Location And Routing Optimization Of Omni-channel Mode Based Pre-positioning Warehouse

E-commerce is an important part of the tertiary industry in China.Meanwhile,novel coronavirus pneumonia has been greatly promoted by the rapid development of ecommerce.The continuous upgrading of consumer demand and the intensification of market competition promote the e-commerce pre-positioning warehouse from warehousing mode to omni-channel mode.The rapid development of omni-channel mode of pre-positioning warehouse reshapes the terminal distribution network of e-commerce.Therefore,it is practical significance to study the location and distribution routing optimization of Omni channel mode based pre-positioning warehouse.It can not only vigorously promote the transformation of omni-channel pre-positioning warehouse business from high speed to high quality,but also effectively meet the escalating needs of consumers and the increasing market competition environment.And then promote the transformation of e-commerce retail enterprises to high-quality development.In this dissertation,system dynamics,multi-objective programming,distributed robust optimization and integer programming are used.Study is based on primary research and secondary research,combined with qualitative and quantitative analysis.It simulates two types of pre-positioning warehouse operation modes from the perspective of commercial operation.Then,it plans pre-positioning warehouse location and distribution routing from the perspective of omni-channel pre-positioning warehouse distribution network design.Finally,it optimizes the location-routing problem from the perspective of omni-channel pre-positioning warehouse distribution network optimization.(1)Simulation analysis and evaluation of two types of pre-positioning warehouse operation mode based on system dynamics.First of all,study defines the operation mode,characteristics and problems of omni-channel mode pre-positioning warehouse.Secondly,dissertation constructs the system dynamics model from the perspective of Omni channel mode and pre-positioning warehouse mode.Combined with the case data of empirical enterprises,study uses VENSIM software to simulate and evaluate the operation mode of two types of pre-positioning warehouse.The results show that the omni-channel mode of pre-positioning warehouse is superior to the warehouse mode in terms of consumers,order volume,operating revenue and enterprise capacity,but it is also higher than the warehouse mode of pre-positioning warehouse in terms of total cost.Therefore,reducing the total cost of omni-channel mode pre-positioning warehouse has become the primary goal of optimizing its location and distribution problem.(2)Joint optimization of location and capacity design of omni-channel mode prepositioning warehouse under determined demand.Based on the description of the deterministic demand of consumptions portrait,dissertation not only considers the distribution service,but also introduces the way of in-store purchase based on the characteristics of the omni-channel model.Considering the limitation of the capacity of the pre-positioning warehouse,total cost,travel time of in-store purchase and time of delivery,dissertation constructs the two-objective programming model with the minimum total cost and the minimum service time by using the multi-objective optimization method.By analyzing the structural characteristics of the model,a multi-objective simulated annealing algorithm based on elitist strategy is designed.Study uses MATLAB software combined with the case data of empirical enterprises to verify the correctness of the model and the effectiveness of the algorithm.The results show that increasing the number of pre-positioning warehouse location and considering the way of delivery and In-store purchase(self-collection)can improve the number of Pareto optimal solutions,and also increase the number of location options.The number of Pareto optimal solutions is significantly affected by the decrease of pre-positioning warehouse distribution capacity,and the location scheme is less improved.(3)Distributed robust optimization of vehicle routing in omni-channel mode prepositioning warehouse under uncertain demand.Based on the uncertainty of demand parameters and distribution,dissertation comprehensively considers multiple modes of distribution and pick-up(In-store purchase),vehicle fixed cost,distribution cost,and customer cost of pick-up(In-store purchase).A distributed robust optimization model with minimum expected total cost and conditional value at risk(CVa R)in the worst case is constructed in this study.Because the model belongs to positive semidefinite programming problem,a box inexact set is constructed to derive the equivalent solvable form of the model.Dissertation uses the case data of empirical enterprises and uses CPLEX software to solve the equivalent model.In this study,the effectiveness of the model is verified by comparing with the stochastic programming model.The results show that the distributed robust method has better stability than the stochastic programming method,and can provide more robust distribution service strategies.(4)Location-routing integrated optimization of omni-channel mode pre-positioning warehouse under the change of customer demand.Base on the background of omnichannel mode,pre-positioning warehouse not only provides delivery services,but also pull-in the omni-channel mode characteristics of in-store purchase.Considering the construction cost of the pre-positioning warehouse,fixed cost of vehicle use,transportation cost from the regional warehouse to the pre-positioning warehouse,and the distribution cost from the pre-positioning warehouse to customer,dissertation uses the integer programming method to construct the location-routing optimization model with minimum total cost for omni-channel mode pre-positioning warehouse.In this study,JAVA language is used to implement the improved genetic algorithm design.Research solves and verifies the model based on the case data of empirical enterprises.The results show that the integrated optimization is better than the single objective optimization.At the same time,setting the appropriate maximum distribution distance and optimizing the performance of distribution vehicles will help to control the total cost of enterprises.51 Figures,35 tables,and 178 references...