Abstract: To solve the batch scheduling problem of identical parallel melting furnaces with nonidentical job weights, an optimization model is established to minimize the makespan on scheduling melting furnaces, and the hybrid particle swarm optimization based jobs sequence (HPSO) designed. In the HPSO, particles are represented by job sequences, the batch first fit (BFF) heuristic used to group jobs to batches, the longest processing time(LPT) heuristic adopted to assign batches to batch processing machines, and the minimum completion time difference(MCD) heuristic proposed to optimize scheduling results by LPT heuristic. In order to avoid the premature convergence problem, the HPSO introduces the crossover and mutation operator to search for the optimal solution. Compared with SA and GA algorithm, simulation experimental results demonstrated that HPSO has a good performance.

Key words: batch scheduling, non-identical jobs, melting furnace, particle swarm optimization