基于顺序启发式进化算法的多目标一维下料问题

Multi-Objective on One-Dimensional Cutting Problem Based on Sequential Heuristic Procedure Evolutionary Algorithm

本文针对原材料有限的一维下料问题,提出降低切割损耗、降低未完成坯料惩罚和降低总成本三个优化目标,运用改进的顺序启发式规则提高算法收敛速度,结合多属性决策的方法,即先用改进的顺序启发式进化算法求出问题的Pareto最优解集,再采用CRITIC法计算各目标的权重,最后运用最小距离法选出一个满意解作为下料方案。通过实验仿真,证明本文的方法可以有效解决该类多目标一维下料问题。

In this paper multi- objective on one- dimensional stock cutting of raw materials limited is examined.The objects are to reduce total trim,to reduce not finished cutting punishment and total cost.A decision- making method for this cutting stock problem is designed.First,the Pareto-optimal set is acquired by using the improved Sequential heuristic procedure evolutionary algorithm.Then the weight of each object is calculated by the CRITIC method.Finally,a satisfactory solution,which is regarded as cutting scheme,is selected from the Pareto-optimal set by minimum distance method.The results of experimentation show that the proposed method is effective.