A Review of Production Scheduling with Artificial Intelligence and Digital Twins
Punit Singh, K. Hari Krishnan, Enkhsaikhan Boldsaikhan
Abstract
Digital twin and artificial intelligence (DT-AI) technologies present hitherto unheard-of possibilities for dynamic production scheduling in smart manufacturing. Nevertheless, a careful examination of several studies reveals significant gaps in the current state of the discipline. This paper attempts to review advancements, gaps, and opportunities in the areas of DT-AI-based production scheduling. Articles chosen for this literature analysis were mostly published within the last eight years. Based on the literature, five enabling challenges that are consistently considered in the literature include Dynamic and Unforeseen Disruptions, High System Complexity, Real-Time Data Management, Integration and Interoperability, and Adaptability and Generalizability. This review not only identifies these enabling challenges but also provides tailored outlines of progress and future directions. The findings pave the way for resilient, scalable, and interpretable DT-AI systems for production scheduling that can handle uncertainty and optimize output in real time. DTs and AI can benefit manufacturing with data-driven intelligent planning and decision-making as well as model-based systems engineering principles. This review examines these advancements and trending research directions in production scheduling.