Hybrid evolutionary algorithm with sequence difference-based differential evolution for multi-objective fuzzy flow-shop scheduling problem

Wenqiang Zhang,Chen Li,Weidong Yang,Mitsuo Gen
DOI: https://doi.org/10.1504/ijims.2022.128636
2022-01-01
International Journal of Internet Manufacturing and Services
Abstract:In the actual production process of a factory, there are often many uncertain factors, and researchers usually use fuzzy time to express this uncertainty. In this regard, a hybrid evolutionary algorithm with sequence difference-based differential evolution (HEA-SDDE) is proposed to solve fuzzy flow-shop scheduling problem (FFSP). Firstly, the algorithm uses a hybrid sampling strategy based a multi-objective evolutionary algorithm to guide the population to quickly converge to multiple areas of the Pareto front (PF). Secondly, the proposed algorithm applies a sequence difference-based differential evolution (SDDE) strategy, which uses exchanging sequences to determine the sequence differences between individuals, thereby improving the poorly performing individuals in the population. The experiment compares HEA-SDDE with multiple algorithms on 12 problems of different scales for the multi-objective fuzzy flow-shop scheduling problem (MoFFSP). The results demonstrate that the proposed HEA-SDDE has good convergence and distribution performance.
What problem does this paper attempt to address?