A Hybrid Large Neighborhood Search Method for Minimizing Makespan on Unrelated Parallel Batch Processing Machines with Incompatible Job Families

This paper studies a scheduling problem with non-identical job sizes, arbitrary ready times, and incompatible family constraints for unrelated parallel batch processing machines, where the batches are limited to jobs from same family. The objective is minimize maximum completion time (makespan). important has wide applications in semiconductor manufacturing industries. study proposes mixed integer programming (MIP) model, which can be efficiently optimally solved by commercial solvers small-scale instances. Since known NP-hard, hybrid large neighborhood search (HLNS) combined tabu strategy local proposed solve large-scale problems, lower bound evaluate effectiveness of algorithm. algorithm evaluated on numerous compatible benchmark instances newly generated results computational experiments indicate that HLNS outperforms solver while existing Meanwhile, comparison strategies.