Feature selection methods for root‐cause analysis among top‐level product attributes

Abstract

Manufacturing companies not only strive to deliver flawless products but also monitor product failures in the field to identify potential quality issues. When product failures occur, quality engineers must identify the root cause to improve any affected product and process. This root‐cause analysis can be supported by feature selection methods that identify relevant product attributes, such as manufacturing dates with an increased number of product failures.In this paper, we present different methods for feature selection and evaluate their ability to identify relevant product attributes in a root‐cause analysis. First, we compile a list of feature selection methods. Then, we summarize the properties of product attributes in warranty case data and discuss these properties regarding the challenges they pose for machine learning algorithms. Next, we simulate datasets of warranty cases, which emulate these product properties. Finally, we compare the feature selection methods based on these simulated datasets. In the end, the univariate filter information gain is determined to be a suitable method for a wide range of applications.The comparison based on simulated data provides a more general result than other publications, which only focus on a single use case. Due to the generic nature of the simulated datasets, the results can be applied to various root‐cause analysis processes in different quality management applications and provide a guideline for readers who wish to explore machine learning methods for their analysis of quality data.

Knowledge mapping of digital twin and physical internet in Supply Chain Management: A systematic literature review

“…When product failures occur, then root-cause analysis is required. A dedicated research by Detzner and Eigner (2021) evaluated different methods and suggested an optimal one for feature selection in root-cause analysis. Other works with similar direction include improving assembly process quality management supported by learning algorithm (Franciosa et al, 2020), and process dynamic optimisation (Wang et al, 2021).…”

Reliability analysis of systems with discrete event data using association rules

“…Association rule learning, a rule‐based machine learning methodology, is often used to identify the significant relationships among the events in a large database 21,22,26,27 . It aims to discover strong rules in databases by using some measures of interest.…”

“…Association rule learning, a rule-based machine learning methodology, is often used to identify the significant relationships among the events in a large database. 21,22,26,27 It aims to discover strong rules in databases by using some measures of interest. Several strong rules are discovered to provide insights into machine operation and can be further applied for decision making.…”

Variable selection wrapper in presence of correlated input variables for random forest models

“…5 The third aim is to improve understanding of the underlying process that generated the data. 6 The identification of all variables, which are in some circumstances relevant for classification or regression purposes, is the so-called all-relevant problem. The goal is to identify and prioritize all influential variables for further investigation with domain expertise.…”

“…The third aim is to improve understanding of the underlying process that generated the data 6 . The identification of all variables, which are in some circumstances relevant for classification or regression purposes, is the so‐called all‐relevant problem.…”