An Efficient and Improved Coronavirus Herd Immunity Algorithm Using Knowledge-Driven Variable Neighborhood Search for Flexible Job-Shop Scheduling Problems

By addressing the flexible job shop scheduling problem (FJSP), this paper proposes a new type of algorithm for the FJSP. We named it the hybrid coronavirus population immunity optimization algorithm. Based on the characteristics of the problem, firstly, this paper redefined the discretized two-stage individual encoding and decoding scheme. Secondly, in order to realize the multi-scale search of the solution space, a multi-population update mechanism is designed, and a collaborative learning method is proposed to ensure the diversity of the population. Then, an adaptive mutation operation is introduced to enrich the diversity of the population, relying on the adaptive adjustment of the mutation operator to balance global search and local search capabilities. In order to realize a directional and efficient neighborhood search, this algorithm proposed a knowledge-driven variable neighborhood search strategy. Finally, the algorithm's performance comparison experiment is carried out. The minimum makespans on the MK06 medium-scale case and MK10 large-scale case are 58 and 201, respectively. The experimental results verify the effectiveness of the hybrid algorithm.

Automated summary

This paper proposes a new algorithm called the hybrid coronavirus population immunity optimization algorithm to solve the flexible job shop scheduling problem. The algorithm redefines the encoding and decoding scheme, designs a multi-population update mechanism and collaborative learning method, introduces an adaptive mutation operation, and proposes a knowledge-driven variable neighborhood search strategy. The experimental results show the effectiveness of the hybrid algorithm.