The Mangrove Ecology Optimization Algorithm For Green Hybrid Flow Shop Scheduling Problem

Resource scheduling optimization is the key to rational use of resources and create economic value.Hybrid flow shop problem(HFSP)is an important resource scheduling problem in the field of industrial manufacturing.At present,green development concepts such as environmental protection,energy conservation are advocated.Therefore,this thesis studies the green HFSP,which is a combinatorial optimization problem and a NP hard problem.Therefore,a new mangrove ecological optimization(MEO)algorithm is proposed to solve the green hybrid flow shop problem.The main work of this thesis is as follows:1.A swarm intelligence optimization algorithm named MEO was proposed.The relationship between major species in the mangrove ecosystem was abstracted,and the characteristics suitable for intelligent computing were extracted to construct MEO.The algorithm consists of three main parts.(1)Mangrove propagation operator.By abstracting the birth and drifting process of mangrove hypocotyls,the dominant position information was made full use of to promote the overall exploration.(2)Crab foraging operator.The internal relationship of crab population was abstracted as a two-dimensional von Neumann topological structure,and the unique foraging behavior of sand crab population in mangrove ecosystem was simulated to search the optimum.It emphasized the use of environmental information to maintain the balance between global exploration and local development.(3)Bird foraging operator.The internal relationship of bird population was abstracted as a ring topology structure,and the unique foraging behavior of Spoonbill in mangrove ecosystem was simulated.Based on the information of dominant food and habitat,the angle matrix and transformation H were used to update the position,which gave the operator the ability to accelerate convergence and jumped out of local optimum.MEO was compared with six comparison algorithms in IEEE CEC2017 benchmark function f1-f20 of 30 dimensions,then performed T-test.It proved that MEO had good optimization ability.2.MEO was used to solve green HFSP considering the electricity consumption.The mathematical model of the single objective green HFSP considering the electricity consumption was presented.Compared with six algorithms,experimental results showed that MEO in solving the green HFSP considering the electricity consumption was excellent.3.The discrete multi-objective MEO algorithm was proposed to solve green HFSP considering makespan,noise and dust.The mathematical model of multi-objective green HFSP considering makespan,noise and dust was presented.On the basis of MEO,the discrete multi-objective MEO algorithm(DMOMEO)was proposed by using the minimum position rule discretization,multi-objective optimization operator and multi-objective update operator.DMOMEO was used to calculate green HFSP considering makespan,noise and dust.The test result was compared with NSGA2 and MOPSO depended on four evaluation criterion including coverage,number of non-dominated solutions,spread and MRDI,and showed DMOMEO had good performance in solving this problem.