{"title":"A review of strategies, challenges, and ethical implications of machine learning in smart manufacturing","authors":"Yassmin Seid Ahmed , Abbas S. Milani","doi":"10.1016/j.dajour.2025.100591","DOIUrl":null,"url":null,"abstract":"<div><div>Manufacturing organizations continuously need to innovative production strategies and advance their machinery to adapt to evolving business objectives. Machine learning and data mining are now essential techniques for solving various complex manufacturing problems promptly and intelligently. This article reviews recent research from multiple sectors that have employed machine learning to develop intelligent manufacturing processes, while highlighting key challenges and areas that have been partly overlooked. Over the last two decades, scholars have developed numerous AI-based algorithms and approaches to improve manufacturing processes outputs, with scheduling, monitoring, quality, and fault detection being among the main focus areas. The review categorizes smart manufacturing problems into clustering, classification, and regression tasks, and discusses the underlying performance metrics associated with each category. Additionally, the study tackles ethical issues by discussing such important considerations as data privacy, transparency, and fairness in industrial machine-learning implementations. Finally, it emphasizes that many users remain concerned about compliance with global data protection legislations and the need to build trust in autonomous decision-making systems.</div></div>","PeriodicalId":100357,"journal":{"name":"Decision Analytics Journal","volume":"16 ","pages":"Article 100591"},"PeriodicalIF":0.0000,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Decision Analytics Journal","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772662225000475","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Manufacturing organizations continuously need to innovative production strategies and advance their machinery to adapt to evolving business objectives. Machine learning and data mining are now essential techniques for solving various complex manufacturing problems promptly and intelligently. This article reviews recent research from multiple sectors that have employed machine learning to develop intelligent manufacturing processes, while highlighting key challenges and areas that have been partly overlooked. Over the last two decades, scholars have developed numerous AI-based algorithms and approaches to improve manufacturing processes outputs, with scheduling, monitoring, quality, and fault detection being among the main focus areas. The review categorizes smart manufacturing problems into clustering, classification, and regression tasks, and discusses the underlying performance metrics associated with each category. Additionally, the study tackles ethical issues by discussing such important considerations as data privacy, transparency, and fairness in industrial machine-learning implementations. Finally, it emphasizes that many users remain concerned about compliance with global data protection legislations and the need to build trust in autonomous decision-making systems.