Data-Driven Decision-Making in Shop Floor Quality Management – A Systematic Literature Review

This paper presents a systematic literature review of applied methods for data-driven decision-making (DDD) in shop floor quality management (QM). The goal is to give an overview of publications in DDD QM and to highlight areas where future research can contribute to the advancement of the field. Relevant publications of the past decade are categorized across the following key dimensions: contribution area in QM DDD (e.g., ‘know-what’), characteristic elements of applied methods within DDD (e.g., unsupervised machine learning), and type of manufacturing process (e.g., additive manufacturing). The review reveals a prevalent examination of initial DDD stages like detection (‘know-what’) and a predominance of supervised machine learning approaches. This indicates potential research opportunities in integrating advanced DDD stages (e.g., ‘know-why’) and exploring underutilized methodologies like semi-supervised learning. The findings also suggest a need for broader application across manufacturing processes and a deeper examination of data quality and explainable AI within DDD in QM. This review not only maps the current landscape but also identifies potential areas for future exploration, providing valuable insights for advancing DDD in QM.