Research On Dynamic Demand Terminal Distribution Route Optimization Of Y Fresh Supermarket Under The New Retail Format

With the upgrading of people’s consumption structure and the trend of consumption concept towards service-oriented experience,traditional online retail is gradually transforming into new retail format,thus many large Internet enterprises have expanded offline fresh electricity business for a new retail format of online and offline integration development.For fresh e-commerce in the new retail state,the quality of terminal delivery service which connects enterprises and consumers is not only the core element but also an important means to improve the experience of consumer services.In view of the high cost,distribution delay,route chaos and low vehicles utilization in terminal distribution of Y Fresh Supermarket,this thesis further considers the situation of dynamic new orders for the distribution process,and studied the route optimization problem of the dynamic demand terminal distribution of fresh supermarkets.In order to realize the path optimization of the dynamic demand terminal distribution of fresh supermarkets,this thesis have taken into account the cost of vehicle distribution,the fixed cost of vehicles and the cost of time punishment,and built the two-stage model of dynamic demand terminal distribution combined with vehicle load and soft time window constraints:(1)For the state of initial order delivery,an distribution route optimization model with the target of smallest cost is established,as the model involves plenty of customers and scattered distribution,an improved genetic algorithm is employed.(2)For the state of dynamic new orders during distribution process,the local optimization of new order is carried out,which uses a dynamic insertion algorithm to insert new orders through key points into the initial distribution vehicle route,to establish a dynamic order distribution route optimization model with the target of smallest cost,and then get the dynamic order distribution route optimization scheme.No matter to take the distribution business data of Y Fresh Supermarket’s offline stores in a certain area of Nanjing in a certain period of time as a sample for analysis,by substituting it into the model and using the improved genetic algorithm,the initial order distribution scheme is obtained.Whenever it appears dynamic new orders,on the one hand,the dynamic insertion algorithm is used to insert the new orders into the initial order distribution scheme based on the two-stage processing route optimization strategy,and the vehicles being distributed are selected to return to the store for pickup to complete the distribution service with the goal of minimizing the total variable cost.On the other hand,based on the initial order route optimization strategy,stores directly arrange new vehicles to complete the delivery service of new orders,and use improved genetic algorithm to solve the delivery scheme.The two-stage processing route optimization strategy and the initial order path optimization strategy are analyzed and compared to solve the delivery scheme,it is found that the two-stage processing path optimization strategy is much better than the initial order route optimization strategy in saving vehicle resources and improving vehicle utilization ratio.Therefore,the proposed model effectively improves the problems of Y Fresh Supermarket with high distribution cost and low delivery punctuality and vehicles utilization,as well as provides certain reference for research on dynamic new order route optimization of fresh supermarkets under the new retail format.