基于混合遗传算法的物流路径优化方法研究

Study on Optimizing of Physical Routing Method Based on Hybrid Genetic Algorithm

从传统的遗传算法机制出发,针对传统遗传算法在解决物流配送路径优化问题中存在早熟和局部搜索能力不足的缺点,提出一种通过聚类分析中K-means算法与改进遗传算法相结合的混合遗传算法。其中,选择操作采用精英保留模型的锦标赛选择策略,交叉操作采用双切点交叉,变异算子引入k-交换变异操作保证个体逐代进化。通过选择、交叉和变异操作,实现目标函数的最小化,大大缩减了车辆行驶距离,优化了配送路线,并根据数学模型利用实验数据进行仿真实验,结果表明混合遗传算法相对于原有的遗传算法提高了全局寻优能力和算法的收敛速度。

By analyzing the optimization scheme of traditional genetic algorithm,according to the precocious and insufficient ability of the genetic algorithm in solving the optimization of logistics distribution routing problems of local search,we propose a hybrid genetic algorithm combined with K-means algorithm and improved genetic algorithm.In this algorithm,selection operation adopts the championship selection strategy of elitist model of tournament,crossover operation with double-cut point crossover,and mutation operator introduces the k-exchange mutation operation to guarantee the individual evolution by generation. By selection,crossover and mutation operations,the objective function is minimized,the vehicle distance is reduced greatly,and the distribution route is optimized.The experiments are carried on according to the mathematical model and the use of experimental data,which showthat the proposed algorithm with respect to the original genetic algorithm can improve the global search ability and convergence speed.