Abstract
Multi-depot vehicle routing problem is a NP-hard combinatorial optimization problem. In this paper, we proposed an improved genetic algorithm (GA), which combined GA with fitnessscaling and local search. The experiments compared the proposed approach with standard GA, simulated annealing, Tabu search, and particle swarm optimization. The results showed that the proposed method was superior to GA, SA, TS, and PSO, w.r.t. solution accuracy.