Minimizing Transfer Times in Public Transit Network with Genetic Algorithm

Abstract

This paper presents a systemwide approach based on a genetic algorithm for the optimization of bus transit system transfer times. The algorithm attempts to find the best feasible solution for the transfer time optimization problem by shifting existing timetables. It makes use of existing scheduled timetables and ridership data at all transfer locations and takes into consideration the randomness of bus arrivals. The complexity of the problem is mainly due to the use of a large set of binary and discrete variables. The combinatorial nature of the problem results in a significant computational burden, and thus it is difficult to solve with classical methods. Scheduling data from Broward County Transit, Florida, were used to calculate total transfer times for the existing and proposed systems. Results showed that the algorithm produced significant transfer time savings.