Root Cause Identification Approach Based on FP-Growth for Product Quality Accident

The Product Quality Accident (PQA) is a kind of accident mainly caused by product quality defects originated in design and production. To decrease the number and the severity of PQA in usage is a routine task of quality and reliability engineer. Especially with the advent of the era of intelligent manufacturing and big data, the functional structure of product is becoming increasingly complicated and the dimension of big data in product lifecycle is high. As a result, it is difficult to artificially identify the root cause of PQA by traditional method. Therefore, this paper proposes a novel method based on an advanced data mining algorithm for root cause identification. Firstly, the quality accident formation mechanism is introduced in detail. Secondly, with the aid of the domain mapping theory, PQA relevance tree is contributed. Thirdly, with the setting of reasonable support and confidence, the FP-Growth algorithm is used to further mine the relevance rules and construct a complete PQA relevance tree, which could reduce the ambiguity of PQA root cause identification. Finally, the effectiveness of the proposed technique is verified by a root cause analysis example of an engine quality accident.