People-Centric Evolutionary System for Dynamic Production Scheduling

Evolving production scheduling heuristics is a challenging task because of the dynamic and complex environments interdependency multiple decisions. Different genetic programming (GP) methods have been developed for this achieved very encouraging results. However, these usually trouble in discovering powerful compact heuristics, especially difficult problems. Moreover, there no systematic approach decision makers to intervene embed their knowledge preferences evolutionary process. This article develops novel people-centric system scheduling. The two key components are new mapping technique incrementally monitor process adaptive surrogate model improve efficiency GP. experimental results with flexible job shop show that proposed outperforms existing algorithms evolving terms performance heuristic sizes. also allows interact on fly guide evolution toward desired solutions.