Application of particle swarm optimization to transportation network design problem

Transportation network design problem (tndp) aims to choose from among a set of alternatives (e.g., set of new arcs) which minimizes an objective (e.g., total travel time), while keeping consumption of resources (e.g., budget) within their limits. tndp is formulated as a bilevel programming problem, which is difficult to solve on account of its combinatorial nature. following a recent, heuristic by ant colony optimization (aco), a hybridized aco (haco) has been devised and tested on the network of sioux falls, showing that the hybrid is more effective to solve the problem. in this paper, employing the heuristic of particle swarm optimization (pso), an algorithm is designed to solve the tndp. application of the algorithm on the sioux falls test network shows that the performance of pso algorithm is comparable with haco.