基于改进型GERT网络的多工序制造过程质量损失预测研究  

Prediction on Quality Loss for aMulti-StageManufacturing Process Based on an Improved GERT Network

作　　者：李亚平[1] 陶良彦[2] LI Yaping;TAO Liangyan(College of Economics and Management,Nanjing Forestry University,Nanjing,Jiangsu 210037,China;College of Economics and Management,Nanjing University of Aeronautics and Astronautics,Nanjing,Jiangsu 210016,China)

机构地区：[1]南京林业大学经济管理学院,江苏南京210037 [2]南京航空航天大学经济与管理学院,江苏南京210016

出　　处：《工业工程与管理》2021年第5期150-160,共11页Industrial Engineering and Management

基　　金：国家自然科学基金项目(71701098,72171120),教育部人文社会科学青年基金项目(17YJC630070),江苏省自然科学基金项目(BK20160940)。

摘　　要：质量损失是评价质量稳健性和质量水平的一个重要指标。多工序制造产品的质量损失可由末端工序的质量特性波动得到,但质量损失是如何由多工序不断累积得到并不清楚。针对该问题,在剖析多工序之间质量损失传递演化关系的基础上,提出一种基于改进型GERT网络的质量损失预测方法。由于工序间质量损失的耦合关系以及返工对质量的改善作用,均无法用传统的GERT网络直接描述,于是设计两类算子表征这些关系。进一步,设计带有算子的网络结构单元,证明其运算规则,并构建一类融合算子的改进型GERT网络。在此基础上,研究多工序质量损失预测问题。案例说明了所提方法的有效性。Quality loss is an important index for evaluating quality robustness or quality level.The quality loss for a multi-stage manufacturing process can be calculated based on the variations of quality characteristics of terminal stage.However,it is not clear that how this quality loss is obtained according to the quality loss of multiple stages during the manufacturing process.The relationships of quality loss pass between stages were analyzed.Based on this,a prediction method for quality loss for a multi-stage manufacturing process was proposed.The traditional GERT network could not describe some relationships of quality loss,for example,the coupling effect of different stages and improving effect of feedback loop,directly.Two types of operators were designed to represent these relationships.Further,a set of computation rules for the network units of different structures that containing operators were developed.Based on this,an improved GERT network with operators were established.At last,how to predict quality loss was studied.The case study validates the effectiveness of the method.

关 键 词：GERT网络 多工序制造过程 质量损失 预测 

分 类 号：F273[经济管理—企业管理]