Improving the Performance of Genetic Algorithm in Capacitated Vehicle Routing Problem using Self Imposed Constraints

The capacitated vehicle routing problem (CVRP) is a well known member of the family of NP hard problems. In the past few decades, a number of heuristics was introduced to solve this problem but no heuristic can claim to work well in all possible scenarios. In the literature, genetic algorithm (GA) even lags behind the other heuristics. In this paper, we reveal some of the reasons for the inferior performance of GA, and propose a number of mechanisms to improve its performance. A number of test problems are solved to demonstrate the usefulness of the algorithm.