Vehicle Routing Problem in Distribution with Two-Dimensional Loading Constraint

WANG Zheng,HU Xiang-pei,WANG Xu-ping

2011-01-01

Abstract:In real distributions,there is a great amount of transportation problems called vehicle routing problem with two-dimensional loading constraint,in which items demanded by customers are usually fragile and needed to be packed into vehicles.This is a new problem that combines the two classical problems of vehicle routing problem and bin packing problem.To solve the problem,a mathematical model and a Memetic algorithm were presented after an explicit problem definition.The key modules of the algorithm, such as a depth-first based heuristic loading method,an encoding and splitting method of chromosomes, an initial solution generation method,crossover and local search methods,were explained in details.Based on some preliminary experiments,the best combination of parameters' values in Memetic algorithm was given.Finally,the robustness and the effectiveness of the Memetic algorithm were tested on Iori's 30 instances whose numbers of customers are from 20 to 199 and a comparison with other algorithm in the literature was made,which shows that the memetic algorithm greatly surpasses the algorithm on aspects of solving capacity and solution quality.