An External Archive Guided NSGA-II Algorithm for Multi-depot Green Vehicle Routing Problem

Abstract

AbstractThe paper proposes a new elitist non-dominated sorting genetic Algorithm 2, called External Archive Guided Elitist Non-dominated Sorting Genetic Algorithm 2 (EAG-NSGA-II) to solve the multi-depot green vehicle routing problem (MDGVRP). In the proposed algorithm, the elitist non-dominated sorting genetic Algorithm 2 (NSGA-II) is incorporated with adaptive local search to improve the solution’s quality and greatly accelerate the convergence. To ensure a good balance between convergence and diversity, we employ the external archive based on adaptive epsilon dominance. The experiments are conducted over famous 11 Cordeau’s data sets and compared against three well-known algorithms, including Strength Pareto evolutionary algorithm (SPEA2), Elitist non-dominated sorting genetic algorithm (NSGA-II), and Multi-objective evolutionary algorithm based on decomposition (MOEA/D). The results indicate that the proposed algorithm is highly competitive and outperforms the selected state-of-the-art multiobjective optimization algorithms.KeywordsMulti-depot green vehicle routing problemMultiobjective optimizationEvolutionary algorithmsEpsilon dominanceCrowding distanceGenetic algorithm