A knowledge-based evolutionary algorithm for the multiobjective vehicle routing problem with time windows |
| |
| Affiliation: | 1. Technical University of Crete, School of Production Engineering and Management, University Campus, Chania 73100, Greece;2. Aristotle University of Thessalonike, Department of Civil Engineering, Thessalonike 54124, Greece;3. Luleå Technical University, Industrial Logistics, Luleå 97187, Sweden;1. Guangdong Key Laboratory of Brain-Inspired Intelligent Computation, Department of Computer Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, China;2. Research Institute of Trustworthy Autonomous Systems, Southern University of Science and Technology, Shenzhen 518055, China |
| |
| Abstract: | This paper addresses the multiobjective vehicle routing problem with time windows (MOVRPTW). The objectives are to minimize the number of vehicles and the total distance simultaneously. Our approach is based on an evolutionary algorithm and aims to find the set of Pareto optimal solutions. We incorporate problem-specific knowledge into the genetic operators. The crossover operator exchanges one of the best routes, which has the shortest average distance, the relocation mutation operator relocates a large number of customers in non-decreasing order of the length of the time window, and the split mutation operator breaks the longest-distance link in the routes. Our algorithm is compared with 10 existing algorithms by standard 100-customer and 200-customer problem instances. It shows competitive performance and updates more than 1/3 of the net set of the non-dominated solutions. |
| |
| Keywords: | Vehicle routing problem Time windows Multiobjective Pareto optimal Evolutionary algorithm |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
