Early fault root cause identification method based on domain mapping and weighted association rules

Abstract

The invention discloses an infant failure root cause identification method based on domain mapping and a weighting association rule. The method comprises the steps of 1, establishing an association tree frame model of infant failures based on axiomatic domain mapping; 2, decomposing and mapping a function domain; 3, decomposing and mapping a design structure domain; 4, decomposing and mapping a process technology and finishing establishing a failure root cause association tree; 5, collecting failure symptom data; 6, mining node initial weights of failure association tree nodes; 7, evaluating failure association weights; and 8, measuring weight coefficients, thereby finishing identifying root causes. According to the method, the cause node weight coefficients are subjectively quantified by employing the weighting association rule; the potential root causes resulting in the product infant failures are sorted scientifically; the root cause identification qualitative and one-sided defects due to the fact that a traditional method is subject to infant failure mechanism understanding deficiency is made up; the intelligence and scientificity of the root cause identification process can be effectively improved; and the new thought is provided for preventing, controlling and improving the infant failures.

Description

technical field [0001] The invention provides an early fault root cause identification method based on domain mapping and weighted association rules, which belongs to the technical field of reliability modeling and analysis. Background technique [0002] Early product failures are mainly caused by manufacturing defects. The shortcoming of process reliability research is mainly reflected in the high early failure rate of products. The current understanding of early failure rates is only estimated by using data or superficially removed by aging tests. How to prevent and control early failures from the source of manufacturing and design has not been a good way. The bottleneck lies in the lack of systematic research on the early failure mechanism of products from the perspective of product reliability formation, and the inability to accurately identify and locate products that lead to The root cause of early failures (key design and manufacturing parameters) has not been studied...

AI Technical Summary

Problems solved by technology

In view of the unsystematic mining of the detailed root causes of early product failures in the past, which hindered the targeted solution to the reliability problems at the root, the present invention provides an early failure root cause identification method based on domain mapping and weighted association rules, forming a design function problem, The hierarchical structure of physical structure defects and process parameter fluctuations establishes the mapping relationship between early product failure characteristics and influencing factors at all levels to quantitatively trace back to the root of the problem

Images