An exploratory study of taxi sharing schemas

Taxi sharing services have been increasing in the past few years in several countries. In this paper we propose a solution to taxi sharing problem based on Vehicle Routing Problem (VRP) with Temporal Restriction. We approach the problem in a static way, for testing purposes. Two metaheuristics were used to determine the best travel plans for taxi sharing service, taking into account the pickup and delivery time of each passenger. Tabu Search (TS) and Simulated Annealing (SA) were used to solve the problem, as literature review suggested, presenting good results. Four test scenario were used with 10, 20, 30 and 40 passengers, knowing their starting location. Analysis of solutions found by TS and SA are compared to the average cost to conclude which is the best travel plan to be used by taxi sharing service.