Spatial Analysis of the Location of Banks, Financial and Credit Institutions in 6th District of Tehran by DEMATEL techniques and Analytic Network Process (ANP)

Extended Abstract
Introduction
Growth of banks, financial and credit institutes in the recent years, their competition for more
advantages and also the subject of customer satisfaction made it essential to use scientific
methods for an optimized performance. An optimal decision, especially in site selection of
financial institution or services, plays an important role in successful achievement of the goals
and is also effective in customer attracting. A wrong site selection of the financial institution or
services will reduce the efficiency, increase the costs and may also cause the irreversible losses
in the competition with other institutions. Therefore, it is necessary to make an optimal decision
in site selection of the banks and financial and credit institutes to achieve the highest rates of
return rather than the costs of setting up in order to provide customer service, and establishing
the highest possible use of site capacity. For example, evidence shows that customer satisfaction
has a direct relationship with access to the ATMs. Thus, urban economic zoning to select the
suitable areas for such activity, and to evaluate the current performance and subsidiaries of
future potential, is very important for closing the branches and move them to the appropriate
locations.
The most important aim of this study is to provide a model for economic zoning and site
selection of banks, financial and credit institutes and their services.
Methodology
This research has been executed in a development-application approach and employed
descriptive-analytical methods. According to the research objectives, the economic criteria in
the banking system have primarily been identified through a literature review (previous
research) and also completed through expert opinions. Then, in a field survey the required data
have been collected to prepare the criteria for future analysis. DEMATEL1 techniques have been
used for identification of the internal relationships among the criteria. After completing the
pairwise comparisons questionnaire, and obtaining the results by the experts based on Copeland
method and by Analytical Network Process (ANP) model the weights has been determined for
each pair of the criteria. To determine the potential radius of influence for each criterion in a
given area, the model presented by Keyter has been used. Fuzzy membership function for each
of the indices have been calculated and then combined by Fuzzy Algebraic Sum operator. In this
research, MATLAB software has been applied to implement DEMATEL techniques. Super
Decision software has also been applied for calculation of analysis network process model, and
ArcGIS software for spatial modeling and zonation. Afterwards, the status of each branch of the
banks and the financial and credit institutes of the region was obtained by plugging it into the
model output in tables. Finally, the SPSS software was used to assess homogeneity of the results
obtained from models and the results of field observations.
Results and Discussion
The first step in the site selection of the banks and the financial and credit institutes is to
determine the effective criteria, identify and prepare them by expert opinion taken from
literature and previous evidence (Table 1). For identification of the relations between the
criteria, DEMATEL technique has been performed. The weights of criteria and sub criteria have
been determined using the Analytic Network Process.
Table1. The final weights for each of the following criteria and sub criteria using ANP
Transport and
Demand / consumption Municipal services and facilities Traffic
Educational Traffic Transport
Traffic Transport and cultural Educational
Transport and cultural Economic and
commercial
0.257521 0.143788 0.040825 0.200450 0.143788 0.040825 0.200450 0.143788
For Spatial Modeling and zonation of the area, we have determined fuzzy membership
function by the potential radius of influence of each criterion in the supposed area. They have 
been calculated by in ArcGIS by the model presented by Keyter and then combined with each
other by Fuzzy Algebraic Sum operator.
In order to assess economic features of the area and to evaluate the results of the study, the
area on the basis of standard deviation method was classified into 6 classes. Class 1 regarding to
the more radius of influence is the best economic class for establishment of the branches and
afterward presentation of services and facilities. The other classes decrease from 1 to 6, orderly,
in the importance of preferences. The results show that only 0.443 percent of the area is in class
1, 6.306 percent in class 2 and proportion of classes 3, 4, 5 and 6 are 26.022, 34.142, 25.390 and
7.696 percent, respectively. About the branches of the banks and the financial and credit
institutes which in this study are just in classes 1-4 , the proportion of these branches in these 4
classes is, respectively, equal to 8.01, 50.904,40.052 and 1.034 percent (31,197,155 and 4
branches).
For evaluating the results of the model, in each class of banks and the financial and credit
institutes some branches were chosen randomly and have been observed using survey methods.
The results of a qualitative field observation have been ranked in the range of numbers from 1 to
4. The number 1 has been considered as the regional economy with high ability and 4 as the
regional economy with low ability. The coefficient Kendall's tau-c in SPSS software was used
to determine the homogeneity among the results obtained in the model and observations results.
The results reported the coefficient equal to 0.748 with Significance less than 5 percent that
indicates the strong relationship between two variables.
Conclusion
This Classification is useful for managers of the banks and the financial and credit institutes and
also the planners to identify the area of economic-potential for the construction of new branches
and establishment of ATMs and also to identify the current status against other competitors for
the current and future planning. The results also can be very useful for closing the non-optimal
branches and reestablish them in the appropriate areas. This can also be utilized for allocation of
ATM machines to each of the branches, according to their location and adjustment of ATMs in
non-optimal branches to transfer them to other optimized branches. The results will provide the
economic managers with a broad insight into the most important world economy positions.