基于改进遗传算法的多约束VRP求解

Solving Multi-restriction VRP by an Improved Genetic Algorithm

建立了多约束条件车辆路径问题的数学模型和求解流程。先采用最近插入法生成初始解,然后基于遗传算法和模拟退火算法改进初始解。实验结果表明:结合模拟退火与遗传算法求解车辆路径问题,可以在一定程度上解决遗传算法易“早熟收敛”问题,从而得到更优的解。

This paper established the mathematic model and solving flow of multi-restriction vehicle routing problem. After obtained the original results with ＂nearest insertion heuristic＂ algorithm, a hybrid solution which combined ＂genetic algorithm＂ and ＂simulated annealing＂ was designed to improve the results. The numerical analysis showed this solution overcame the problem of premature convergence of genetic algorithm, and so the results were more optimized.