Application of Discrete Event Simulation, AI and Machine Learning (ML) in Optimizing Supply Chain and Operations Management: A Systematic Literature Review

Abstract

This review paper examines the integration of Industry 4.0 technologies—specifically, artificial intelligence (AI), discrete event simulation (DES), machine learning (ML), and metaheuristics—in enhancing supply chain and operations management. By optimising manufacturing parameters, these techniques not only reduce overall costs but also increase production efficiency. Digital twins of manufacturing facilities offer additional benefits by validating optimisation results and identifying bottlenecks within existing systems. A systematic literature review (SLR) was conducted using 19 academic papers and journal articles published from 2010 to recent years. Papers were selected through thematic deduction, employing keywords such as algorithm, discrete event simulation, machine learning, artificial intelligence, supply chain, and Industry 4.0. The review first highlights how AI and ML are revolutionising supply chain management by enabling efficient decision-making, optimising process arrangements, and supporting the adoption of innovative methods through simulation-based insights derived from digital twin models. Furthermore, evidence is presented demonstrating significant improvements in production efficiency, cost management, and planning when integrating AI and ML into supply chain operations. The study also addresses the challenges hindering the implementation of these technological advancements in Bangladesh’s textile and ready-made garment (RMG) industries—where, in 2023, Bangladesh ranked as the world’s second-largest exporter of RMG with a revenue of $38 billion. Key challenges include data quality issues, insufficient digital infrastructure, and inadequate data documentation. Given that much of the existing research relies on case studies or reviews, this paper also suggests future research directions, including simulation-based experimentation and the exploration of additional AI techniques such as data mining (DM), long short-term memory (LSTM) networks, and natural language processing (NLP).