.png){kind=icon}
返回
Solution knowledge mining and recommendation for quality problem-solving
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cie.2021.107313.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Quality problem-solving (QPS) is one of the most important processes to ensure product quality. In the context of Industry 4.0, increasingly more manufacturing companies have collected rich QPS data yet remain mostly untapped by many firms. To this end, we propose a method of mining solution knowledge from QPS data and recommending the best solutions to problem-solvers. In this method, we first construct a problem-solution bipartite graph model and a problem cluster-solution toolbox (PCST) model, where dispersed solutions solving the same or similar problems are converged to form a solution toolbox so that problem-solvers will have more and reliable options when solving new problems. Given the particularity of the content composition of problems, a two-stage problem clustering method is proposed to acquire common problem clusters. Then the PCST knowledge is obtained according to the problem-solution bipartite relations. Considering the differences in the effectiveness of different solutions, we design a novel reasoning method to select the best solutions from the solution toolbox to solve new quality problems. Finally, the QPS data of an automobile company is used to verify the feasibility and effectiveness of the proposed method, and a prototype system for solution recommendation and problem analysis is designed. The results highlight that the proposed method is of considerable significance to improve the efficiency and effectiveness of QPS in manufacturing companies.
Keyword:
Quality problem-solving
Quality management
Solution knowledge
Knowledge mining
Recommender system
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.5
论文数:
1.0W
被引数:
4.4W
机构
引用论文
.png){kind=icon}
查看更多内容