Energy Efficiency Optimization Scheduling Based On Multi-objective Hybrid Artificial Bee Colony Algorithm

The development of manufacturing system has been plagued by the inefficient energyuse. The issue of energy research has been a hot topic in recent years. In actual production,reducing the idle and wait time of machines are one of the effective ways to improve theutilization of workshop equipment and reduce the emission of some harmful gas. Amethod to make each processing step more continuous by scheduling can effectivelyreduce the waiting time of a machine, which has broad application prospects in themanufacturing sector.The vast majority of scheduling problems are NP-hard problem. In recent yearsswarm intelligence algorithm were successfully used in solving scheduling problems andhas achieved good results. In this paper two kinds of hybrid multi-objective artificial beecolony algorithm (MOHABC) was designed to solving the scheduling problem aboutenergy efficiency optimization. The algorithm connecting the global search ability of ABCwith local search features of variable neighborhood search can greatly improve the searchability of a single intelligent algorithm. We mainly consider two problems of the energyconsumption in this paper, the single-machine with two objectives and permutation flowshop problem with three objectives.In this paper, a framework to solve the single machine scheduling problems thatminimize the total energy consumption and total tardiness is proposed. In the framework,the working state of the machine was considered to avoid wasting energy in thescheduling. Finally the proposed framework was utilized on a case study. Meanwhile thegrey relational analysis was proposed to select the best solution from the Pareto front.Through computing and analysis of the solution from the Pareto by the multi-objectivedecision grey relation grade, giving the decision-making solution.Constrained to the machining sequence and the process time, there was a lot of idletime during processing in the permutation flow shop. Affected by the traditional processmethod, the spindle was always run at maximum speed, although it can decrease theprocessing time in some extent, but the relatively utilization efficiency of energy andmachine will greatly decreasing. Based on the above research, a permutation flow shopscheduling model based on energy with the optimization objectives of total completiontime, total energy consumption and maximum peak load was established. The machinecutting parameters was introduced to scheduling for the first time and an example wasdesigned to test this problem. A local search strategy with variable neighborhood search and simulate annealing was used in the algorithm, through the study of specific cases thefeasibility of algorithm and mode were verified.Finally, a summary of our work based on the research is given. At the same time, thefurther research directions of both are analyzed and forecasted, which can be regarded as areference for subsequent researchers.