基于多目标果蝇算法面向低碳的车间布局与调度集成优化

Integrated Optimization of Workshop Layout and Scheduling to Reduce Carbon Emissions Based on a Multi-objective Fruit Fly Optimization Algorithm

以往车间布局和调度优化都是各自分开进行的,单独车间布局优化时一般假设待加工工件各工序的加工设备已经确定;而单独调度优化则在车间布局确定后进行的,这种分开优化的方式忽略了不同布局对工序加工设备间距离的影响,由此影响工序间的搬运时间,从而影响调度结果。为此,提出以车间制造过程总碳排放和总完工时间最小为优化目标的车间布局和调度集成优化模型。为了求解该模型,设计多目标果蝇优化算法(Multi-objective fruit fly optimization algorithm,MFOA)。为了增强MFOA全局搜索能力和维持算法的稳定性,设计一种基于混合步长的嗅觉搜索;为了增大果蝇种群的协作,避免陷入局部最优引入了全局协作机制。将非支配等级排序方法引入MFOA处理多目标问题,并用算例验证了多目标果蝇优化算法的有效性。将集成优化结果与车间机群式布局下的调度优化结果和将车间布局、调度分开优化的结果分别进行对比,说明提出的集成优化模型可以得到更低的碳排放,验证了模型的有效性。

The traditional way of workshop layout and scheduling are optimized independently. The optimization of workshop layout is generally based on an assumption that machining equipment of each process had been confirmed, and scheduling is optimized on a given workshop layout. The influence of different layout schemes on distances between two machining equipment for each process are ignored and the handling time of workpieces between two processes might also be changed, which will led to different results of scheduling. Therefore, an integrated optimization model of workshop layout and scheduling aimed at minimizing carbon emissions in its manufacturing process and minimizing completion time is proposed. To solve the model, a multi-objective fruit fly optimization algorithm （MFOA） is designed. To improve the global search ability and maintain the stability of the algorithm, a smell search based on hybrid step is developed. To increase the fruit fly population collaboration and avoid to local optimum of MFOA, global collaboration mechanism is introduced. Meanwhile, the dominance hierarchy ordering method is adopted to deal with the multi-objective problem. The effectiveness of the proposed algorithm is verified by a case study. The results of proposed model are compared respectively with the results of scheduling based on an equipment group Layout, and the results of traditional way of workshop layout and scheduling in which layout and scheduling are optimized independently respectively. The effective of proposed model and algorithm are validated by the case study.