Research On Optimization Of The Combined Delivery Problem For O2O Take-out With Single Pickup Point

In recent years,the rapid development of the O2 O takeout market and the continuous increase in the requirements for the timeliness of takeout delivery by residents have made immediate delivery of takeout with the goal of improving customer satisfaction an urgent problem to be solved.In reality,when the delivery men on the O2 O takeout platform grab orders,they often face a dilemma between guaranteeing customer satisfaction and reducing delivery costs in a single delivery.Considering the characteristics of delivery men who are used to serving fixed business districts,researching the optimization of O2 O take-out order delivery optimization at a single pickup point is of great significance.Based on the systematic description of the research background,significance and development status of O2 O takeout crowdsourcing logistics,the characteristics of O2 O takeout crowdsourcing logistics are analyzed in detail,and the optimization problem of the combined delivery for O2 O take-out with single pickup point is proposed and studied.The research content is as follows:(1)For the single vehicle optimization problem of the combined delivery for O2 O take-out with single pickup point,considering the demand orders arriving real-time,we establish an online model of order allocation and path optimization to minimize the sum of completion time of each order,and we use online theory and competitive analysis method to prove the lower bound of the problem on the general metric space.The WR strategy and the WI strategy are designed for the case when the demand orders release only on the positive half axis.The worst case analysis method is used to prove the competitive ratios of these two strategies.It is found that the WR strategy is better than the WI strategy in the worst case.for the case when the demand orders release on the general metric space,we prove the competitive ratio of the WI strategy.The numerical simulation method is used to compare the advantages and disadvantages of the WR strategy and the WI strategy on the general metric space.The results show that the WI strategy is better than the WR strategy in a random situation.(2)For the multi vehicle optimization problem of the combined delivery for O2 O take-out with single pickup point,a theoretical model is established to minimize the average completion time of each order,and the dynamic problem of real-time delivery is statically processed using the concept of delayed delivery.Then we design a greedy adjacent bundling strategy and an optimized bundling strategy based on VRP(vehicle routing problem).And the numerical simulation method is used to study the effectiveness of these two strategies,and the stability and applicability through the model parameters sensitivity simulation research,including the network structure,order quantity,number of delivery staff,car capacity and so on.to verify Further,we compare the advantages and disadvantages of the two strategies in different situations.Some suggestions for the adoption of strategies are provided for the future development of the take-out instant delivery platform.This research is an innovative study of the dynamic vehicle routing problem in the new context of instant delivery.Applying it to the logistics crowdsourcing platform can promote the platform function upgrade,thereby promoting the development and application of intelligent logistics.