Non-identical parallel machines batch processing problem to minimize the makespan: Models and algorithms
研究了加权工件先分批、再分配到不同容量机器上的调度问题，目标是让最后完工的机器总加工时间最短，并提出了数学模型和元启发式算法。
This paper studies a parallel heterogeneous machine batching and scheduling problem in which weighted jobs are first batched, and the batches are then assigned and sequenced on machines of varying capacities. The duration of a batch is the longest time needed to process a job, and the objective is that of minimizing the makespan, or the sum of the batches durations on the machine finishing last. The authors develop polynomial-size mathematical formulations and a variable neighborhood search metaheuristic. Extensive computational results suggest that the flow-based formulation outperforms a compact formulation, despite its larger number of variables. The metaheuristic is capable of producing high-quality solutions within a limited computing time.