Adapting a reinforcement learning method for the distributed blocking hybrid flow shop scheduling problem

With the continuous emission of energy in the past years, the environmental problems are becoming more and more serious. For example, in manufacturing, the energy efficient scheduling problem has become particularly prominent, and attracted much attention of the researchers. As a common scheduling problem in the real world, the research on distributed blocking hybrid flow shop (DBHFSP) is very few. In this paper, we will carry out a study of the problem. Because of its NP-hard character, therefore, we use the intelligent optimization algorithm to solve the problem. we firstly introduce the MILP model of the DBHFSP, then, the Q-learning method combined with the IG algorithm framework (IGQ) is proposed to solve this problem. In the experimental part, through the experiment results and comparison with other algorithms in the recent literatures, the proposed algorithm shows the excellent performance in the simulation experiment with the objective of minimizing the energy consumption.