Abstract: To solve the multiple Automated Guided Vehicles(AGVs)scheduling problem in a batching area for materials transportation, a mathematical model was built and an improved genetic algorithm was proposed to minimize material handling time for materials supplement in a batching area of flexible production system. The combination method of integer coding and multi-parameter coding was used in the algorithm. To solve the problem of illegal solutions were easily produced by the conventional crossovers and mutations，a sectional type of crossover method and mutating method were used, and method of expansion on chromosome in discrete random population was also used to initialize the population generation. From the optimization results, this multi-AGVs task allocation and scheduling scheme could be obtained. Through the example of heavy machine assembly workshop and the comparison between this improved genetic algorithm and branch and barrier algorithm as well as conventional genetic algorithm，the feasibility and effectiveness of the method were proved.

Key words: automated guided vehicle, scheduling, mathematical optimization model, task assignment, 　genetic algorithms

摘要： 为解决配料区物料运输多自动导航小车的调度问题，以自动导航小车给柔性生产系统配料区补料搬运时间最短为目标建立数学模型，提出了一种改进的遗传算法进行自动导航小车的任务分配和任务排序优化。算法采用了整数编码和多参数编码相结合的方式，同时为避免常规交叉变异算子易出现非法解的状况，采用了对离散随机种群中染色体进行扩展的初始种群生成方法和分段式交叉变异算子，优化结果给出了多自动导航小车的任务分配方案。通过某重型机械公司的装配车间内多自动导航小车调度的优化实例，并与分支定界算法和未改进的遗传算法进行结果对比，验证了该方法的有效性和可行性。

关键词: 自动导航小车, 调度, 数学优化模型, 任务分配, 遗传算法