Machine Learning Applications for Continuous Improvement in Integrated Management Systems: A Short Review

Abstract

The machine learning (ML) field is increasingly impacting industries within the Industry 4.0 paradigm, which among other things, enables the usage of vast amounts of data. Combining data and ML consequently enables continuous improvement that differs from conventional continuous improvement, especially in higher efficiency on default detection. However, scientific literature needs to be updated continuously. Hence, this work aims to gather research information on advancements in ML applications in organisations for continuous improvement purposes. Following the PRISMA Statement, 11 relevant articles were analysed. Results evidenced that most research is case-specific and focuses on applicability. Furthermore, it was observed that most papers are from the manufacturing industry. Although some theoretical research indicates the potential of ML applications in the operations domain (i.e., before and post-production processes), the literature in this segment is lacking. It can be concluded that the potential of ML in management systems is dependent on the amount and quality of data. However, this latter comes at a high cost and therefore, careful costbenefit analysis should be made before adopting ML in businesses. Overall, this review provided promising perspectives for applying ML within quality systems and potential leading ways for continuous improvement in safety, environment and integrated management systems. (c) 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.