Digital-twin-based decision support of dynamic maintenance task prioritization using simulation-based optimization and genetic programming

Modern decision support systems need to be connected online equipment so that the large amount of data available can used guide decisions shop floor operators, making full use potential industrial manufacturing systems. This paper investigates a novel optimization and analytic method implement such system, based on heuristic generation using genetic programming simulation-based running digital twin. Such digital-twin-based system allows proactively searching best attribute combinations in data-driven composite dispatching rule for short-term corrective maintenance task prioritization. Both job (e.g., bottlenecks) operator priorities multiple criteria, including competence, utilization, walking distances floor, bottlenecks, work-in-process, parallel resource availability. The rules are evaluated twin, built real-world machining line, which simulates effects regarding disruptions. Experimental results show improved productivity because generated by compared priority similar attributes methods. improvement is more pronounced when number operators reduced. thus offers new insights about how transformed into useful knowledge with enhance efficiency.