A new diversification measurement technique in metaheuristics for vehicle routing problems

Abstract

This article presents a technique to measure the diversification effect of local search moves in metaheuristic algorithms for solving capacitated vehicle routing problems. The diversification values calculated through this technique can be used to modify the algorithms, pick the best neighborhoods in each move and improve the searching process to find the optimal solution. This technique may be implemented to a wide variety of metaheuristics. It is shown that metaheuristic algorithms for vehicle routing problems may be examined efficiently and improved with this technique. Moreover, we examine the solutions for different instances produced by this technique on simulated annealing algorithm and we interpret the relations between the optimal solution and the diversification scores