Performance Comparison between the Multi-Colony and Multi-Pheromone ACO Algorithms for Solving the Multi-objective Routing Problem in a Public Transportation Network

What is it about?

There would be one common experience between anybody who lives in metropolitans; you would be confused about choosing your prefereed path in the complex large multimodal public transport network. Our main goal was to solve this problem considering and focusing on real preferences and demands of users. We try to design a system in order to consider users' rights to choose their favorite path. Despite of technically grows of navigation systems in the last decades, there are lots of disadvantages about them; so, we solved and improved current navigation systems considering following features: • We solve a complex routing problem (path finding) in a real large public transportation network. • Three objective functions (Time, Expense, Comfort) are considered to solve user-friendly the problem. • We find several non-dominated paths for providing users' rights in choosing their path. • Two different concepts of MOACO, as the optimization algorithms,, are modeled to the space of mentioned problem. • Optimum initial parameters of the algorithms for the problem are determined.

Featured Image

Why is it important?

First, we used a new point of view with concentration on users' demands to solve the multi-objective path finding problem which is really important in current complex network of public transport. Our view point would increas satisfaction level of its users. Despite the classic methods in solving routing problem, we used two modes of multi-objective ACO algorithm (as metaheuristic optimization algorithm) for solving the problem. Our results show the capability of the algorithms in finding several various non-dominated paths for each specific origin and destination in public transport network of Tehran city.

Perspectives