Research Of Multi-objective Flow Shop Scheduling Based On Fast Non-dominated Sorting Genetic Algorithm

In an era of increasing trend of economic globalization, increasingly diverse of modern production, enterprise put forward an urgent demand in production technology innovation and efficient production management, production scheduling is a core part of enterprise production management, as well as what many scholars have focused on. In the other hand, with the growing production, environmental issues have become important issues to lots of enterprises in production, the energy crisis and carbon emissions problem make our country even most countries in the world facing enormous pressure on energy conservation. Flow shop scheduling has been widely applied in enterprise production, but in the research of flow shop scheduling, most of the literature consider makespan, delay time, production cost and others as main object of optimization, but there is few literature considering workshop energy consumption about shop scheduling problem, the study of such problems has important and practical significance to enterprise production.This paper studies the multi-objective flow shop scheduling problem which considers workshop energy consumption and solves the multi-objective optimization problem with fast non-dominated sorting genetic algorithm(NSGA-II algorithm), focus on a new type of flow shop scheduling of green energy which mainly considers workshop process energy consumption, stop-boot energy consumption, machine idle energy consumption, In addition, this paper do further optimization to the machine idle energy consumption, and presents the analysis of the energy saving effect after dynamic adjustment;Do comprehensive research on multi-objective flow shop scheduling of optimization objects including total energy consumption, makespan, delay time, using multi-objective genetic algorithm to solve the problem, give the Pareto optimal scheduling results and the analysis of results. Optimization results is not ideal because of target duplication exists in late evolutionary of NSGA-II algorithm, this paper make some improvements on NSGA-II algorithm: do further variation for the repeat individuals of each generation to get a new population, and get the next generation of population through competition with the population obtained by genetic manipulation. The improved NSGA-II algorithm and NSGA-II algorithm were comparatively analyzed to verify the validity of the improved algorithm on different scheduling scale;Explore the shop scheduling problem under the constraint of a special kind- flow shop scheduling problems with sequence-dependent setup times. First, the analysis is given in the case of a single optimization objective- makespan, total energy consumption separately for the workshop scheduling, we proposed heuristic LST-GA algorithm, LSE-GA algorithm to study and verify the validity of algorithms on different scheduling scale. Then we use the heuristic rules combined with improved NSGA-II algorithms to analysis the flow shop scheduling with sequence dependent setup times considering energy consumption. Experimental results show that the proposed heuristic rules is effective in solving flow shop scheduling problems with sequence-dependent setup times.