Study On Optimization Of Low Carbon Collection And Transportation Routes Of Municipal Solid Waste In Uncertain Environment

Municipal solid waste management（MSW）is a series of process,including classification,collection,transportation and disposal of solid waste.And the process of waste collection and transportation is the most critical link because of its high cost and carbon emissions.Thus,with the demand of rapid urban development and low-carbon economy,how to reasonably deal with the uncertainty factors in the process of waste collection and transportation to reduce cost and carbon emissions and improve the efficiency is a big challenge that the city waste management is facing.In order to consider both economic and environmental benefits of the process of waste collection and transportation,a Two-stage Low-carbon Vehicle Routing Problem in Uncertain Environment（TSLCVRP-UE）is proposed.At the pre-optimization stage,the comprehensive cost composed of routing cost and carbon emission cost is taken as the optimization objective,and the pre-optimization routes are generated and implemented based on the expected value of waste at each collection point.And at the re-optimization stage,the random variables are treated with chance constraint deterministic equivalence,and the collection failure points are determined.Then the re-optimization scheme is implemented:vehicle returns before failure points and the rescheduling strategy is uniformly implied for all the re-optimization points.In order to solve the model,PSOSA,a hybrid heuristic algorithm,is designed based on PSO（Particle Swarm Optimization）and SA（Simulated Annealing）.At first,the initial optimal solution is obtained by PSO algorithm,and then based on the random selection,a series of neighborhood search is operated on the initial optimal solution for new solution.SA with a Metropolis rule accepts non-optimal solutions based on a certain probability is used to improve PSO to avoid falling into local optimum.Finally,the classical CVRP database cases are used to test the optimization ability of the algorithm,and it is proved that the solving effect of PSOSA is better than PSO.Finally,a simulation experiment is carried out on a real case of waste collection and transportation in a certain area.Through the research on the waste collection and transportation in certain environment,the optimization results under different objective functions are compared.Then,the sensitivity analysis of confidence leveland vehicle number6)₁ in the pre-optimization stage was carried out with uncertain conditions.The numerical analysis results show that with the same value of6)₁,the number of vehicles in the pre-optimization stage,the less the number of vehicles6)is,the lower the comprehensive cost will be.Meanwhile,the increase of confidence levelwithin a certain range will not increase the total cost.with the same value of,confidence level,there is an optimal value of6)₁ to obtain the optimal value of route distance,carbon emission or comprehensive cost.Finally,based on the research results,some management suggestions are provided for enterprises and governments.