Accessible taxi routing strategy based on travel behavior of people with disabilities incorporating vehicle routing problem and Gaussian mixture model

Abstract

The government of Seoul has implemented accessible taxis, following the Convention on the Rights of Persons with Disabilities established by the United Nations in 2006. Accessible taxis are an emerging transportation mode that improves transportation accessibility for people with disabilities (PWD). The main idea of this study is to classify PWD into short-distance and long-distance groups and operate corresponding short-distance and long-distance taxis to improve the operational efficiency and user convenience of accessible taxi services. This study aims to find an optimal routing strategy for accessible taxis considering the travel behavior of PWD. Specifically, the travel distance distribution of PWD was divided into multiple distributions, i.e., short-distance and long-distance distributions, using Gaussian Mixture Model. Then, a vehicle routing problem with time windows was developed to find the optimal routing strategy for accessible taxis. The objective function was set to minimize users’ total wait time, and the decision variables included the number of user classifications, the number of accessible taxis, and the route for each vehicle. The results of the proposed strategy showed about an 8.0% reduction in total wait time and a 1.3% reduction in the total operating distance of accessible taxis compared to the current strategy. Specifically, the wait time of the short-distance user group with a distance of less than 4.3 km was reduced by about 26%. An economic benefit analysis was also performed to identify the applicability of the proposed strategy, and the results showed that the proposed model was economically beneficial.