000 0467 2000 3000 2600 2533 2067 3467 34671198622

2000 0000 2600 2533 2533 2067 3000 2067 3533 30671198623

2000 2600 0000 2067 2533 2067 2533 2533 3533 35331198624

2467 2533 2533 0000 2533 2533 3000 3000 3533 26001198625

2533 2533 2533 3000 0000 2533 3000 2533 3533 30671198626

2600 2067 2533 2533 2533 0000 3000 3000 2533 25331198627

3000 2533 2067 3000 3000 2533 0000 2533 3000 35331198628

2067 2067 2067 2067 2533 2533 2533 0000 2533 25331198629

3000 3067 3067 3533 2533 2533 2533 2533 0000 400011986210

3467 3533 3533 3067 2533 2067 3067 2067 4000 0000Note Consider (1(119899(119899 minus 1)))sum119899

119894=1sum119901

119895=1(|119905119906

119894119895minus 119905119906minus1

119894119895|119905119906

119894119895) times 100 = 2362 lt 5 that is confidence is 97368 where 119906 = 15 denotes the number of experts

and 119905119906119894119895is the average influence of 119894 criterion on 119895 n denotes the number of criteria here 119899 = 10 and 119899 times 119899matrix

Table 2 The normalized direct-influence matrix G

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 1357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 1336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 1385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 1387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 1326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 1408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 1306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 1504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 1494

Table 3 The total influence matrix T

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 0357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 0336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 0385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 0387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 0326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 0408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 0306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 0504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 0494

Table 4 The sums of giving and received influences

Dimensionscriteria 119903119894(effects) 119889

119894(received influences) 119903

119894+ 119889119894(centrality) 119903

119894minus 119889119894(causality)

Environment (1198631) 4135 4403 8538 minus0268

Agrological capacity (1198621) 4135 4403 8538 minus0268

Orography (1198632) 13515 12994 26509 0521

Slope (1198622) 4441 4369 8810 0072

Orientation (1198623) 4436 4103 8539 0334

Area (1198624) 4638 4523 9161 0115

Location (1198633) 17839 17427 35267 0412

Distance to roads (1198625) 4746 4455 9202 0291

Distance to power lines (1198626) 4381 4063 8444 0319

Distance to villages (1198627) 4731 4709 9440 0023

Distance to substations (1198628) 3981 4201 8181 minus0220

Climatology (1198634) 10094 10758 20852 minus0664

Solar radiation (1198629) 5001 5486 10487 minus0484

Average temperature (11986210) 5092 5273 10365 minus0180

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

International Journal of Photoenergy 3

the interdependent interrelationship among criteria can beestablished by the INRM [18 19]

23 DANP Method Comparing with [11ndash13] the advantageof the proposed model is to improve the equal weightingassumption of the traditionalmethod for solving complicatedproblems in influential weights [16] ANP was developedby Saaty [7] to reduce the limitations of the AHP and todetermine nonlinear and complicated network relationshipsTherefore after investigating the influential relationship thisstudy integrates DEMATEL with basic concept of ANP(called DEMATEL-based ANP DANP) to obtain the influ-ential weights of criteria [20] The DANP includes foursteps First build the influence network structure based onDEMATEL Second find the unweighted supermatrix Thetotal influence matrix T presented in (5) is derived fromDEMATEL

T119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T11119888

sdot sdot sdot T1119895119888

sdot sdot sdot T1119899119888

T1198941119888

sdot sdot sdot T119894119895119888

sdot sdot sdot T119894119899119888

T1198991119888

sdot sdot sdot T119899119895119888

sdot sdot sdot T119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(5)

Normalize each level of T119888by using the total degree of

influence to obtain T120572119888via

T120572119888=

1198631

119863119894

119863119899

11988811

11988812

11988811198981

1198881198941

1198881198942

119888119894119898119894

1198881198991

1198881198992

119888119899119898119899

1198631

119863119895

119863119899

1198881111988811198981 119888

1198951119888119895119898119895 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

T12057211119888

sdot sdot sdot T1205721119895119888

sdot sdot sdot T1205721119899119888

T1205721198941119888

sdot sdot sdot T120572119894119895119888

sdot sdot sdot T120572119894119899119888

T1205721198991119888

sdot sdot sdot T120572119899119895119888

sdot sdot sdot T120572119899119899119888

]]]]]]]]]]]]]]]]]]]]]]]

]

(6)

whereT12057211119888

can be calculated through (7) and (8) by the sameway T120572119899119899

119888can be obtained

11988911

119894=

1198981

sum

119895=1

11990511

119888119894119895 119894 = 1 2 119898

1 (7)

T12057211119888

=

[[[[[[[[[[[[[[[

[

11990511

11986211

11988911

1

sdot sdot sdot11990511

1198621119895

11988911

1

sdot sdot sdot11990511

11986211198981

11988911

1

11990511

1198621198941

11988911

119894

sdot sdot sdot11990511

119862119894119895

11988911

119894

sdot sdot sdot11990511

1198621198941198981

11988911

119894

11990511

11986211989811

119889111198981

sdot sdot sdot11990511

1198621198981119895

119889111198981

sdot sdot sdot11990511

11986211989811198981

119889111198981

]]]]]]]]]]]]]]]

]

=

[[[[[[[

[

11990512057211

11986211 sdot sdot sdot 119905

12057211

1198621119895 sdot sdot sdot 119905

12057211

11986211198981

11990512057211

1198621198941 sdot sdot sdot 119905

12057211

119862119894119895 sdot sdot sdot 119905

12057211

1198621198941198981

11990512057211

11986211989811

sdot sdot sdot 11990512057211

1198621198981119895 sdot sdot sdot 119905

12057211

11986211989811198981

]]]]]]]

]

(8)

The unweighted supermatrix can be acquired by adoptingthe interdependent relationship in group to array T120572

119888via

W = (T120572119888)1015840

=

1198631

119863119895

119863119899

11988811

11988812

11988811198981

1198881198951

1198881198952

119888119895119898119895

1198881198991

1198881198992

119888119899119898119899

1198631

119863119894

119863119899

1198881111988811198981 119888

1198941119888119894119898119894 119888

1198991119888119899119898119899

[[[[[[[[[[[[[[[[[[[[[[[

[

W11 sdot sdot sdot W1198941 sdot sdot sdot W1198991

W1119895 sdot sdot sdot W119894119895 sdot sdot sdot W119899119895

W1119899 sdot sdot sdot W119894119899 sdot sdot sdot W119899119899

]]]]]]]]]]]]]]]]]]]]]]]

]

(9)

where W11 is exhibited by (10) and W119899119899 can be obtained inthe same way A blank space or 0 in the matrix stands for

4 International Journal of Photoenergy

independence of the group of criteria or a single criterion inrelation to other criteria Consider

W11 = (T11)1015840

=

11988811

sdot sdot sdot 1198881119894

sdot sdot sdot 11988811198981

11988811

1198881119895

11988811198981

[[[[[[[[[

[

11990512057211

11988811sdot sdot sdot 11990512057211

1198881198941sdot sdot sdot 119905

12057211

11988811989811

11990512057211

1198881119895sdot sdot sdot 11990512057211

119888119894119895sdot sdot sdot 119905

12057211

1198881198981119895

11990512057211

11988811198981

11990512057211

1198881198941198981

sdot sdot sdot 11990512057211

11988811989811198981

]]]]]]]]]

]

(10)

The third step is to calculate the weighted supermatrixThe total influence matrix of dimensions T

119863is acquired by

(12) Utilize the total degree of influence to normalize eachlevel of T

119863through (13) to receive T120572

119863

119889119894=

119899

sum

119895=1

119905119894119895

119863 119894 = 1 2 119899 (11)

T119863

=

[[[[[[[[

[

11990511

119863sdot sdot sdot 1199051119895

119863sdot sdot sdot 1199051119899

119863

1199051198941

119863sdot sdot sdot 119905119894119895

119863sdot sdot sdot 119905119894119899

119863

1199051198991

119863sdot sdot sdot 119905119899119895

119863sdot sdot sdot 119905119899119899

119863

]]]]]]]]

]

(12)

T120572119863

=

[[[[[[[[[[[[[[[

[

11990511

119863

1198891

sdot sdot sdot1199051119895

119863

1198891

sdot sdot sdot1199051119899

119863

1198891

1199051198941

119863

119889119894

sdot sdot sdot119905119894119895

119863

119889119894

sdot sdot sdot119905119894119899

119863

119889119894

1199051198991

119863

119889119899

sdot sdot sdot119905119899119895

119863

119889119899

sdot sdot sdot119905119899119899

119863

119889119899

]]]]]]]]]]]]]]]

]

=

[[[[[[[[

[

11990512057211

119863sdot sdot sdot 1199051205721119895

119863sdot sdot sdot 1199051205721119899

119863

1199051205721198941

119863sdot sdot sdot 119905120572119894119895

119863sdot sdot sdot 119905120572119894119899

119863

1199051205721198991

119863sdot sdot sdot 119905120572119899119895

119863sdot sdot sdot 119905120572119899119899

119863

]]]]]]]]

]

(13)

The weighted supermatrix W120572 can be obtained by nor-malizing T120572

119863into the unweighted supermatrix W displayed

in

W120572 = T120572119863W

=

[[[[[[[[

[

11990512057211

119863timesW11 sdot sdot sdot 119905

1205721198941

119863timesW1198941 sdot sdot sdot 119905

1205721198991

119863timesW1198991

1199051205721119895

119863timesW1119895 sdot sdot sdot 119905

120572119894119895

119863timesW119894119895 sdot sdot sdot 119905

120572119899119895

119863timesW119899119895

1199051205721119899

119863timesW1119899 sdot sdot sdot 119905

120572119894119899

119863timesW119894119899 sdot sdot sdot 119905

120572119899119899

119863timesW119899119899

]]]]]]]]

]

(14)

Fourthly receive the influential weights of DANP Theweighted supermatrixW120572 ismultiplied by itself enough timesto calculate the limit supermatrix according to the conceptof Markov Chain The influential weight of criteria can thusbe calculated by lim

119911rarrinfin(W120572)119911 The influential weights of

DANP are received by the limit supermatrix application W120572with power 119911 a large enough integer until the supermatrixW120572 has converged and turns a long-term stable supermatrixto acquire the global priority vectors

3 Results and Discussion

31 Establishing the INRM for Comprehending the Interre-lationship The DEMATEL technique is used to investigatethe problems of interdependence and feedback among tencriteria from a literature review Thereafter the study con-struction of the influence network is displayed as Figure 1The influencematrixA is displayed in the beginning (Table 1)Second the normalized influence matrix G exhibited inTable 2 can be obtained using (1) Thirdly the total influencematrixT shown in Table 3 is calculated via (3) The INRM ofinfluential interrelationship for solar farms site is eventuallyconstructed by the vector r and vector d (Table 4) derivedfrom the total influence matrix T shown in Figure 2

32 Calculating Influential Weights of Criteria for DecisionMaking DANP is utilized by this study to obtain the level ofinfluential weights of ten criteria for solar farms site selectionshown in Tables 5 6 and 7 based on the constructionof the influence network from DEMATEL The empiricalfindings present that experts pay more attention to solarradiation (119862

9) temperature (119862

10) and distance to villages

(1198627) however less on orientation (119862

3) and distance to power

lines (1198626) The outcomes reveal that the level of influential

weights is much higher in solar radiation temperature anddistance to villages More specifically solar radiation gets thehighest influential weight of 012 followed by temperature(0116) and distance to villages (0103) Moreover the level ofinfluential weight of orientation and distance to power linesis relatively lower averaging 009

When comparing criteria within dimension the influen-tial weight of agrological capacity is 0097 in the dimension ofenvironment (119863

1) Experts think area is the most important

criterion in the dimension of orography (1198632) As for location

(1198633) the influential weight of distance to villages is the

highest Solar radiation is regarded by experts as the mostimportant criterion in the dimension of climatology (119863

4)

Received results present that solar radiation (ranked top one)is the last criterion which can be neglected when improvingthe performance of solar farms Experts are much concernedwith dimension of climatology (119863

4) in that the mean (0118)

of its criteria is much higher than others in the standpoint ofdimensions

In addition this study collects comparative data fromthree different regions of China for empirical analysis by thepurposed model to reveal the improving of the performanceof solar farms site because China has one of the biggest

International Journal of Photoenergy 5

Table 1 The initial influence matrix A

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0000 2000 0467 2000 3000 2600 2533 2067 3467 34671198622

2000 0000 2600 2533 2533 2067 3000 2067 3533 30671198623

2000 2600 0000 2067 2533 2067 2533 2533 3533 35331198624

2467 2533 2533 0000 2533 2533 3000 3000 3533 26001198625

2533 2533 2533 3000 0000 2533 3000 2533 3533 30671198626

2600 2067 2533 2533 2533 0000 3000 3000 2533 25331198627

3000 2533 2067 3000 3000 2533 0000 2533 3000 35331198628

2067 2067 2067 2067 2533 2533 2533 0000 2533 25331198629

3000 3067 3067 3533 2533 2533 2533 2533 0000 400011986210

3467 3533 3533 3067 2533 2067 3067 2067 4000 0000Note Consider (1(119899(119899 minus 1)))sum119899

119894=1sum119901

119895=1(|119905119906

119894119895minus 119905119906minus1

119894119895|119905119906

119894119895) times 100 = 2362 lt 5 that is confidence is 97368 where 119906 = 15 denotes the number of experts

and 119905119906119894119895is the average influence of 119894 criterion on 119895 n denotes the number of criteria here 119899 = 10 and 119899 times 119899matrix

Table 2 The normalized direct-influence matrix G

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 1357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 1336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 1385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 1387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 1326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 1408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 1306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 1504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 1494

Table 3 The total influence matrix T

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 0357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 0336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 0385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 0387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 0326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 0408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 0306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 0504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 0494

Table 4 The sums of giving and received influences

Dimensionscriteria 119903119894(effects) 119889

119894(received influences) 119903

119894+ 119889119894(centrality) 119903

119894minus 119889119894(causality)

Environment (1198631) 4135 4403 8538 minus0268

Agrological capacity (1198621) 4135 4403 8538 minus0268

Orography (1198632) 13515 12994 26509 0521

Slope (1198622) 4441 4369 8810 0072

Orientation (1198623) 4436 4103 8539 0334

Area (1198624) 4638 4523 9161 0115

Location (1198633) 17839 17427 35267 0412

Distance to roads (1198625) 4746 4455 9202 0291

Distance to power lines (1198626) 4381 4063 8444 0319

Distance to villages (1198627) 4731 4709 9440 0023

Distance to substations (1198628) 3981 4201 8181 minus0220

Climatology (1198634) 10094 10758 20852 minus0664

Solar radiation (1198629) 5001 5486 10487 minus0484

Average temperature (11986210) 5092 5273 10365 minus0180

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

4 International Journal of Photoenergy

independence of the group of criteria or a single criterion inrelation to other criteria Consider

W11 = (T11)1015840

=

11988811

sdot sdot sdot 1198881119894

sdot sdot sdot 11988811198981

11988811

1198881119895

11988811198981

[[[[[[[[[

[

11990512057211

11988811sdot sdot sdot 11990512057211

1198881198941sdot sdot sdot 119905

12057211

11988811989811

11990512057211

1198881119895sdot sdot sdot 11990512057211

119888119894119895sdot sdot sdot 119905

12057211

1198881198981119895

11990512057211

11988811198981

11990512057211

1198881198941198981

sdot sdot sdot 11990512057211

11988811989811198981

]]]]]]]]]

]

(10)

The third step is to calculate the weighted supermatrixThe total influence matrix of dimensions T

119863is acquired by

(12) Utilize the total degree of influence to normalize eachlevel of T

119863through (13) to receive T120572

119863

119889119894=

119899

sum

119895=1

119905119894119895

119863 119894 = 1 2 119899 (11)

T119863

=

[[[[[[[[

[

11990511

119863sdot sdot sdot 1199051119895

119863sdot sdot sdot 1199051119899

119863

1199051198941

119863sdot sdot sdot 119905119894119895

119863sdot sdot sdot 119905119894119899

119863

1199051198991

119863sdot sdot sdot 119905119899119895

119863sdot sdot sdot 119905119899119899

119863

]]]]]]]]

]

(12)

T120572119863

=

[[[[[[[[[[[[[[[

[

11990511

119863

1198891

sdot sdot sdot1199051119895

119863

1198891

sdot sdot sdot1199051119899

119863

1198891

1199051198941

119863

119889119894

sdot sdot sdot119905119894119895

119863

119889119894

sdot sdot sdot119905119894119899

119863

119889119894

1199051198991

119863

119889119899

sdot sdot sdot119905119899119895

119863

119889119899

sdot sdot sdot119905119899119899

119863

119889119899

]]]]]]]]]]]]]]]

]

=

[[[[[[[[

[

11990512057211

119863sdot sdot sdot 1199051205721119895

119863sdot sdot sdot 1199051205721119899

119863

1199051205721198941

119863sdot sdot sdot 119905120572119894119895

119863sdot sdot sdot 119905120572119894119899

119863

1199051205721198991

119863sdot sdot sdot 119905120572119899119895

119863sdot sdot sdot 119905120572119899119899

119863

]]]]]]]]

]

(13)

The weighted supermatrix W120572 can be obtained by nor-malizing T120572

119863into the unweighted supermatrix W displayed

in

W120572 = T120572119863W

=

[[[[[[[[

[

11990512057211

119863timesW11 sdot sdot sdot 119905

1205721198941

119863timesW1198941 sdot sdot sdot 119905

1205721198991

119863timesW1198991

1199051205721119895

119863timesW1119895 sdot sdot sdot 119905

120572119894119895

119863timesW119894119895 sdot sdot sdot 119905

120572119899119895

119863timesW119899119895

1199051205721119899

119863timesW1119899 sdot sdot sdot 119905

120572119894119899

119863timesW119894119899 sdot sdot sdot 119905

120572119899119899

119863timesW119899119899

]]]]]]]]

]

(14)

Fourthly receive the influential weights of DANP Theweighted supermatrixW120572 ismultiplied by itself enough timesto calculate the limit supermatrix according to the conceptof Markov Chain The influential weight of criteria can thusbe calculated by lim

119911rarrinfin(W120572)119911 The influential weights of

DANP are received by the limit supermatrix application W120572with power 119911 a large enough integer until the supermatrixW120572 has converged and turns a long-term stable supermatrixto acquire the global priority vectors

3 Results and Discussion

31 Establishing the INRM for Comprehending the Interre-lationship The DEMATEL technique is used to investigatethe problems of interdependence and feedback among tencriteria from a literature review Thereafter the study con-struction of the influence network is displayed as Figure 1The influencematrixA is displayed in the beginning (Table 1)Second the normalized influence matrix G exhibited inTable 2 can be obtained using (1) Thirdly the total influencematrixT shown in Table 3 is calculated via (3) The INRM ofinfluential interrelationship for solar farms site is eventuallyconstructed by the vector r and vector d (Table 4) derivedfrom the total influence matrix T shown in Figure 2

32 Calculating Influential Weights of Criteria for DecisionMaking DANP is utilized by this study to obtain the level ofinfluential weights of ten criteria for solar farms site selectionshown in Tables 5 6 and 7 based on the constructionof the influence network from DEMATEL The empiricalfindings present that experts pay more attention to solarradiation (119862

9) temperature (119862

10) and distance to villages

(1198627) however less on orientation (119862

3) and distance to power

lines (1198626) The outcomes reveal that the level of influential

weights is much higher in solar radiation temperature anddistance to villages More specifically solar radiation gets thehighest influential weight of 012 followed by temperature(0116) and distance to villages (0103) Moreover the level ofinfluential weight of orientation and distance to power linesis relatively lower averaging 009

When comparing criteria within dimension the influen-tial weight of agrological capacity is 0097 in the dimension ofenvironment (119863

1) Experts think area is the most important

criterion in the dimension of orography (1198632) As for location

(1198633) the influential weight of distance to villages is the

highest Solar radiation is regarded by experts as the mostimportant criterion in the dimension of climatology (119863

4)

Received results present that solar radiation (ranked top one)is the last criterion which can be neglected when improvingthe performance of solar farms Experts are much concernedwith dimension of climatology (119863

4) in that the mean (0118)

of its criteria is much higher than others in the standpoint ofdimensions

In addition this study collects comparative data fromthree different regions of China for empirical analysis by thepurposed model to reveal the improving of the performanceof solar farms site because China has one of the biggest

International Journal of Photoenergy 5

Table 1 The initial influence matrix A

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0000 2000 0467 2000 3000 2600 2533 2067 3467 34671198622

2000 0000 2600 2533 2533 2067 3000 2067 3533 30671198623

2000 2600 0000 2067 2533 2067 2533 2533 3533 35331198624

2467 2533 2533 0000 2533 2533 3000 3000 3533 26001198625

2533 2533 2533 3000 0000 2533 3000 2533 3533 30671198626

2600 2067 2533 2533 2533 0000 3000 3000 2533 25331198627

3000 2533 2067 3000 3000 2533 0000 2533 3000 35331198628

2067 2067 2067 2067 2533 2533 2533 0000 2533 25331198629

3000 3067 3067 3533 2533 2533 2533 2533 0000 400011986210

3467 3533 3533 3067 2533 2067 3067 2067 4000 0000Note Consider (1(119899(119899 minus 1)))sum119899

119894=1sum119901

119895=1(|119905119906

119894119895minus 119905119906minus1

119894119895|119905119906

119894119895) times 100 = 2362 lt 5 that is confidence is 97368 where 119906 = 15 denotes the number of experts

and 119905119906119894119895is the average influence of 119894 criterion on 119895 n denotes the number of criteria here 119899 = 10 and 119899 times 119899matrix

Table 2 The normalized direct-influence matrix G

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 1357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 1336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 1385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 1387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 1326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 1408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 1306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 1504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 1494

Table 3 The total influence matrix T

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 0357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 0336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 0385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 0387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 0326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 0408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 0306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 0504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 0494

Table 4 The sums of giving and received influences

Dimensionscriteria 119903119894(effects) 119889

119894(received influences) 119903

119894+ 119889119894(centrality) 119903

119894minus 119889119894(causality)

Environment (1198631) 4135 4403 8538 minus0268

Agrological capacity (1198621) 4135 4403 8538 minus0268

Orography (1198632) 13515 12994 26509 0521

Slope (1198622) 4441 4369 8810 0072

Orientation (1198623) 4436 4103 8539 0334

Area (1198624) 4638 4523 9161 0115

Location (1198633) 17839 17427 35267 0412

Distance to roads (1198625) 4746 4455 9202 0291

Distance to power lines (1198626) 4381 4063 8444 0319

Distance to villages (1198627) 4731 4709 9440 0023

Distance to substations (1198628) 3981 4201 8181 minus0220

Climatology (1198634) 10094 10758 20852 minus0664

Solar radiation (1198629) 5001 5486 10487 minus0484

Average temperature (11986210) 5092 5273 10365 minus0180

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

International Journal of Photoenergy 5

Table 1 The initial influence matrix A

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0000 2000 0467 2000 3000 2600 2533 2067 3467 34671198622

2000 0000 2600 2533 2533 2067 3000 2067 3533 30671198623

2000 2600 0000 2067 2533 2067 2533 2533 3533 35331198624

2467 2533 2533 0000 2533 2533 3000 3000 3533 26001198625

2533 2533 2533 3000 0000 2533 3000 2533 3533 30671198626

2600 2067 2533 2533 2533 0000 3000 3000 2533 25331198627

3000 2533 2067 3000 3000 2533 0000 2533 3000 35331198628

2067 2067 2067 2067 2533 2533 2533 0000 2533 25331198629

3000 3067 3067 3533 2533 2533 2533 2533 0000 400011986210

3467 3533 3533 3067 2533 2067 3067 2067 4000 0000Note Consider (1(119899(119899 minus 1)))sum119899

119894=1sum119901

119895=1(|119905119906

119894119895minus 119905119906minus1

119894119895|119905119906

119894119895) times 100 = 2362 lt 5 that is confidence is 97368 where 119906 = 15 denotes the number of experts

and 119905119906119894119895is the average influence of 119894 criterion on 119895 n denotes the number of criteria here 119899 = 10 and 119899 times 119899matrix

Table 2 The normalized direct-influence matrix G

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 1357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 1336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 1385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 1387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 1326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 1408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 1306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 1504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 1494

Table 3 The total influence matrix T

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0336 0396 0331 0409 0430 0389 0436 0383 0521 05051198622

0423 0357 0416 0448 0441 0396 0475 0407 0554 05241198623

0422 0437 0336 0434 0441 0395 0461 0419 0553 05361198624

0451 0450 0428 0385 0457 0424 0492 0449 0572 05291198625

0462 0459 0437 0486 0387 0432 0501 0444 0584 05531198626

0434 0416 0408 0442 0436 0326 0470 0429 0518 05021198627

0475 0458 0421 0485 0478 0431 0408 0442 0568 05641198628

0387 0384 0365 0396 0404 0375 0422 0306 0478 04631198629

0496 0495 0471 0522 0487 0450 0510 0463 0504 060311986210

0516 0516 0490 0516 0495 0444 0533 0457 0633 0494

Table 4 The sums of giving and received influences

Dimensionscriteria 119903119894(effects) 119889

119894(received influences) 119903

119894+ 119889119894(centrality) 119903

119894minus 119889119894(causality)

Environment (1198631) 4135 4403 8538 minus0268

Agrological capacity (1198621) 4135 4403 8538 minus0268

Orography (1198632) 13515 12994 26509 0521

Slope (1198622) 4441 4369 8810 0072

Orientation (1198623) 4436 4103 8539 0334

Area (1198624) 4638 4523 9161 0115

Location (1198633) 17839 17427 35267 0412

Distance to roads (1198625) 4746 4455 9202 0291

Distance to power lines (1198626) 4381 4063 8444 0319

Distance to villages (1198627) 4731 4709 9440 0023

Distance to substations (1198628) 3981 4201 8181 minus0220

Climatology (1198634) 10094 10758 20852 minus0664

Solar radiation (1198629) 5001 5486 10487 minus0484

Average temperature (11986210) 5092 5273 10365 minus0180

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

6 International Journal of Photoenergy

Table 5 The unweighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

1000 1000 1000 1000 1000 1000 1000 1000 1000 10001198622

0348 0292 0362 0356 0332 0329 0336 0336 0333 03391198623

0292 0341 0278 0339 0316 0322 0309 0319 0317 03221198624

0360 0367 0360 0305 0352 0349 0355 0346 0351 03391198625

0262 0257 0257 0251 0220 0263 0272 0268 0255 02571198626

0237 0230 0230 0233 0245 0196 0245 0249 0236 02301198627

0266 0276 0269 0270 0284 0283 0232 0280 0267 02761198628

0234 0237 0244 0246 0252 0258 0251 0203 0242 02371198629

0508 0514 0508 0520 0514 0508 0502 0508 0455 056211986210

0492 0486 0492 0480 0486 0492 0498 0492 0545 0438

Table 6 The weighted supermatrix

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0081 0096 0096 0096 0099 0099 0099 0099 0100 01001198622

0096 0080 0099 0097 0096 0095 0097 0097 0099 01011198623

0080 0093 0076 0093 0091 0093 0089 0092 0094 00961198624

0099 0100 0098 0083 0102 0101 0103 0100 0105 01011198625

0104 0100 0100 0098 0082 0099 0102 0100 0097 00981198626

0094 0090 0090 0091 0092 0074 0092 0093 0090 00881198627

0105 0108 0105 0105 0107 0106 0087 0105 0102 01051198628

0093 0092 0095 0096 0094 0097 0094 0076 0092 00901198629

0126 0124 0123 0126 0122 0121 0119 0120 0101 012411986210

0122 0118 0119 0116 0115 0117 0118 0117 0121 0097

Table 7 The stable matrix of DANP

Criteria 1198621

1198622

1198623

1198624

1198625

1198626

1198627

1198628

1198629

11986210

1198621

0097 0097 0097 0097 0097 0097 0097 0097 0097 00971198622

0096 0096 0096 0096 0096 0096 0096 0096 0096 00961198623

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198624

0099 0099 0099 0099 0099 0099 0099 0099 0099 00991198625

0098 0098 0098 0098 0098 0098 0098 0098 0098 00981198626

0090 0090 0090 0090 0090 0090 0090 0090 0090 00901198627

0103 0103 0103 0103 0103 0103 0103 0103 0103 01031198628

0092 0092 0092 0092 0092 0092 0092 0092 0092 00921198629

0120 0120 0120 0120 0120 0120 0120 0120 0120 012011986210

0116 0116 0116 0116 0116 0116 0116 0116 0116 0116

markets of solar industry in the world Table 8 shows the inte-grated values by utilizing simple additive weighting (SAW)method to receive the total performances of three regionsselected in China The empirical results present that westernChina has the highest total performance level It is followedby eastern and southern China with this regard Thereforedecision makers of solar farms site selection are suggestedto take western China as an example when improving theperformance of solar farms site according to the decisionmodel provided by this research

33 Implication and Discussion Discussion of empiricalresults and innovation strategies for improving the perfor-

mance for solar farms site is presented as follows In the firstplace the influential relationships within solar farms suggestthat what administrators should improve first is orography(1198632) for enhancing the performance for solar farms based

on INRM built by DEMATEL It is meaningful to improveother dimensions after having an excellent geomorphologicalsolar farms base It should be well located on an even andfacing south land to take advantage of natural resourceefficiently

Second the most important criterion found by DANPwhen improving solar farms is solar radiation (119862

9) whose

influential weight equals 012 It plays a significant role in theeffective functioning of a prosperous solar energy industry

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

International Journal of Photoenergy 7

Table 8 Influential weights of solar farms site and performances of selected regions

Dimensionscriteria Local weights Global weights Eastern China Western China Southern ChinaEnvironment (119863

1) 0097 3667 5667 2400

Agrological capacity (1198621) 1000 0097 3667 5667 2400

Orography (1198632) 0285 4529 6051 3043

Slope (1198622) 0336 0096 3667 5333 2800

Orientation (1198623) 0315 0090 6400 8400 4400

Area (1198624) 0348 0099 3667 5600 2400

Location (1198633) 0383 3548 5144 3336

Distance to roads (1198625) 0256 0098 3333 5667 3200

Distance to power lines (1198626) 0234 0090 3800 6400 4200

Distance to villages (1198627) 0270 0103 3667 3333 3667

Distance to substations (1198628) 0240 0092 3400 5400 2267

Climatology (1198634) 0235 5193 7336 4087

Solar radiation (1198629) 0508 0120 4800 7467 3333

Average temperature (11986210) 0492 0116 5600 7200 4867

Total performances mdash mdash 4227 5969 3339

Goal Optimal solar farms site

Dimensions Environment (D1) Orography (D2)

Slope (C2)

CriteriaOrientation (C3)

Outer-dependentArea (C4)

Dimensions Climatology (D4)

Distance to roads (C5) Solar radiation (C9)

Distance to power lines (C6)Average temperature (C10)

CriteriaDistance to villages (C7)Distance to substations (C8)

Eastern China Western China Southern ChinaAlternatives

Agrological capacity (C1)

Location (D3)

Interdependent

Interdependent Interdependent

Interdependent

Figure 1 Analytic framework for influence network of solar farms site

To follow the arc of the sun for generating the optimal amountof power solar panels are typically mounted on rotatingtowers Therefore solar farms should better situate panels tomake the most of the available solar radiation

4 Conclusions

The proposed hybrid MCDM model based on GIS can beapplied by managers of solar energy industry worldwideThey can adjust the influential weights of the ten criteria

according to the situations of various countries to obtainvaluable information for decision making when improvingthe performance of solar farms Moreover they can select apotential base to evaluate if it is suitable or not

Furthermore only few preceding study attempts areconcerned about the interdependent interrelationship amongcriteria and the influential weights of criteriaThis study thusproposes a hybrid MCDMmodel based on GIS and exploresthe perspectives of employing experts for examining theseissues for solar farms Associating past theoretical research

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

8 International Journal of Photoenergy

climatology ( C9 C10)

orography ( C2 C3 C4)location (C5 C6 C7 C8)

D1 (8538 minus0268)

D2 (26509 0521)

5000 10000 15000 20000 25000 30000 35000 40000

0600

0400

0200

0000

0000

minus0400

minus0600

minus 0800D4 (20852 minus0664)

minus0200

environment (C1)

D3 (35267 0412)

ri + di

r iminusdi

Figure 2 The INRM of influential relationships within solar farms

with opinions of professional and experienced experts makesthe proposed model a more suitable tool for improving solarfarms site selection It is not provided by preceding studyattempts This study mainly utilizes a hybrid MCDM modelbased on GIS for exploring solar farms site so some criticalfactors (eg incentive from government and purchase pricefor electricity from utility) can be included by taking extradimensions into consideration to make the research morecomplete in the future

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This work was supported in part by the National ScienceCouncil in Taiwan under the project title Caltech-TaiwanCollaboration on Energy Research-Uncertainty Mitigationfor Renewable Energy Integration Project no NSC 101-3113-P-008-001

References

[1] N S Lewis and D G Nocera ldquoPowering the planet chem-ical challenges in solar energy utilizationrdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 103 no 43 pp 15729ndash15735 2006

[2] MOEA ldquoThe analysis and investment opportunities in pho-tovoltaic industryrdquo Department of Investment Services Min-istry of Economic Affairs (MOEA) 2008 httpwwwdoismoeagovtw

[3] M Socorro Garcıa-Cascales M Teresa Lamata and J MiguelSanchez-Lozano ldquoEvaluation of photovoltaic cells in a multi-criteria decision making processrdquo Annals of OperationsResearch vol 199 no 1 pp 373ndash391 2012

[4] B Parida S Iniyan and R Goic ldquoA review of solar photovoltaictechnologiesrdquo Renewable and Sustainable Energy Reviews vol15 no 3 pp 1625ndash1636 2011

[5] M A Bhaskar S S Dash R Magdal et al ldquoApplication ofintegrated wind energy conversion system (WECS) and pho-tovoltaic (PV) solar farm as STATCOM to regulate grid voltageduring night timerdquo Energy Procedia vol 14 pp 1536ndash1541 2012

[6] H Hodson ldquoGiant solar farm uses molten salt to keep powercomingrdquo New Scientist vol 222 no 2965 p 22 2014

[7] T L Saaty Decision Making with Dependence and FeedbackTheAnalytic Network Process RWSPublications Pittsburgh PaUSA 1996

[8] A Rikalovic I Cosic and D Lazarevic ldquoGIS based multi-cri-teria analysis for industrial site selectionrdquo Procedia Engineeringvol 69 pp 1054ndash1063 2014

[9] H H Chen and C Pang ldquoOrganizational forms for knowledgemanagement in photovoltaic solar energy industryrdquoKnowledge-Based Systems vol 23 no 8 pp 924ndash933 2010

[10] J A Carrion A E Estrella F ADolsM Z ToroMRodrıguezand A R Ridao ldquoEnvironmental decision-support systemsfor evaluating the carrying capacity of land areas optimalsite selection for grid-connected photovoltaic power plantsrdquoRenewable and Sustainable Energy Reviews vol 12 no 9 pp2358ndash2380 2008

[11] JM Sanchez-Lozano J Teruel-Solano P L Soto-Elvira andMSocorro Garcıa-Cascales ldquoGeographical Information Systems(GIS) and Multi-Criteria Decision Making (MCDM) methodsfor the evaluation of solar farms locations case study in south-eastern Spainrdquo Renewable and Sustainable Energy Reviews vol24 pp 544ndash556 2013

[12] M Uyan ldquoGIS-based solar farms site selection using analytichierarchy process (AHP) in Karapinar region KonyaTurkeyrdquoRenewable and Sustainable Energy Reviews vol 28 pp 11ndash172013

[13] J M Sanchez-Lozano C H Antunes M S Garcıa-Cascalesand L C Dias ldquoGIS-based photovoltaic solar farms site selec-tion using ELECTRE-TRI evaluating the case for TorrePachecoMurcia Southeast of SpainrdquoRenewable Energy vol 66pp 478ndash494 2014

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

International Journal of Photoenergy 9

[14] C H Chen and G H Tzeng ldquoCreating the aspired intelligentassessment systems for teaching materialsrdquo Expert Systems withApplications vol 38 no 10 pp 12168ndash12179 2011

[15] E Fontela and A Gabus The DEMATEL Observer DEMATEL1976 Report Battelle Geneva Research Centre Geneva Switzer-land 1976

[16] C- Hsu F Wang and G Tzeng ldquoThe best vendor selectionfor conducting the recycled material based on a hybrid MCDMmodel combining DANP with VIKORrdquo Resources Conserva-tion and Recycling vol 66 pp 95ndash111 2012

[17] Y Shen G T R Lin and G Tzeng ldquoCombined DEMATELtechniques with novel MCDM for the organic light emittingdiode technology selectionrdquo Expert Systems with Applicationsvol 38 no 3 pp 1468ndash1481 2011

[18] J J H Liou G Tzeng and H Chang ldquoAirline safety measure-ment using a hybrid modelrdquo Journal of Air Transport Manage-ment vol 13 no 4 pp 243ndash249 2007

[19] G Tzeng C Chiang and C Li ldquoEvaluating intertwinedeffects in e-learning programs A novel hybrid MCDM modelbased on factor analysis and DEMATELrdquo Expert Systems withApplications vol 32 no 4 pp 1028ndash1044 2007

[20] Y P Ou Yang H M Shieh J D Leu and G H Tzeng ldquoA novelhybrid MCDM model combined with DEMATEL and ANPwith applicationsrdquo International Journal of Operations Researchvol 5 no 3 pp 160ndash168 2008

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Inorganic ChemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal ofPhotoenergy

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Carbohydrate Chemistry

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Physical Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom

Analytical Methods in Chemistry

Journal of

Volume 2014

Bioinorganic Chemistry and ApplicationsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

SpectroscopyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Medicinal ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chromatography Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Applied ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Theoretical ChemistryJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Spectroscopy

Analytical ChemistryInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Quantum Chemistry

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Organic Chemistry International

ElectrochemistryInternational Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CatalystsJournal of