An effective hybrid collaborative algorithm for energy-efficient distributed permutation flow-shop inverse scheduling

Jianhui Mou,Peiyong Duan,Liang Gao,Xinhua Liu,Junqing Li

DOI: https://doi.org/10.1016/j.future.2021.10.003

IF: 7.307

2022-03-01

Future Generation Computer Systems

Abstract:Distributed scheduling problem, a novel model of intelligent manufacturing, urgently needs new scheduling methods to meet the dynamic market demand. The inverse scheduling in a distributed shop greatly impacts both its energy consumption and productivity. This paper proposes the energy-efficient distributed permutation flow-shop inverse scheduling problem to minimize adjustment and energy consumption simultaneously. This model contains some realistic constraints, controllable processing times and energy consumption factors. To solve the problem effectively, an effective hybrid collaborative algorithm with cooperative search scheme is designed. The heuristic method and random method are improved to initialize the population. In order to balance the global exploration and local development ability of the algorithm, a double-population cooperative search link based on learning mechanism is presented. Moreover, a dual-mode local search based on switching mechanism is addressed. The influence of key parameters on the performance of the algorithm is proved by using the ANOVA of design-of-experiment. Finally, the developed algorithm has been compared with other approaches for distributed inverse scheduling and assessed by satisfactory results on different set of distributed inverse scheduling problems.