作业车间区间型多属性瓶颈识别方法

摘要

Aiming at the Job shop bottleneck identification problem under random disturbance resulting in the difficulty of obtaining the determinate value of machine feature attribute, interval form was used to describe these uncertain attributes of machine. Furthermore, a new interval multi-attribute bottleneck identification model was established, and an interval TOPSIS bottleneck identification approach was proposed. By considering the close relationship between bottleneck utilization and bottleneck identification, an integrated framework, under which they could be solved simultaneously, was presented. This framework included two layers. In the first layer of bottleneck utilization, the Plant-Simulation platform was used to simulate random disturbance including equipment failure. Genetic algorithm (GA) was applied to perform optimization and simulation for the scheduling problems under the random disturbance and the optimum scheduling solution was obtained. In the second layer of bottleneck identification, based on scheduling optimization, interval TOPSIS bottleneck identification approach was proposed to identify bottleneck machines with considerations of multiple feature attributes. Comparing the proposed approach with machine utilization, bottleneck occurrence rate and shifting bottleneck detection method in the existing literatures, the results demonstrated the effectiveness of this approach. Finally, the influence of machining cost and material cost on bottleneck identification was analyzed.%针对扰动情形下作业车间瓶颈识别时机器的特征属性难以用确定值表示的问题,采用区间形式描述机器特征属性,构建了区间型多属性瓶颈识别模型,提出了区间TOPSIS多属性瓶颈识别方法.考虑瓶颈识别与瓶颈利用的紧密关系,提出了先进行瓶颈利用再进行瓶颈识别的统一框架.其中瓶颈利用层基于Plant-Simulation 仿真平台设置了机器故障等随机扰动,采用遗传算法对扰动情形下的调度问题进行了优化仿真,获得了适应扰动情形的最优调度优化方案；瓶颈识别层基于调度优化方案,综合考虑了瓶颈的多维特征属性,采用区间TOPSIS多属性瓶颈识别方法识别了瓶颈机器.通过与机器利用率、瓶颈出现率和移动瓶颈识别法等进行比较,验证了所提方法的有效性.最后,分析了制造成本和原材料成本两个参数对瓶颈识别的影响.