基于改进遗传算法的物料配送多AGV调度优化

Scheduling of automated guided vehicles for material distribution based on improved genetic algorithm

为解决混流作业车间中物料配送多自动导引车(AGV)的调度优化问题,以AGV配送物料行驶时间最短为目标建立数学优化模型,提出了一种改进的遗传算法进行AGV的任务分配和配送路径优化。在算法设计过程中,采用直接反映AGV配送路径和任务分配的整数编码方式,为避免常规交叉变异过程中产生不可行解的情况,改进了交叉变异算子,采用最好-最坏交叉模式和基因段随机交换的变异模式,获得了优化的调度方案。最后,以某重型机械公司装配车间内物料输送AGV调度优化为实例,并与遗传算法和分支定界法进行对比,验证了所提方法的可行性和有效性。

To solve multiple Automated Guided Vehicles（ AGVs） scheduling problem for material distribution in hybrid flow job shop,it establishes an scheduling optimization model to minimize material handling time for AGVs of materials distribution,proposes an improved genetic algorithm for optimal solution. In the design process of algorithm,it uses integer coding to reflect directly AGV distribution routing and allocation of tasks. In order to face with illegal solutions due to the conventional crossovers and mutations,it applies best-worst route crossover and mutating method. The optimization process provides the multi-AGVs task allocation and scheduling. Finally,taking material distribution of heavy machine assembly workshop as the example,it compares this improved hybrid genetic algorithm with conventional genetic algorithm as well as branch and bound algorithm,proves the feasibility of the method.