Reinforcement Learning-Based Model for Optimization of Cloud Manufacturing-Based Multi Objective Resource Scheduling: A Review

Cloud-based resource scheduling problem is a typical combination optimization problem, and the task allocation problem refers to how to use resources most efficiently for the completion of a limited set of tasks in a cloud manufacturing system (CMfg). Due to their dynamic nature, and real-time requirements, CMfg faces great challenges in optimizing resource scheduling problems. Since machine learning was developed, a variety of decision-making problems have been solved using Reinforcement Learning (RL). This review paper is aiming to discuss aspects of an RL-based algorithm for optimization of resource scheduling in CMfg through investigating the literature to date. To this end, first, multi-objective resource scheduling is defined and elaborated. Subsequently, the aspects of RL algorithms are presented through their fundamental elements to optimize the scheduling model. Finally, the findings of the review paper are discussed and some suggestions for potential future research to further consolidate this field have been enumerated.