DesertDesert

Online at http://desert.ut.ac.ir

Desert 19-2 (2014) 111-119

Solar desalination plant site suitability through composingdecision-making systems and fuzzy logic in Iran (using the desert

areas approach)H. Paktinata*, H.A. Farajia, A. Rahimi Kianb

a Faculty of Geography, University of Tehran ,Tehran, Iranb Faculty of Electrical & Computer Engineering, University of Tehran, Tehran, Iran

Received: 5 November 2013; Received in revised form: 24 August 2014; Accepted: 30 September 2014

Abstract

Freshwater resources represent around 3% of all water on Earth, and less than 1% of that is available. Consideringcurrent conditions, as well as future predictions of need, freshwater resources cannot meet human needs. Thus,sweetening of the brackish water can be performed to provide freshwater for human use. Solar energy, because ofIran's climatic conditions, may be used for sweetening the brackish water. The objective of this study was to surveythe lands using groundwater resources for the installation of solar desalinations. According to the goal, criteria andindicators were identified by the analyses of previous studies data, Delphi method and the internal structure ofmeasurements and indicators determined using the DEMATEL technique. Then, indicators were weighted using theanalytic network process (ANP) method, and indicators suitable membership functions were defined using fuzzylogic. Indicators have been combined by the means of minimum function. Eventually, the areas were classified intofour classes. The results showed that among 250 scope studies, 20 scopes have been put in class one, and these arelocated in five provinces. The results also indicated that among the provinces of Iran, Yazd is located in an area withthe highest percentage of class one, and is ranked first. Comparisons of the obtained results with the climaticconditions of Iran confirmed the results.

Keywords: Solar desalination; ANP; Fuzzy logic; DEMATEL; Delphi

1. Introduction

More than two thirds of the Earths surface iscovered by water. Oceans contain about 97% ofthe Earths water, and only about 3% of allwater on the Earth is freshwater; less than onepercent of that is available for human uses.Generally, the human need for water is doubledevery 20 years and, currently, most of the waterresources have been consumed in the countriesof the Middle East. It is estimated that thepopulation will be increased by 50% in Africa,25% in Asia and 14% in America in the next 20years (FAOSTAT, 2000). It is clear that thepopulation in developing countries such as Iranhas increased more than the average level of

Corresponding author. Tel.: +98 913 9527062,Fax: +98 21 66404366.E-mail address: paktinat.hadi@gmail.com

Asia (World Bank, 1990-2010). Therefore,considering current conditions as well as futurepredictions, current freshwater resources arelimited for future use.

Furthermore, in many areas, due to climaticand geological conditions, freshwater may notbe available. Humans should therefore supplythe missing portion of water by sweeteningbrackish water. Desalination refers to a processthat separates salt from brackish water (saltywater). Desalination is performed for differentpurposes including drinking water (freshwater),agriculture, industry, and the militaryworldwide. Water obtained by the means ofdesalination can be an important water resourcefor many countries of the Middle East, PersianGulf and northern Africa.

Water and energy are two human needs thatare always going to be together, and with the

Paktinat et al. / Desert 19-2 (2014) 111-119112

increase in population their correlations areincreased (Gude et al., 2011). The solardesalination process is energy consuming(Rodriguez, 2003). Therefore, the increased useof desalination plants causes one main problemfor energy consumption. Energy may besupplied from the electrical, mechanical andheating resources. The benefits and efficiency ofwater desalination strongly depend on the costsof energy consumption (Sagie, 2001). Energy isthe biggest variable for the cost of a waterdesalination system, while more than one-thirdand even more than half of the cost of waterproduction belongs to it (Chealhry, 2003).

Factors that have more of an influence on thecost of water sweetening are the quality of inputwater (level of salinity), quality of producedwater, costs of energy, and scale (Alatigi et al.,1999; Dore, 2005). Economic factors are themost important factor determining final successand the development of water desalinationsystems. Worldwide, 77% of fuels are suppliedby fossil resources, and they produce pollutantgases and greenhouse gases in the conversionprocess, which damages ozone layer, isdangerous for the environment, produces heat,and causes an increase in the temperature of theEarth (Rashidi & Gharib, 2011). Furthermore,resources of fossil fuels are reducing whilst theirprices are increasing, so the use of fossil fuelsfor a water desalination system cannot be aneconomical process. Solar energy has the largestcorrelation with the need for water as comparedto the other renewable energies because it is themain factor for water reduction (Gastli et al.,2010).

Solar energy is a natural energy and ischeaper than fossil fuels, but primaryinvestments for the collection and conversion ofsolar energy into a usable form are high (Sagie,2001). Solar energy is one of the best renewableenergies, and Iran has good climatic conditionsfor its development. The amount of sunradiation in Iran is almost 5kwh/m2/year, andIran has sun radiation 300 days per year in morethan 90% of its areas (Saghafi, 2009). The dataused in this study were obtained from NASA1.These data extend the temporal coverage of thesolar data from approximately 11 years to 22 ormore years (e.g., July 1983 through June 2005)and related to the (SSE2) NASA project.

The objective of this study was to survey thelands for the installation of solar desalinationusing knowledge-based systems (Delphi methodand questionnaires), the DEMATEL technique,

http://eosweb.larc.nasa.gov1nergyeolarsSurface meteorology and2

multi-criteria decision-making systems andfuzzy logic. Effective site selection for theinstallation of solar desalinations depends onseveral factors, including morphology,hydrology and sun radiation.

Geographical information system (GIS) witha multi-criteria decision-making process(MCDM), a spatial decision support system andfuzzy logic can be used for the selection ofoptimum areas for the installation of solardesalinations.

Many investigations have been performedusing an analytic network process, but multi-criteria decision-making systems, structuremaking systems and incorporated knowledgesystems are not used frequently. Only oneinvestigation has been done with the help ofmulti-criteria decision-making systems andgeographical information systems to determinesuitable areas for the installation of solardesalinations using groundwater. Salim groupedEgypt into several classes (Salim, 2012). Rujula(2009) presented the best energy resource inMauritania using multi-criteria analysis, and hasfound that the best resources for energy inMauritania for water desalination are wind andsolar energy, respectively (Rujula & Dia, 2010).

The objective of this study was to survey thelands using groundwater resources for theinstallation of solar desalinations with the helpof geographical information systems and multi-criteria decision-making systems.

2. Material and methods

The area of this study is located in Iran, becauseall of the areas in Iran do not need the waterdesalination and considering two indicatorsincluding precipitation rate and salinity level ofgroundwater, some areas of Iran withprecipitation rates higher than threshold of dryareas and those with non-salty water wereremoved from this investigation. According tothe new definition of desert, with the exceptionof a narrow strip in the north of Iran, other partsof the country encounter the problem ofdesertification (Shakerian et al., 2011). Someclimatologists used the annual meanprecipitation indicator to obtain different valuesfor the separation of dry areas from non-dryones. Based on Fink's findings, areas with anannual mean precipitation less than 500 mm aredry, and those with an annual mean precipitationless than 250 mm are deserts (Jazirei, 1991).Gansen noted that the areas with an annual meanprecipitation less than 350 mm are located in dryareas, and if this value is less than 125 mm,these areas are deserts (Hossein Zadeh, 1998).

Paktinat et al. / Desert 19-2 (2014) 111-119 113

Another drought indicator is the reconnaissancedrought index (RDI). This index is based bothon precipitation (P) and potentialevapotranspiration (Dastorani et al., 2011).

Considering Irans climatic conditions, theareas studied with an annual mean precipitationmore than 300mm were removed. Isohyet line(Iran Water Resources Management Company)was used to identify these areas. On the otherhand, according to the World HealthOrganization (WHO), the allowable limit for thesalinity of freshwater (drinking water) is 500ppm, and for specific cases this value is 1000ppm (Eltawil et al., 2009). Thus, groundwaterwith a salinity of less than 1000 ppm wasremoved from the study. Eventually, because ofdata availability limitations, the areas withoutsufficient data were removed, and finally 250areas out of 609 in Iran concerning the WaterResources Management Company wereselected. Figure 1 shows these areas.

Fig. 1. Case study boundary

Determination of the appropriate place forthe installation of solar desalination depends onfull and correct knowledge of effective factorsand means of selection in the current study.Previous investigations and the Delphi methodwere used to identify the effective factors on theclassification of water desalination systems.Meriam Salim (2012) used sun radiation, depthof groundwater, salinity of groundwater anddistances from the Nile Delta and valley todetermine suitable places for the installation ofsolar desalination using groundwater.

In this study, regarding the distance from theNile valley, two indicators, including thedistance from communication lines (roads) andresidential centres were obtained and. as hasbeen mentioned by Salim (2012), this is due to

the reduction of population compression,residential places, and communication linesbecause of an increasing distance from the Nilevalley (according to Egypts conditions). Charbi(2011) used a slope indicator for surveying thesuitability of places (areas) for waterdesalination systems (Charbi & Gastli, 2011).By considering Irans conditions and nationaleffective factors in this study in addition to theuse of resources, the Delphi method was used toincorporate the experts viewpoints and identifynew factors.

The Delphi method is a structured processfor the collection and classification of existingdata with respect to some groups of experts,while distributing a questionnaire to them andtaking control feedback from the receivedanswers. The Delphi method is used to discovercreative and reliable ideas or supply suitableinformation for making decisions (Jafari &Montazer, 2007).

Thus, unlike scanning investigations, thereliability of the Delphi method does not dependon the number of participants in theinvestigation, but its reliability depends on thescientific reliability of specialists that participatein the investigation. In the current investigation,three specialists in the Delphi method attendedand their viewpoints were taken and collectedusing questionnaires. Using the Delphi method,slope, dam and precipitation were added to theeffective factors in this study.

After the identification of criteria andindicators at the next step, it should be clear thatthese criteria and indicators independently affecteach other, or work together. Therefore, if theywork together, what are their communicationstructures? In the current study, because of thecorrelations between indicators, the DEMATELtechnique was used to identify thecommunication structures for the desired criteriaand indicators. There are many multi-criteriadecision-making methods with regard to theassumption of existing independence betweenthe elements of system, but for all the cases thisindependence does not exist (Buyokozkan &Cifci, 2012). The DEMATEL technique waspresented in 1993 as a method for structuralmodelling concerning a problem (Fontela &Gabus, 1976).

This method specifies knowledge ofspecialists for determining the communicationstructures between criteria and drawing thenetwork map. The DEMATEL technique is oneof the pairing comparison-based decision-making methods that use experts views todetermine the elements of a system and theirsystemic structures using the graphic theory of

Paktinat et al. / Desert 19-2 (2014) 111-119114

principal structures of existing elements in thesystem, while influencing and interacting witheach other to obtain the elements. Expertsjurisdictions in pairing comparisons are simple,and specialists do not need to know about theDEMATEL process, but the quality of theirviewpoints and extent of their insights intodifferent sides of the problem have stronginfluences on the results given by DEMATEL(Aghaebrahimi, 2008).

Thus, the DEMATEL questionnaire has beenprovided and given to the specialists, and theywere asked to complete the questionnaires. Afterthis, the DEMATEL technique wasimplemented using the obtained questionnairesaccording to the following steps.Step 1: Find the initial average matrix and theaverage matrix obtained based on pairing

matrices completed by specialists.Step 2: Calculate the primary matrix of normaland direct relationships.Step 3: Calculate the final relationship matrix(T).Step 4: Define a threshold value to achieve afinal effect map.

In the final step, threshold limits aredetermined by calculating the average of theelements matrix. Because the matrix providesinformation on interactions between two factors,decision makers should determine a thresholdlimit for filtering the insignificant effects. Finalresults for this procedure are shown in Table 1.Zero (0) in Table 1 means that there is nointeraction between the elements in the rows andcolumns, while number 1 indicates the effects ofelements in the rows on those in the columns.

Table 1. Final results of DEMATEL techniques

GroundwaterDepth

Distance fromResidential

Centers

Distancefromroads

PrecipitationRadiationSlopeECDistance

fromDam

Criteria

00000000Distance from Dam00000000EC00001000Slope00000000Radiation10000011Precipitation

01000000Distance fromroads

00100000Distance fromResidential Centers00000000Groundwater Depth

For solving the problems of site selection,because a set of purposes should be optimized atthe same time, a multi-criteria decision-makingprocess is used (Forghani et al., 2007). As hasbeen mentioned before, there are many multi-criteria decision-making methods that assumethe elements of the system as independent ones,but for all cases this independence does not exist(Wu, 2008). For solving this problem, theanalytic network process is a relatively newmethod that was presented in 1996 by Saaty(Saaty, 1996). Saaty has presented the ANPmethod for solving the problems of correlationsbetween choices or criteria (Najafi, 2010; Dorri& Hamzei, 2010).

The analytic network process is one of themulti-criteria decision-making techniques, andis considered as one of the compensatorymodels (Faraji-Sabokbar et al., 2008). In fact,the analytic network process is one of the multi-criteria decision-making techniques that canremove the limitations of the analytic hierarchyprocess (Kiani et al., 2010). The analyticnetwork process can be used as an effective toolfor cases in which the interactions between the

elements of a system are formed as a networkstructure (Saaty, 1996).

Developing an ANP model requiresidentification of the problems, definition of therelated criteria and sub-criteria, anddetermination of their relationships andinteractions. As mentioned before, unlike theanalytic hierarchy process, in which therelationships between elements of a system areunidirectional, in the analytic network process,one element of a model influences the otherelements, and may even be affected by the otherelements so that when using a method theinternal relationship between elements should bespecified in the analytic network model. In thecurrent study, the DEMATEL technique wasused to find internal relationships between theelements of a system for the analytic networkmodel (Fig. 2), which has been described in fullabove (Saeidi & Najafi, 2010).

The modelling process in the ANP techniqueincludes the following steps:1. Pairing comparisons matrix and estimation ofrelative weighting:

Paktinat et al. / Desert 19-2 (2014) 111-119 115

After developing the network model, pairingcomparisons between criteria and sub-criteriawere performed using the relative significancescale. Pairing comparisons and element-pairingmatrices in each level are performed accordingto relative significance rather than controlcriteria, which may be similar to the AHPmethod. Experts viewpoints were used toobtain the pairing matrices. To avoid specificproblems in the decision-making process, we tryto select participants with the same levels ofspecialists and sufficient knowledge on therelated issues.2. Formation of primary supermatrix

ANP elements have specific interactions.These elements can include decision maker unit,criteria, results and choices. Relative weight ofeach matrix has been calculated based onpairing comparisons similar to AHP method.Obtained weights entered to supper matrixshowing the interactions between elements ofsystem.3. Calculation of general weight vector

In order to implement analytic networktechnique in current study, Super Decisionssoftware, version 2008, was applied. Thus, atfirst step, internal structures of indicators andcriterion network models were simulated usingthe results reported by the DEMATEL technique(Table 1). Then, in order to collect the expertsviewpoints, a questionnaire was designed andgiven to the specialists, and, afterwards, theobtained results were collected. At the final step,using questionnaires, pairing comparisons weremade between criteria and indicators. Weightsobtained from the analytic network techniqueare shown in Table 2.4. Combination of layers:

After identifying the criteria and indicatorsusing previous studies and the Delphi method,relationships between these criteria andindicators were identified and then weighted bythe analytic network process. Afterwards, these

indicators needed to be combined. In order to dothis, fuzzy logic was used in the currentinvestigation.

Table 2. Criteria and indicators used in the model andweights obtained from ANP

WeightIndicatorsCriteria0.051Distance from Dam

Technical 0.1821EC0.099Slope0.0459Distance from Roads

Economic 0.2144Groundwater Depth

0.1458Distance fromPopulation Centers0.1368RadiationClimate 0.125Precipitation

Fuzzy theory was first presented by anIranian scientist named Asqar Lotfizadeh,professor of the University of America workingon non-reliable conditions (Kahraman et al.,2006). Considering fuzzy logic, the degree ofmembership of an element within a set isdefined by a value in the interval between zero(uncompleted membership) to one (completedmembership) (Pour Ahmad, 2007). Degrees ofmembership are usually defined by themembership function by which they can belinear, non-linear, continuous or non-continuous(Bonham & Carter, 1991).

Considering the fuzzy model, for each pixelin each map, the value of zero to one (0-1) isgiven to indicate suitable places for the pixelswith respect to the related criteria for theobjectives of this study. On the other hand,concerning fuzzy logic, each area according tothe value adapted with the criteria has amembership value indicating the amount of areasuitability. To achieve success, the use of fuzzymathematics for different applications isstrongly dependent on the proper membershipfunction (Sui, 1992). In the current investigationregarding different factors, three kinds of fuzzyfunction were used (Table 3).

In the current study, the first kinds offunctions were used to model the indicator ofdistance from a road, distance from residentialcentres, distance from a dam, precipitation andradiation. Generally, elements related todistances and continuous issues such asradiation can be modelled by these functions. Inorder to model the slope and salinity indicators,because of non-linear variations for these twoindicators, a sinusoidal function was used forthis case. Boundary values and membershipfunctions have been defined for the indicatormap related to six factors shown in Table 3.

RoadsGroundwater

depthResidential

Economic

RadiationPrecipitatio

n

Climate

Dam SlopeEC

Technical

Fig. 2. General structure of model

Paktinat et al. / Desert 19-2 (2014) 111-119116

Table 3. Membership functions and domains defined for the indicatorsDomainsMembership FunctionIndicators

min x maxU=Distance from Roads

Distance from Residential CentersPrecipitation

min x maxU=RadiationDistance from Damc x dU=Slope

a x 1500U=EC X=1500U=1

1500 < x dU=

Regarding the natural depth of groundwater andexisting data, it was not possible to surveygradual changes for the suitability of differentareas in a specific map, so for this indicator, themembership function was utilized. However, allthe points with the same depth have a samedegree of membership. Different classes for thementioned factors and the given fuzzy values forthem are shown in Table 4. After applying thefuzzy functions and creating fuzzy layers, all thelayers were multiplied by the weights obtainedfrom the analytic network process (Fig. 3); then,they were combined using minimum function(Fig. 4).

Table 4. Different classes of groundwater depth and degreeof fuzzy membership in each class

IndicatorClassMembership Value

Groundwater Depth

0-25125-500.850-750.675-1000.4

100-1250.2

Using minimum function instead of thesimple additive weighting method (SAW) has amain advantage. Minimum function is a non-compensation function, while the SAW methodis a compensation one. On the other hand, inminimum function, unlike the SAW method, theeffects of an indicator cannot be removed by theimpacts of the other indicators.

Fig. 3. Weighted fuzzy maps of indicators used in the study

Fig. 4. Final phase map obtained by minimum function

3. Results

Regarding the results obtained from this study,technical, economical and climatic criteria andeight indicators (Table 2) were selected as themost important factors in the survey forselecting the suitable lands for the installation ofwater desalination systems.

The analytic network process indicated thedepth of groundwater and level of salinity tohave the largest weights. After combining thelayers by the minimum function and preparing afinal fuzzy map for the study, the fuzzy map wasclassified by the interval of 0.015 and alllimitations were classified as four classes in amanner where classes one and four refer to themost and least suitable lands for the installationof water desalination systems, respectively (Fig.5). The results have shown that class one islocated in five provinces of Yazd, Isfahan,Kerman, Hormozgan and Semnan. Yazdprovince is located in the area with the highestpercentage of class one and is ranked first.Kerman, Isfahan, Hormozgan and Semnanprovinces have occupied the other positions,respectively.

The results showed that among 250 scopestudies, 20 were put in class one. Largest area islocated in Yazd-Ardakan plain in Yazd provincewith the average EC of 8538 ppm and theaverage piezometric level of 50 to 75maccording to the inspections, field studies andinvestigation information on these areas. Theresults obtained from this investigation are fullylogical and practical.

On the other hand, a good agreement can befound between the current investigation andclimatic conditions in Iran confirming theseresults. Generally, class one determines the mostsuitable areas. In central areas in Iran, with hot

and dry weather conditions, groundwater isoften salty and the precipitation rate is very low.It should be noted that the defined classes areconventional and the final fuzzy map can beclassified in different classes with differentintervals.

Fig. 5. Final classification map of study area

4. Discussion and Conclusion

Because of hot weather and a low rate ofprecipitation, and the increased use ofgroundwater, water levels in desert areas in Iranare to be decreased and the salinity of waterincreased. Desalination of brackish water is theonly solution when there are no water resources.One of the most important factors influencingthe development and final success related to awater sweetening system is the economic factor:desalination is a very energy-consumingprocess. The highest varying cost is spent forenergy regarding the desalination process, sothat more than one third and even sometimesmore than half of the costs of water desalinationmay belong to it (Chaudhry, 2003).

Solar energy is a natural form of energy, andis cheaper than fossil fuels. Iran is located on theworldwide solar belt and is a country that hasgood access to solar energy, with susceptibleareas for using this energy. According todifferent developed technologies related todesalination using solar energy, we can now useboth parts of solar energy including photon andheating for the desalination process. One of theunique advantages for using solar energy in Iranis the applicability of it in all of the areas, eventhose with a low solar heating rate. The resultsobtained from the current study have shown thatcentral areas including Yazd, Isfahan, Kerman,Hormozgan and Semnan provinces are more

Paktinat et al. / Desert 19-2 (2014) 111-119 117

Paktinat et al. / Desert 19-2 (2014) 111-119118

susceptible for the installation of waterdesalination systems. Solar desalinations requirelow costs for operation and maintenanceprocesses, but need high primary investmentsfor the installation. Otherwise, this technologycould be the best solution for areas wherepopulation centres and water resources arereducing. The results obtained from the currentinvestigation have shown that suitable selectionsfor the installation of water desalination can beperformed using multi-criteria decision-makingand geographic information systems (GIS).

References

AghaEbrahimiSamani, B., A. Makoi, M.H. SadrLahijani,2008. Challenges of Iranian companies in oil and gasprojects to DEMATEL. Sharif Journal (IndustrialEngineering & Management), 45; 121-129.

Alatiqi, I., H. Ettouney, H. El-Dessouky, H., 1999.Process control in water desalination industry: anoverview. Desalination, 126; 1532.

Bayod Rujula, A.A., N. Khalidou Dia, 2010.Application of a multi-criteria analysis for theselection of the most suitable energy source andwater desalination system in Mauritania. EnergyPolicy, 38; 99115.

Beheshtifar, S., M.S. Mesgari, M.J. Voldan-Zoj, M.Karimi, 2010. The use of fuzzy logic in GIS to locategas stations. Journal of Civil and SurveyingEngineering, 44; 583-595.

Bonham-Carter, G.F., 1991. Geographic InformationSystem for Geoscie ntists: Modeling with GIS.Pergamon, Ontario, 291-300.

Buyokozkan, G., G. Cifci, 2012. A novel hybrid MCDMapproach based on fuzzy DEMATEL, fuzzy ANPand fuzzy TOPSIS to evaluate green suppliers.Expert Systems with Application, 39; 3000-3011.

Chaibi, M.T., 2000. An overview of solar desalinationfor domestic and agriculture water needs in remotearid areas. Desalination, 127; 119-133.

Charabi, Y., A. Gastli, 2011. PV site suitability analysisusing GIS-based spatial fuzzy multi-criteriaevaluation. Renewable Energy, 36; 2554-61.

Chaudhry, S., 2003. Unit cost of desalination. CaliforniaDesalination Task Force, California EnergyCommission. Sacramento, California.

Dari, B., E. Hamzehee, 2010. Determine risk responsestrategies using ANP technique, (Case Study: NorthAzadegan oil field development project). Journal ofTechnology Management, 4; 92-75.

Dastorani, M.T., A.R. Massah Bavani, S.Poormohammadi, M.H. Rahimian, 2011.Assessment of potential climate change impacts ondrought indicators (Case study: Yazd station, CentralIran). Desert, 16; 159-167.

Dore, MHI., 2005. Forecasting the economic costs ofdesalination technology. Desalination, 172; 20714.

Eltawil, M.A., Z. Zhengming, L. Yuan, 2009. A reviewof renewable energy technologies integrated withdesalination systems. Renewable and SustainableEnergy Reviews, 13; 22452262.

Faraji-Sabokbar, H.A., M. Salmani, F. Ferydooni, H.Karimzadeh, H. Rahimi, 2010. Rural Waste Landfilllocation using network analysis process (ANP): A

case study rural areas of Ghoochan city. ModaresJournal, 14; 137-149.

Forghani, A., N. Yazdan-shenas, A. Akhoondi, 2007.Provide a framework for locating industrial centersnationally with a case study. Journal of KnowledgeManagement, 77; 81-104.

Garcia-Rodriguez, L., 2003. Renewable energyapplications in desalination: state of the art. SolarEnergy, 75; 38193.

Gastli, A., Y. Charabi, S. Zekri, 2010. GIS-basedassessment of combined CSP electric power andseawater desalination plant for Duqum-Oman.Renewable and Sustainable Energy Reviews, 14;821827.

Gude, V.G., N. Nirmalakhandan, S. Deng, 2011.Desalination using solar energy: Towardssustainability. Energy, 36; 78-85.

HosiynZadeh, S.R., 1999. Tabas territory based ondetailed studies of the role of geomorphic andexternal dynamics processes in resonance the desertphenomena and desertification. PhD thesis, Facultyof Literature and Humanities, Tehran University.

Jazirehee, M.H., 1992. Drought phenomenon. Journal ofForest and Range, 13; 23-34.

Kahraman, C., T. Ertay, G. Buyukozkan, 2006. A fuzzyoptimization model for QFD planning process usinganalytic network approach. European Journal ofOperational Research, 171; 390411.

Kiyani, A., A. Khanjari, Q. Fazelniya, 2010. The usageANP model for proposed ICT patterns to optimizethe relationship between urban and rural in city ofAlashtar. Agricultural Modares Journal, 14; 267-249.

Montazer, G.A., N. Jafari, 2008. Fuzzy-Delphi Methodto determine the financial policies, Humanities andCultural Studies. Journal of SustainableDevelopment, 8; 1-10.

Najafi, A., 2010. Applying the analytic network process(ANP) in the analysis of structural challenges andthe Executive environmental in the implementationof project management. International Journal ofIndustrial Engineering and Production Management,21; 63-76.

Poorahmad, A. Q. Habibi, S. Mohammad-zahrayi, S.Nazari-Adli, 2007. Fuzzy algorithms and GIS forlocating urban facilities. Environmental Studies, 42;42-31.

Rashidi, M., H. Qarib, 2011. The use of Solar Energy;Strategy for Sustainable Development RenewableEnergy. First International Conference on EmergingTrends in Energy Conservation, Tehran, Iran.

Saeedi, H.R., A. Najafi, 2010. Using analytic networkprocess (ANP) to determine the movement prioritiesof animals from forest, and organization ofimmigrants Forest, Case Study: Bababkvh Gilandistrict. Journal of Forestry, 4; 309-321.

Saghafi, M., 2009. Renewable energy. Tehran UniversityPress, Tehran.

Sagie, D., E. Feinerman, E. Aharoni, E., 2001. Potentialof solar desalination in Israel and in its closevicinity. Desalination, 139; 21-33.

Shakerian, N., Gh.R. Zehtabian, H. Azarnivand, H.Khosravi, 2011. Evaluation of desertificationintensity based on soil and water criteria inJarghooyeh region. Desert, 16; 23-32.

Sui, D.Z., 1992. A Fuzzy GIS Modeling Approach forUrban Land Evaluation. Journal of Computers,Environment and Urban Systems, 16; 101-115.

Paktinat et al. / Desert 19-2 (2014) 111-119 119

The FAOSTAT database, Food and AgricultureOrganization of the United Nations. 2000.Population: annual time series. Rome: Food andAgriculture Organization of the United Nations.

Wu, W.W., 2008. Choosing knowledge managementstrategies by using a combined ANP and DEMATELapproach. Expert Systems with Applications, 35;828835.