Step Fixed Charge Transportation Problems via Genetic Algorithm

Abstract

In this paper, we consider the step fixed-charge transportation problem where is one of the most important problems in transportation research area. To tackle such an NP-hard problem, we present Genetic Algorithm (GA). Since crossover and mutation operators have significant role on the algorithm's quality, some crossover and mutation operators are tested in this work. For this aim, several problem sizes are generated at random and then through extensive computational experiments, appropriate GA parameter values were chosen. Besides, the efficiency and convergence of the proposed algorithms was evaluated by solution quality. The results showed that the GA was more robust and consistently outperformed Simulated Annealing (SA) for all instances.