Research On The Route Of Fresh Food Picking And Delivery Vehicles In The Same City Based On Ant Colony Algorithm

Today,the city fresh sales market has gradually entered the white-hot stage.The two-way demand of the supply end and the demand end continues to emerge,and the fresh trading in the same city has gradually become a business with high frequency,high growth rate and strong demand at this stage.From the tuyere monitoring and forecasting report given by the E-commerce Research Center,it can be intuitively found that the industry chain related to fresh delivery has a trillion-level market scale,which is expected to become one of the large tuyere in the middle of the 21 st century.And in the city fresh market,the most important thing is experience and convenience,not only to the demand side users convenient and fast quality experience,but also to the supply side merchants with efficient and low-cost convenient service.When more consumption scenes go back offline,the pick-up and delivery personnel are the last and only link between the whole business model and the supply side and the demand side.Especially in the early stage of seizing users,on the premise that online experience tends to be homogenized,the pick-up and delivery link has become a decisive factor affecting the overall experience.From the off,transport costs has become influence overall cost and experience the biggest single,so the vehicle routing problem,through the adoption of a more reasonable path planning,from making off link in meet the requirements under the premise of end user requirements,make the supply side minimization of cost,thus reducing the overall cost to optimize the user experience,has the vital significance.This paper studies the following two kinds of problems with great practical significance: the path planning problem of fresh pickup and delivery vehicles in multiple depots with time Windows and the dynamic path planning problem of fresh pickup and delivery vehicles with time Windows.The main work of this paper includes:1.In this paper,the basic ant colony algorithm is optimized and improved from two aspects,namely,the setting logic of individual search starting point and the updating logic of pheromone along the path of ants,and the simulation experiment proves that the optimized algorithm has better performance.2.Two kinds of realistic problem scenarios are analyzed in detail,and corresponding mathematical models are established,and the optimized ant colony algorithm is used to solve them experimentally.The experimental results based on the desensitization data of Jingdong Logistics prove that the optimized ant colony algorithm in this paper is more reasonable.After using the improved algorithm for practical problems,the result data index obtained is significantly improved compared with the historical data.The availability,robustness and computation time of the algorithm are also within a reasonable range.