Predictive waste management in SMEs through lean manufacturing and machine learning

Abstract

This study examines the integration of Lean Manufacturing (LM) and Machine Learning (ML) for waste prediction and production process optimization in Small and Medium Enterprises (SMEs) in the digital era. The study employed a Systematic Literature Review (SLR) approach following the PRISMA framework by analyzing publications retrieved from Scopus, Web of Science, ScienceDirect, IEEE Xplore, Google Scholar, and national accredited journal databases. The review focused on studies published within the last ten years and applied explicit inclusion and exclusion criteria. From an initial pool of 485 records, 15 studies were selected for final analysis. Data were synthesized using thematic analysis to identify key themes, integration mechanisms, benefits, and implementation challenges associated with Lean Manufacturing and Machine Learning integration. The findings revealed five dominant themes: waste identification and reduction, predictive analytics and forecasting, production process optimization, implementation barriers in SMEs, and digital transformation readiness. The review indicates that Lean Manufacturing contributes to the systematic identification and elimination of operational waste through continuous improvement practices, while Machine Learning enhances predictive capabilities through data-driven analysis, real-time monitoring, and early detection of operational inefficiencies. The synthesis further suggests that integrating LM and ML may enable SMEs to shift from reactive waste management toward more predictive and proactive production management. However, several implementation barriers were identified, including limitations in human resources, technological infrastructure, data availability, and organizational readiness. This study contributes a conceptual framework explaining the theoretical mechanisms linking LM and ML in SME production systems. The framework highlights how predictive analytics can support waste reduction, process optimization, and digital transformation initiatives. Nevertheless, the findings are derived from literature synthesis rather than primary empirical evidence and therefore require further validation through future case studies, surveys, and industrial implementation projects.