多目标优化算法在AGV调度的应用研究

Research on Application of Multi-Objective Optimization Algorithm in AGV Scheduling

针对多品种小批量生产的车间AGV调度问题,以客户满意度最大化和配送总成本最小化为优化目标,提出了一种有效的多目标优化算法。首先建立了多目标混合整数线性规划模型,在此基础上提出了多目标进化算法。该算法在种群初始化引入了启发式算法,采用基于理想点的多目标局部搜索增强挖掘能力,在非支配解的有价值信息的基础上设计了两点交叉运算方法,通过重启策略避免了所提算法陷入局部最优。最后通过某企业的调度实例进行了对比实验,结果表明,所提算法求解该问题的性能明显优于现有的多目标算法。

Aiming at the scheduling problem of workshopAGV for multi-variety and small-batch production,an effective multiobjective optimization algorithm is proposed with the optimization goal of maximizing customer satisfaction and minimizing the total cost of distribution.First,a multi-objective mixed integer linear programming model is established,and on this basis,a multi-objective evolutionary algorithm is proposed.This algorithm introduces a heuristic algorithm in the population initialization,uses multi-objective local search based on ideal points to enhance the mining ability,and designs a two-point crossover operation method based on the valuable information of the non-dominated solution.The restart strategy avoids the proposed The algorithm falls into a local optimum.Finally,a comparative experiment was conducted through a scheduling example of a certain enterprise.The results show that the performance of the proposed algorithm to solve the problem is significantly better than the existing multi-objective algorithm.