An efficient multi-objective iterated greedy algorithm for solving distributed flexible job shop scheduling problems

Abstract

The distributed flexible job shop scheduling problem (DFJSP) is a representative challenge in intelligent manufacturing, requiring coordinated decision-making on job-to-factory assignment, machine selection, and operation sequencing under heterogeneous resources. To tackle the trade-off between makespan and energy consumption, this paper proposes a multi-objective iterated greedy (MOIG) algorithm tailored for energy-aware scheduling. Building upon the iterated greedy framework, two complementary reconstruction strategies are introduced: a delay-aware insertion mechanism for makespan reduction, and an idle energy evaluation strategy for energy minimization. A dynamic selection mechanism is employed to adaptively balance the two strategies based on population feedback. To further enhance search capability, five destruction–reconstruction operators are designed to diversify local structures, while a tabu-based local search is integrated to refine solution quality. The proposed MOIG is evaluated on 40 benchmark instances. Experimental results show that MOIG outperforms three representative algorithms in terms of inverted generational distance and set coverage, validating its effectiveness and robustness in solving multi-objective DFJSP.