Decision-analytics-based electric vehicle charging station location selection: A cutting-edge fuzzy rough framework

Abstract

Electric vehicles are of great significance in supporting sustainable transportation and sustainability. In parallel with the increasing demand for such vehicles worldwide, the electric vehicle charging stations (EVCSs) market has grown dramatically. The study presents a practical model for selecting EVCS sites integrating multi-criteria decision-making (MCDM), fuzzy, and rough sets. The research aims to bridge the gap in evaluating EVCS locations by leveraging the superiorities of fuzzy and rough set theories to address vagueness effectively. Firstly, assessment criteria cover the environment, economic, technology, and social drivers. Secondly, a fuzzy Defining Interrelationships Between Ranked criteria (F-DIBR) model is applied to determine the weight values of siting factors. Last, for the first time, the Mixed Aggregation by COmprehensive Normalization Technique (MACONT) with hybrid fuzzy rough numbers (FRN-MACONT) model is proposed to obtain the ranking results. Further, a new approach for defining hybrid fuzzy rough numbers is suggested, based on an improved methodology for determining rough numbers' lower and upper limits, allowing consideration of mutual relations between a set of objects and flexible representation of rough boundary intervals depending on the dynamic environmental conditions. The study's novelties reside in deciding the importance of the driving forces used in determining the EVCS site location with a novel method, F-DIBR, and selecting the optimal site with a new FRN-MACONT approach. The results show that "economy" is the most significant criterion, whereas "system reliability" is the most critical sub-criterion. The findings also indicate that the Konak territory performs the best, whereas the Cigli territory is the second best. Comprehensive sensitivity analysis verifies the proposed framework's validity, robustness, and effectiveness. As per the research findings and analyses, some managerial implications are further discussed. The approach introduced has the potential to contribute to the green transport literature.