Technical Journal of Engineering and Applied Sciences Available online at www.tjeas.com ©2014 TJEAS Journal-2014-4-3/124-134 ISSN 2051-0853 ©2014 TJEAS Locating the bank branches using a hybrid method Lotfalipour Zainab1 , Naji- Azimi Zahra2, Kazemi Mostafa3 1. MBA student, Management department of Ferdowsi University, Mashhad, Iran 2. Faculty member in department of Management , Ferdowsi University, Mashhad, Iran 3. Faculty member in department of Management ,Ferdowsi University, Mashhad, Iran Correspondence author email: [email protected] ABSTRACT: Survival in today's competitive world remains no space for mistakes. In this regard, planning and optimal decision-making are of the affecting factors in achieving success, and appropriate locating by organizations. The banks are of the organizations that the right selection of their branches plays a major role in their success. The purpose of this study is to aid the bank managers in order to select the appropriate location for establishing new branches and decreasing the risk of decision-making. The Pasargad bank was the case of this study and the five regions are nominated for establishing the new bank branches using the views of banking experts. To prioritize and select the appropriate location, the affective factors in locating the branches were selected with the help of experts and the literature review then, their relative weights was determined using the AHP method. But, since the paired comparisons of AHP is dependent on personal opinions, we came to get assistance from the hybrid method of AHP and Monte Carlo simulation in order to decrease the risk of decision-making, prioritize the locations, and select the best one. Finally, we compared the results statistically using both the descriptive and inferential statistics. Keywords: Location problem, Multi Criteria Decision Making, Analytical Hierarchy Process, Monte Carlo Simulation INTRODUCTION Placement and searching for new places has historically been considered by researchers and managers of different organizations (e.g., Charnetski 1976; Hotelling 1929; Isard 1956). They clearly found that selecting the right location would prevent many financial problems of the organizations, provide many benefits for them, and result in their success (Guneri et al. 2009; Kuo et al. 1999). Of the organizations that are closely associated with the concept of locating, are the banks. Although, the volume of practices and customers' traffic to their branches has been reduced because of spreading the electronic banking right now, but still the quality of branches' performance plays a major role in the efficiency of these financial-service enterprises. Therefore, it should be considered that the locating of branches is performed in such a way that leads to customers' satisfaction due to enhancing the quality of branches' performance. Few studies could be found related to locating in the field of banking from which the Min's (1989) article could be mentioned. From interviewing with managers of branches and investigating the statistics of their performance, it appears that the decision-makers of these enterprises still have difficulty with selecting a location for establishing a new branch, because it is a very complex decision-making process. Decision-makers carry out extensive researches and given the obtained information and their working experiences make a decision about selecting a location for establishing a new branch. Since many factors are involved in decision-making, there exists a possibility for personal judgments and making mistake. Furthermore, decision-making is not a responsibility of only one person and it is difficult to consider the opinions of all the decision-makers. These issues along with the other limitations such as finding the desired estate and possibility of buying or necessity of renting and many more, make it difficult to make decision. Since most of today's decision-makings have taken a mathematical form and many qualitative and quantitative factors are involved in this process, it seems to be helpful for managers to utilize the multi-criteria decisionmaking methods. Amongst the multi-criteria decision-making methods, the analytical hierarchy process not only allows you to investigate a wide range of primary and secondary indicators but also to study the association of these criteria with the desired alternatives. This method has been used in many large organizations like IBM, British Airway, Ford Motor, and Xerox and has been provided satisfactory results (Saaty 2008). In addition, Tech J Engin & App Sci., 4 (3): 124-134, 2014 since the decisions about selecting a location for establishing new branches are made by a decision-making team not an individual, the group analytical hierarchy process sounds to be appropriate, because by using this method the opinions of all the team members are considered in decision-making process in such a way that they cannot impose their opinion on the others. Indeed, the group analytical hierarchy process leads to excellent decisions so that combines all the decisions together and optimal decision includes the opinions of all members. However, since the criterion to decide in AHP is the comparisons of individuals, the hybrid method of AHP and Monte Carlo simulation is used in this study in order to decrease the effect of personal opinions. According to our information, this hybrid method is not used yet in locating the bank branches. Rest of the paper is consisted of literature review in section two, hybrid method used in section three, and computational results and conclusion in sections four and five. RESEARCH BACKGROUND LOCATING Locating has an extensive theoretical history. Many attempts have been made to solve the locating problems in which a wide range of objective factors and methodology is applied (Badri 1999). The year 1909 is known as birthdate of locating theory, because in which this theory was exposed by Weber for the first time (Brandeau & Chiu1989). He used the locating problem to determine the location of an industrial center in order to minimize the transportation costs (Tabari et al. 2008). After him, researchers made some changes in his proposed model and exploited new methods in order to locating in different sectors of industry. For instance, two years later, Geoffrion used the decomposition method along with mixed integer linear programming, simulation, and heuristic method in the analysis of locating problems (Badri 1999). And after him, Aikens (1985) took advantage of another method named mathematical programming in order to develop the models of facilities locating (Chou et al. 2008). Chen (1996) is of the other researchers who have focused on locating problem. He used the mathematical programming to offer a model for selecting the optimal location of distribution centers, and then, offered the fuzzy group decision model trying to find a suitable location for plant or retailing store (Chen 1999). In addition, Min (1989) combined the fuzzy goal programming and decision support system in order to locating the bank branches. Using the metaheuristic methods is also another approach to solve the locating problems. For example, in an empirical study, a group of researchers investigated the difference among three methods of tabu search, simulated anealing, and genetic algorithm in locating the facilities (Arostegui JR et al. 2006). Chen (2007) has also used a hybrid heuristic method in locating of hub. Furthermore, Caballero et al. (2007) have used a metahuristic method based on tabu for locating. After them, Chen and Ting (2008) combined a Lagrangian heuristic method and Ant Colony System in order to locating the single source capacitated location problem. LOCATING AND ANALYTICSAL HIERARCHY PROCESS In decision-making in the real world, the decision-maker is faced to many important factors and criteria which judgments about the importance of each of them affect his final decision. In such cases, the multi-criteria decision-making methods are used by decision-makers. Amongst them, the analytical hierarchy process is one of the most useful and widely used methods (Triantaphyllou & Mann 1995), because allows the decision-maker to formulate the decision-making process based on the hierarchical decision tree, consider the various qualitative and quantitative criteria, and analyze the competing alternatives of decision-making based on paired comparisons (Saaty 1990). The AHP method is widely used in various fields. In Turkey, Buyukozkan (2004) applied two methods of fuzzy analytical hierarchy and fuzzy Delphi in order to determine the best selection for e-market place. Furthermore, another researcher in the same country conducted the analytical hierarchy process for locating the industrial plants (Kazancoglu & Ada 2009). In 2011, two other Turkish researchers used the analytical hierarchy process for locating the gas stations (Semih & Seyhan 2011). In Russia, Lorentz (2008) utilized the AHP method for identifying and weighting the affective factors in locating the food factories. Selecting the appropriate location for outsourcing abroad using the analytical hierarchy process is also an issue performed by several American researchers (Boardman et al., 2008). In another research, locating an emergency logistic center was done with the aid of AHP method in China (Liu & Xiaohua a 2011). Another study was also performed in the dry port location selection in the same country using the two methods of Fuzzy-AHP and elimination ET choice translating reality (Ka 2011). Wei and his colleagues (2011) located the fire stations through the AHP method along with GIS. Bagum et al. (2012) also investigated the external and internal factors of locating the wind pumping system in Bangladesh using the Brown-Gibson plant location model and analytical hierarchy process respectively. In the field of services, Tzeng et al. (2002) used the analytical hierarchy process in order to specify the appropriate location for establishing the restaurant. Chou and his colleagues (2008) created a fuzzy multicriteria decision-making using the analytical hierarchy process, Fuzzy, and Ideal and anti-ideal solution, and used it in locating the international tourism hotels. 125 Tech J Engin & App Sci., 4 (3): 124-134, 2014 HYBRID METHID OF AHP AND MONTE CARLO SIMULATION IN LOCATING THE BANK BRANCHES In this paper, prioritizing the regions in order to determine a location for the new branch of Pasargad bank was performed using the hybrid method of AHP and Monte Carlo simulation, because the Monte Carlo method decreases the uncertainty of paired comparisons in the AHP (Hsu & Pan 2009). This method leads to more confidence in the results of decisions and enables the decision-makers to express their priorities with more flexibility than the AHP method. This method applies a triangular distribution in order to more efficiency when the distribution is unreliable and considers three states of minimum, maximum, and the most probable (Momani & Ahmed 2011). In this study, the following stages will be done in order to locate the bank branches using the hybrid method of analytical hierarchy process and Monte Carlo simulation: Modeling Simulating and carrying out the paired comparisons using the AHP method Prioritizing the branches Writing down the results No Is it time to check out the simulation? Yes Conclusion Figure 1. Stages of the hybrid method of AHP and Monte Carlo simulation MODELING The analytical hierarchy process needs to break a problem with several indicators down to a hierarchy of levels (Abu Dabous & Alkass 2008). In this study, we extracted the major locating criteria and sub-criteria after a survey of the heads, associates, and experts of the Pasargad bank and reviewing the literature, and selected five candidate locations for establishing the new branch via the opinions of banking specialists. Figure 2 indicates the hierarchical structure of locating for branches of the Pasargad bank in which the highest level represents for the main objective and the lowest level represents for the decision options. The middle levels belong to major criteria and sub-criteria. As it can be seen in this figure, the first criterion is the place position indicating the effect of proximity with the wealthy residential areas (Min 1989; Tzeng et al. 2002; Okeahalam 2009), commercial areas and markets (Meidan 1993; Boufounou, 1995; Tzeng et al. 2002), plants and huge economic organizations (Boufounou 1995), and industrial towns. The branches distance towards one another is also considered in this criterion (Min 1989). The next one is the economic criterion, which indicates that to what extent each branch candidate increases the bank reserves (Tzeng et al. 2002) and the potential and capacity of branch itself (Cheng et al. 2005). The third one is related to safety (Chou et al. 2008). The accessibility is the fourth criterion (Cheng et al. 2005) related to factors such as locating at the site of traffic plans, existing the 126 Tech J Engin & App Sci., 4 (3): 124-134, 2014 public transportation, and access to parking (Tzeng et al. 2002; Chou et al. 2008). The last one is the competition criterion (Cheng et al. 2005). The existence of branches from the other banks (Min 1989; Boufounou 1995; Tzeng et al. 2002; Chou et al. 2008) as well as the awareness of their reserves is considered in this criterion. Locating the bank branches Competitive Existence of the competitors Financial recourses of the other banks Accessibility Locating in traffic zones Access to parking Ghasemabad Forudgah Economic Security Environmental security Emamreza Financial recourses of the bank Branch's potential and capacity Hefdahshahrivar Positional Closeness to business centers Proximity to wealthy residential areas Proximity to large economic organizations Closeness to industrial towns Geographical distribution of the bank branches Tolab Figure 2. Hierarchical structure of bank locating PAIRED COMPARISONS In this stage, the paired comparison questionnaires were prepared and given to some experts of Pasargad bank, and then, the paird comparisons were made based on the Saaty scaling table (Saaty 2008). Nevertheless, one of the shortcomings of AHP method is to force the decision-makers to choose only one number in paired comparisons (Banuelas & Antony 2004). To cover this weakness, we located the bank branches with the help of complementary method of Monte Carlo simulation, which has not been used for locating in the literature so far. Monte Carlo simulation is a technique through which the uncertainties resulting from the variance of values in input variables that affect the model output are simulated via the random numbers. In this regard, we consider each component of paired comparisons as a random variable with triangular distribution function, which its amounts are shown through an ordinal ternary by the first, second, and third numbers are respectively representing for the least, the most probable, and the most possible values. In other words, the managers and experts allocate a ternary rather than a single number to each paired comparison, which its components are extracted from the following table. 127 Tech J Engin & App Sci., 4 (3): 124-134, 2014 Table 1. The fundamental scale of absolute numbers (saati 2008) Intensity of importance Definition Explanation 1 Equal Importance Two activities contribute equally to the objective 2 Weak or slight 3 Moderate importance Experience and judgement slightly favour one activity over another 4 Moderate plus 5 Strong importance Experience and judgement strongly favour one activity over another 6 Strong plus 7 Very strong or An activity is favoured very demonstrated importance strongly over another; its dominance demonstrated in practice 8 Very, very strong 9 Extreme importance The evidence favouring one activity over another is of the highest possible order of affirmation If activity i has one of the Reciprocals above non-zero numbers A reasonable assumption of above assigned to it when compared with activity j, then j has the reciprocal value when compared with i May be difficult to assign the best 1.1–1.9 value but when compared with If the activities are very other contrasting activities the close size of the small numbers would not be too noticeable, yet they can still indicate the relative importance of the activities. In the present study, which does not exist one decision-maker but there exists a group of decision-makers; the opinions of these individuals should be merged somehow. The first solution is that the members of decisionmaking group come together and reach to a single opinion. But, this procedure was not used here because aggregating the experts and managers together in a meeting was impossible. On the other hand, since prioritizing the branches is being performed in this paper using the AHP method as well as the hybrid method of AHP and Monte Carlo simulation separately, it should be determined how to combine the opinions in either AHP or Hybrid method. According to what was stated, the geometric mean is used in the first method in which prioritizing the alternatives is carried out only using the AHP method. How to combine the opinions in the hybrid method will be mentioned in the next section. SIMULATION AND PRIORITIZING As previously mentioned, in this problem we have a random variable with a triangular distribution function per each component of paired comparisons, which the minimum, maximum, and most probable values are identified by experts. In other words, there exists an interval per each paired comparison in the simulation and every time a random number with a triangular distribution function is elected from this interval as a number relating to comparison. Given the high capabilities of Excel software, this stage and the next ones is done using the simtools tab in this software. Since there exists a group of decision-makers rather than a single decision-maker in this study and the random selection is very important as well, the opinion tables of all the decision-makers are integrated at first. In this case, the minimum, mean, and maximum values of experts' opinions for each component are listed in the relevant table. Then, one number in each interval is selected randomly using the Monte Carlo simulation and considering the triangular probability distribution, and after creating all the required matrices, AHP will continue 128 Tech J Engin & App Sci., 4 (3): 124-134, 2014 as usual and the regions are prioritized based on obtained scores. Then, the whole process is repeated 1000 times, and each time the results are recorded. It should be noted that after generating random numbers, the consistency of tables are also investigated and if the consistency of matrix is greater than .10, another random number is generated so that, we have a total of 1000 iterations of the consistent simulation matrix. COMPUTATIONAL RESULTS OF CASE STUDY Analytical hierarchy process In the first method of the present study, branches have been prioritized solely based on the analytical hierarchy process. As it can be seen in table 2, the branch of Emamreza Street has gained the most score using the AHP method. Furthermore, the prioritization results of the other regions based on this method are provided in the table. Table2. Ranking of candidate places using AHP Place of branch Score Emamreza 0.229 Forudgah 0.199 Tolab 0.198 Ghasemabad 0.190 Hefdahshahrivar 0.185 As it can be seen, the branches are rated closely and it is difficult to give a preference to one over another. For example, if we investigate the table scores with the precision of a one-digit decimal, there is no difference between the branches of Forudgah, Tolab, and Ghasemabad. For this reason, another method is needed to compensate this deficiency and create further certainty in decision-making. Hybrid method of analytical hierarchy process and Monte Carlo simulation In this section, the results of 1000 iteration of the hybrid process of AHP and Monte Carlo simulation was investigated via the STATISTICA software. Figure 3 plots the frequency charts for each region and table 3 indicates the statistical information relating to the five branches of Tolab, Hefdahshahrivar, Emamreza, and Ghasemabad. As it can be seen in table 3, Emamreza branch is in the first rank than the other locations in terms of score means. In these charts, the horizontal axis represents for obtained scores and vertical axis for score frequency of each branch in 1000 iteration of the process. 129 Tech J Engin & App Sci., 4 (3): 124-134, 2014 Figure 3. Frequency charts of branches in the hybrid method of AHP & Monte Carlo De scri ptives Table 3. Statistical information of the five branches based on the hybrid method of AHP & Monte Carlo Sc ore2 Tolab Hefdahshahrivar Emamreza Forudgah Ghasemabad Total N 1000 1000 1000 1000 1000 5000 Mean .28302 .21300 .23113 .14177 .13107 .20000 St d. Deviat ion St d. Error .095701 .003026 .026735 .000845 .030017 .000949 .034677 .001097 .059911 .001895 .079660 .001127 95% Confidenc e Int erval for Mean Lower Bound Upper Bound .27708 .28896 .21135 .21466 .22927 .23300 .13962 .14393 .12735 .13479 .19779 .20221 Minimum .149 .151 .178 .097 .061 .061 Maximum .415 .274 .328 .215 .251 .415 Furthermore, according to information obtained from the 1000 iteration of the simulation, Emamreza branch 495 times, Tolab branch 492 times, Qasemabad branch 11 times, and Hefdah Shahrivar branch 2 times has earned the first place. According to this method, the Forudgah branch could not win the first place in 1000 iteration. The following figure shows the ranking frequency of branches. The horizontal axis represents for braches' ranks and the vertical axis for frequency of ranks in 1000 iteration of the process. 130 Tech J Engin & App Sci., 4 (3): 124-134, 2014 Figure 4. Frequency charts of the branches ranks According to frequency charts, these tables are not lonely enough to decide on the best location. Thus, we conducted an analysis of variance in order to find out whether the differences among the branches' score ANOVA means are statistically significant. Results for this test are provided in table 4. Table 4. Analysis of variance test Sc ore2 Sum of Squares Between Groups 16.172 W ithin Groups 15.551 Total 31.722 df 4 4995 4999 Mean Square 4.043 .003 F 1298.611 Sig. .000 According to results of the one-way analysis of variance, the assumption of equity for score means of branches is strongly rejected, because the resulting P-value is much smaller than the significance level of the test i.e. .05 (p-value = .000 < α = .05). Therefore, it can be concluded with the confidence coefficient of 95% that there are significant differences among the relative score means of the five branches. 131 Tech J Engin & App Sci., 4 (3): 124-134, 2014 After awareness of the significant difference among the score means of branches, it was determined that the difference there exists between which one of the branches using the Duncan paired comparison test. Carmer and Swanson (1973) indicated that the Duncan multiple range test is highly effective in revealing the real difference between the means when we use the Monte Carlo simulation. Table 5 shows the results of this test. Table 5. Duncan test result Score 2 a Duncan Group Ghasemabad Forudgah Hefdahshahrivar Emamreza Tolab Sig. N 1000 1000 1000 1000 1000 1 .13107 Subset for alpha = .05 2 3 4 5 .14177 .21300 .23113 1.000 1.000 1.000 1.000 .28302 1.000 Means for groups in homogeneous subsets are displayed. a. Us es Harmonic Mean Sample Size = 1000.000. As it can be seen, the results of Duncan test reveal that all the relative score means are significantly different and each of them is classified in one distinct category. According to this test, Tolab branch has the highest level with an average of .283 and Ghasemabad branch has the lowest level with an average of .131. The box plot, which is shown in figure 5 obviously indicates the difference among the relative score means of branches. 132 Tech J Engin & App Sci., 4 (3): 124-134, 2014 Categ. Box & Whisker Plot 0.30 م يزان اه م ي ت ي ا وزن ن س ب ي در روش دو م 0.28 0.26 0.24 0.22 0.20 0.18 0.16 0.14 0.12 0.10 Tolab Ghasemabad Emamreza Forudgah Hefdahshahrivar Mean Mean±SE Mean±1.96*SE Figure 5. Box plot for comparing the importance of the five branches DISCUSSION AND CONCLUSION This study tried to identify the affective criteria in locating the bank branches. This was accomplished through the literature review as well as interviewing with the heads, associates, and experts of Pasargad bank in Mashhad. Obtained results indicated five main criteria including economic, place position, safety, accessibility, and competition criterion. Each of these criteria, which include several sub-criteria are also discussed in detail and can be seen in figure 1. Another objective was to prioritize the location of new branches based on the obtained indicators. Obtained criteria as well as the five candidate regions of Tolab, Hefdahshahrivar, Emamreza, Forudgah, and Qasemabad were prioritized using the analytical hierarchy process and the best one was selected. Then, the Monte Carlo simulation method was used as a complementary method in order to decrease errors, due to the large amount of indicators and uncertainty in decision-making. By the group analytical hierarchy process, the branches' scores were close to each other, which hardens the decision-making process, although the Tolab branch was selected as the best, according to the both two methods. But, it is not so in the hybrid method as the conducted statistical tests revealed a significant difference among the branches' scores. This study focused on the banking industry in a limited scale. Time restrictions, lack of access to decisionmakers (headquarters unit), and lack of access to needed information impeded the comparing and scoring processes. Besides this information, using the more precise information obtained from the GIS certainly will result in the more accurate results. Furthermore, the above method can be utilized in the other sections of banking and in the other industries. It is also proposed to use the other locating methods such as goal programming, and the other multi-criteria methods, and to compare these methods in order to reach the most practical method in locating of bank branches. REFERENCES Abu Dabous, S., Alkass, S. 2008. Decision support method for multi-criteria selection of bridge rehabilitation strategy. Construction management and economics 26(9): 883- 893 Arostegui Jr, M.A., Kadipasaoglu, S., Khumawala, B.M. 2006. An empirical comparison of Tabu Search, Simulated annealing, and Genetic Algorithms for facilities location problems. International journal of production economics 103: 742- 754. Badri, M. 1999. Combining the analytic hierarchy process and goal programming for global facility location- allocation problem. International journal of production economics 62: 237- 248 Bagum, N., Rashed, A.A, Masud A.K.M., Islam, Q. 2012. Using multi-criteria analysis in decision making regarding the adoption of wind pump for irrigation in Bangladesh. Review of general management 15: 157-178. Banuelas, R., Antony, J. 2004. Modified analytic hierarchy process to incorporate uncertainty and managerial aspects. International journal of production research 42(18): 3851- 3872. 133 Tech J Engin & App Sci., 4 (3): 124-134, 2014 Boardman Liu, L., Berger, P., Zeng, A., & Gerstenfeld, A. 2008. Applying the analytic hierarchy process to the offshore outsourcing location decision. Supply chain management: An international journal 13:435- 449. Boufounou, P.V. 1995. Theory and methodology, evaluating bank branch location and performance: a case study. European journal of operational research 87: 389- 402. . Brandeau, M.L., Chiu, S.S. 1989. An overview of representative problems in location research. Management Science 35: 645–674 Buyukozkan, G. 2004. Multi- criteria decision making for e-marketplace selection. Internet research 14: 139- 154. Caballero, R., Gonzalez, M., Guerrero, F.M., Molina, J., Paralera, C. 2007. Solving a multiobjective location routing problem with a metaheuristic based on tabu search. Application to a real case in Andalusia. European journal of operational research 177: 1751- 1763. Carmer, S.G., Swanson, M.R.S. 1973. Evaluation of Ten Pairwise Multiple Comparison Procedures by Monte Carlo Methods. Journal of the American Statistical Association 68: 66-74. Charnetski, J.R. 1976. Multiple criteria decision making with partial information: a site selection problem, Space Location-Regional Development, Pion, London. Chen, C.H., Ting, C.J. 2008. Combining Lagrangian heuristic and Ant Colony System to solve the Single Source Capacitated Facility Location Problem. Transportation research part E 44: 1099- 1122. Chen, C.T. 1999. A fuzzy group decision model of location selection for distribution center. Journal of Management and Systems 6 (4): 459– 480. Chen, J.F. 2007. A hybrid heuristic for the uncapacitated single allocation hub location problem. Omega 35: 211- 220. Cheng, E.W.L., Li, H., & Yu, L. 2005. The analytic network process(ANP) approach location selection: a shopping mall illustration. Construction innovation: information, process, management 5: 83- 97. Chou, T.Y., Hsu, C.L., & Chen, M.C. 2008. A fuzzy multi- criteria decision model for international tourist hotels location selection. International journal of hospitality management 27: 293- 301. Guneri, A. F. ,Cengiz, M., Seker, S. 2009. A fuzzy ANP approach to shipyard location selection. Expert System with Application 36: 213-24. Hong, L., Xiaohua a, Z. 2011. Study on location selection of multi objective emergency logistics center based on AHP. Procedia engineering 15: 2128- 2132. Hotelling, H. 1929. Stability in competition. Economic Journal 39: 41–57. dental quality attributes. Expert Systems with Applications 36; Hsu, T.H., Pan, F.F.C. 2009. Application of Monte Carlo AHP in ranking .2310-2316 Isard, W. 1956. Location and Space Economy, Technology Press, MIT, Cambridge. Ka, B. 2011. Application of fuzzy AHP and Electre to China dy port location selection. The asian journal of shipping and logistics 27: 331- 354. Kazancoglu, Y., Ada, E. 2009. The location problem of a plant and a warehouse by an expanded linear programming model integrated with AHP. Ege academic review 9: 29- 42. Kuo, R.J., Chi, S.C., & Kao, S.S. 1999. A Decision Support System for Locating Convenience Store through Fuzzy AHP. Computers & Industrial Engineering 37: 323-326. Lorentz, H. 2008. Production locations for the internationalising food industry: case stydy from Russia. British Food Journal 110: 310- 334. Meidan, A. 1993. Distribution of bank services and branch location. International journal of bank marketing 2: 60- 72. Min, H. 1989. A model- based decision suport system for locating banks. information and management 17: 215- 207. Momani, A. M. , Ahmed, A.A. 2011. Material annealing equipment selection using Hybrid Mont Carlo simulation and Analytic Hierarchy Process. World academy of science engineering and technology 59: 953- 957. Okeahalam, C. 2009. Bank branch location: a count analysis. Spatial economic analysis 4: 275- 300. Saati, T.L. 1990. How to make a decision: the analytic hierarchy process. European journal of operational research 48: 9-26. Saati, T.L. 2008. Decision making with the analytic hierarchy process. Int. J. Services 1: 83- 98. Saati, T.L. 2008. Relative measurement and generalization in decision making, why pair wise comparisons are central in mathematics for the measurement of intangible factors, the analytic hierarchy/network process. Statistics and operations research 102( 2): 251-318. Semih, T., Seyhan, S. 2011. A multi- criteria factor evaluation model for gas station site selection. Journal of global management 2: 12- 21. Tabari, M., Kaboli, A., Aryanezhad, M., Shahanaghi, K., & Siadat, A. 2008. A new method for location selection: A hybrid analysis. Applied mathematics and computation 206: 598- 606. Triantaphyllou, E., Mann, S.H. 1995. Using the analytic hierarchy process for decision making in engineering applications: some challenges. Inter’ l journal of industrial engineering: applications and practice 2(1): 35-44. Tzeng, G.H., Teng, M.H., Chen, J.J., Opricovic, S. 2002. Multicriteria selection for a restaurant location inTaipei. Hospitality management 21: 171- 187. Wei, L. , Li, H.L. , Liu, Q., Chen, J.Y., Cui, Y.J. 2011. Study and implementation of fire sites planning based on GIS and AHP. Procedia Engineering 11: 486- 495. 134
© Copyright 2026 Paperzz