Big data oriented root cause identification approach based on Axiomatic domain mapping and weighted association rule mining for product infant failure. (July 2017)

• Record Type: Journal Article
• Authors: He, Yihai; Zhu, Chunling; He, Zhenzhen; Gu, Changchao; Cui, Jiaming
• Is Part Of: Computers & industrial engineering. Volume 109(2017)
• Page Start: 253
• Page End: 265
• Publication Date: 2017-07
• Subjects: Product infant failure -- Root cause identification -- Big data -- Axiomatic domain mapping -- WARM; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2017.05.012
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Big data of product infant failure is defined. A failure analysis framework of associated tree for product infant failure is brought forth. The WARM is adopted to mine the weight of the possible root cause node in the associated tree. The applicability of the proposed method is illustrated using a washing machine failure analysis example. Abstract: Product infant failure formation mechanism is a maze that remains unclear to most manufacturers. Root cause analysis is an important and challenging task in exploring this mechanism in the era of quality and big data. Therefore, a novel big data oriented root cause identification approach based on weighted association rule mining (WARM) is proposed in this paper. First, the mechanism is expounded based on big data in the product lifecycle, and the requirements of root cause identification are determined simultaneously. Second, in view of domain mapping theory in Axiomatic design, the associated tree is proposed to provide a framework for the root cause search and identification. Then, the big data of root causes is defined based on the proposed associated tree. Third, a root cause mining technique using the WARM is presented, and the weight computation approach for the node on the associated tree based on the weight confidence is provided. Finally, the validity of the proposed method is verified by a case study on mining root causes for severe infant failure of an automatic washing machine. The final result shows that the proposed approach is conducive to heuristically identify the root causes of the complicated product infant failure from the big data of product lifecycle.