Considering the peak power consumption problem with learning and deterioration effect in flow shop scheduling

Abstract

This paper investigates the permutation flow shop scheduling problem with peak power constraints under sequence-dependent setup time, learning, and deterioration effects to minimize the makespan, where the peak power consumption satisfies a given upper bound at any time. We establish relevant mathematical models based on the characteristics of the scheduling environment and set up five setup time-based heuristics, including the earliest start time, the latest setup time based on balance job–machine, latest setup time based on balance machine–job, latest setup time insert based on balance job–machine, and latest setup time insert based on balance machine–job. Similarly, a hybrid genetic algorithm combined with simulated annealing is proposed to prevent premature trapping in local optima. The algorithms are evaluated through a large number of data experiments, and the results show that it can effectively solve this scheduling problem.