A Hybrid Nested Partitions Method for Bi-objective Job Shop Scheduling Problem Considering Energy Consumption and Makespan

摘要

The energy consumption in discrete manufacturing draws extensive attention as the global warming and air pollution. It is critical to consider energy consumption as well as traditional optimization objective of job shop scheduling. This paper addresses a bi-objective optimization problem for job shop scheduling and proposes a hybrid nested partition algorithm with NSGA-II. The nested partition method is a partitioning- and sampling-based strategy that focuses computational effort on the most promising region of the solution space while maintaining a global perspective. The sampling strategy of NSGA-II samples from each of the sub-regions and from the aggregated surrounding region with Pareto-optimal solutions, so that the algorithm quickly converges. The proposed method is evaluated in an application case and the results show that it really optimize the energy consumption as well as makspan.