Optimization And Integration Of Fresh Products’ Harvest And Distribution Considering Multi-time Windows

With the improvement of people’s living standards,China’s desire for consumption of fresh products has become stronger,and various fresh e-commerce businesses have emerged.However,high fresh food storage costs,transportation costs,and the problem of rapid fresh food corruption after picking have always been important factors restricting the rapid development of fresh food.Using a reasonable operation model can effectively reduce the loss of fresh food in transportation and storage.At present,domestic cold chain operation is still the mainstay,and other methods are supplementary.In order to quickly respond to consumers ’fresh needs,and to meet consumers’ demand for ready-to-eat food,the harvest-distribution model of fresh food is particularly important.On the other hand,the vehicle routing problem(VRP)has always been one of the hotspots in the research field,and it is also an important means to effectively reduce the cost of logistics distribution.With the strengthening of customers’ management of fragmented time,the problem of multi-time window vehicle routing has gradually gained more and more attention.In this paper,the multi-window vehicle routing problem and the harvest-distribution model are effectively combined to form a multi-time window harvest-distribution integration optimization problem.This question mainly studies the following three aspects:(1)Correlation analysis based on multi-time window vehicle routing problem and harvest distribution problem.Starting from the current logistics dilemma,the harvest-distribution model has a crucial impact on reducing costs and reducing fresh quality corruption.Then comb the research status of harvesting and distribution problems and multi-time window vehicle routing problems,and summarize the key factors of multi-time window fresh harvest-distribution integration optimization.(2)Carry out model and algorithm research on vehicle routing problem with multiple time windows.Starting from the original VRP problem,this paper gradually sorts out the development of the multi-time window vehicle routing problem and the characteristics of the model and algorithm.A model of a standard multi-time window vehicle routing problem with a stronger logical relationship is constructed.On the basis of reviewing the hybrid genetic algorithm,a hybrid genetic algorithm with secondary domain,which is closely combined with neighborhood search and genetic algorithm,is proposed to solve the model.In the absence of a standard test set for vehicle routing problems with multiple time windows,a VRPMTW standard test set is constructed based on the VRPTW standard test set and tested.(3)Research on the integrated optimization model of fresh harvest-distribution with multiple time windows.Due to its inherent fragility,perishability and short life cycle,fresh products are particularly important to reduce the total time required for fresh harvest and distribution while taking into account the cost.This article focuses on the integration of the fresh harvest-distribution scheduling problem in consideration of multiple time windows to reduce the quality loss and operating costs caused by the fresh harvest and fresh distribution processes.In response to this problem,this paper constructs a nonlinear mixed integer programming model that targets the quality loss in the harvest stage,the quality loss in the delivery stage,and the distribution operating cost.Due to the difficulty of solving the multi-objective,this paper uses a standardized method to convert the multi-objective into a single objective The big M method is used to transform the nonlinear constraints into linear constraints,so as to realize the perfection of the objectives and constraints.Afterwards,a hybrid genetic algorithm with secondary domain structure is used to perform numerical experiments and sensitivity analysis.The results show that the algorithm has a good solution effect.At the same time,sensitivity analysis gives us important management implications.This paper constructs a multi-time window fresh harvest-distribution integrated optimization model,and solves it by using a hybrid genetic algorithm with a secondary domain structure.Make decision support to provide management decision support for fresh harvest-distribution operations.