Public Transportation Mode Selection in an Urban Corridor: Application of Multi-Criteria Decision Making Methods

Transportation decisions including transportation mode selection، vehicle routing، and scheduling are directly related to the location and land uses. Changes in location and land uses can lead to considerable changes in transportation decisions at the strategic، tactical، and operational levels. At the strategic level، selection of the appropriate transportation mode is the most important concept in urban transportation planning. Each of the transportation modes has its own advantages، limitations، and characteristics. Therefore any of them can be considered as the best choice under different circumstances. In the selection of a particular transportation mode، all of the corresponding advantages and limitations have to be considered. Although choice of transportation mode may seem simple، but a systematic comparison between modes by using a variety of criteria is required. For this purpose، different criteria including the cost، time and traveling distance، availability، reliability، capacity، and security may be considered. Each of these criteria may conflict with each other. Therefore، identification of the qualitative and quantitative evaluation criteria، definition of their interrelations and feedbacks، assessment of their importance، for choosing the best transportation mode is an important problem for transportation planners and decision makers. This problem can be solved using a well known multi-criteria decision-making (MCDM) process. Various multi-criteria decision-making (MCDM) methods such as the OWA، ELECTRE، PROMETHEE، TOPSIS، and AHP have been used in past works for comparison and determination of the appropriate transportation modes. In all of these methods، evaluation criteria are assumed to be independent with no dependencies within or between different criteria. But، in many decision making problems including the transportation mode selection، considerable inner and outer dependencies can be expected between the criteria and alternatives. Due to the mutual dependencies and feedback effects of the criteria، the analytic network process (ANP) can be used for systematic evaluation and selection of the most appropriate transportation mode. The ANP is a multi-attribute approach for decision-making which enables consideration of interactions and feedbacks within (inner dependences) and between (outer dependences) the criteria. In this paper، the best public transportation mode in an urban corridor، has been selected using the ANP and TOPSIS Multi-Criteria Decision Making (MCDM) techniques and Geospatial Information System (GIS). The Analytic Network Process (ANP) has been used for evaluation of the criteria and determination of the preferences. Then TOPSIS technique has been used for evaluation and ranking the alternatives. Theoretical bases: 2-1- Analytic Network Process (ANP) Many decision problems may not be structured hierarchically because they involve interactions and dependencies between the higher-level and lower-level elements. Saaty propose a network model for structuring these problems. Analytic Network Process (ANP) is a more general form of the Analytic Hierarchy Process (AHP)، first introduced by Saaty. The ANP is a nonlinear structure that deals with dependences within (inner dependencies) and among the different sets of elements (outer dependences). Generally، the ANP consists of three stages: first is the network construction، pair-wise comparisons within and between criteria، and relative weight estimation of criteria and indicators، second is formation of the initial supermatrix، and third is formation of the weighted and limiting supermatrix. To construct the network structure of the problem، all indicators are classified to multiple criteria and their interactions are considered. All of these relations are evaluated using the pairwise comparisons which uses the local priority vectors (eigenvectors)، as an estimate of the relative importance of the indicators. To obtain the global priorities، local priority vectors are entered in the appropriate columns of the matrix of influences among the elements، known as the supermatrix. The initial supermatrix must be normalized to a weighted supermatrix in which the sum of each of its columns equals one. The weighted supermatrix is raised to limiting powers to calculate the limiting supermatrix and overall priorities. 2-2- TOPSIS TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) is a multi-attribute decision making method to identify solutions from a finite set of alternatives. TOPSIS method chooses an alternative which respectively has the shortest and highest distance from the positive and negative ideal solutions. The positive and negative ideal solutions are respectively composed of the best and worst performance values. Steps of the TOPSIS are as follow:  Formation of the decision matrix،  Normalization of the decision matrix،  Calculation of the weighted normalized decision matrix،  Determination of the positive ideal and negative ideal solutions،  Calculation of the separation measures of each alternative from the ideal solutions using the m-dimensional Euclidean distance،  Calculation of the relative closeness of each alternative to the ideal solutions and the rank of alternatives. The main purpose of this paper was choice of the best public transportation mode in an urban corridor، using Multi-Criteria Decision Making (MCDM) techniques and Geographic Information System (GIS). For this purpose، a sample corridor in Tehran city was selected as the case study. This corridor is located between Azadi and Pounak Squares and on the boundary regions of 2 and 5 of Tehran. Length of the corridor is 6. 75 km and it contains a considerable daily trips. Due to high travel demand and traffic، development of the public transportation modes is of prime importance. To choose the most appropriate public transportation mode in the corridor، four alternatives including the bus، taxi، Bus Rapid Transit (BRT)، and metro together with the twenty indicators were considered. The indicators were classified to seven criteria including the cost، time، flexibility، reliability، security، pollution، and mode properties. Afterwards، relations and feedbacks between the criteria were determined and a network structure in ANP model was constructed. The decision-making and evaluation processes were implemented in GIS environment. Finally، weights of indicators were obtained from the ANP method. Then، TOPSIS technique was used to evaluate and rank the alternatives. Results of the evaluation showed that respectively BRT، metro، bus، and taxi transportation modes had highest scores in the selected urban corridor. The main contribution of this study was consideration of many detailed criteria and evaluation of the relations and feedbacks between them to select the most appropriate public transportation mode in an urban corridor. The evaluation criteria for transportation mode selection are not generally independent from each other، but they are often interdependent. Therefore، the AHP method that neglects the relations and feedbacks between the criteria is not suitable for solving this problem. To handle the inner and outer dependencies among the criteria، the ANP method was used in this paper. Complex decision making problems، such as the public transportation mode selection، which consist of many interdependent criteria، are effectively solved using the ANP. Whereas the AHP can be only used for the hierarchical decision structures. Suggestions: There are different studies that attempt to solve transport optimization problems، which include the mode choice phase. In the future researches، the weights of the criteria were obtained in this study can be used in different mathematical transportation models. Moreover، Instead of using precise numbers to compare the criteria and ANP method to obtain the weight of the criteria، fuzzy numbers and Fuzzy ANP (FANP) method can be used for transportation mode choice problem. The FANP encompasses and solves the ambiguity and imprecision of the pair-wise comparisons.