Research On The Optimization Of Campus Unmanned Vehicle Path With Two Stages And Multiple Journeys

In the new retail context,online shopping has become one of the important ways of daily consumption,especially among college students.But in addition to making life easier for people,online shopping has also created some problems.First,the growth of express packages,especially during the holiday rush,can lead to an influx of deliveries from colleges and universities,which can easily "overload" campus delivery sites.Secondly,because of the particularity of school logistics,students are more concentrated in picking up express delivery,more likely to lead to crowded delivery points,long pick-up time,resulting in poor user experience at the end of the distribution chain.Third,the number of parcels at courier stations does not match the capacity of the site.If the number of couriers is increased,it will not only lead to an increase in the human cost of courier stations,but also cause management difficulties and bring security risks to campus safety and order.The introduction of driverless cars can alleviate some of these existing problems,and because of their powerful computing and information interaction capabilities,they can adjust road traffic information and customer needs in real time to bring about a better delivery experience for customers.Therefore,this paper presents a two-stage multi-trip vehicle path optimization model for campus unmanned vehicles based on the characteristics of campus logistics and unmanned vehicles.Firstly,the paper analyzes the theoretical basis of path optimization,vehicle path problem solving algorithm,campus express and unmanned vehicle distribution,and combs the related literature.Then,according to the characteristics of unmanned vehicles and campus logistics,a two-stage multi-trip vehicle path optimization model of unmanned vehicles is constructed.The model considers multi-trip vehicle paths because drones are more expensive.Choosing multiple trips minimizes the number of drones and makes them more efficient,reducing the financial pressure on Stagecoach to introduce drones.The two-stage modeling approach is designed to take advantage of the drone’s powerful environmental awareness and information interaction capabilities.When road information or customer needs change,the path is updated accordingly,making delivery on campus more dynamic,flexible,and efficient.Secondly,in view of the above modeling,a hybrid heuristic algorithm is designed to solve the problem.The genetic algorithm is improved on the selection and variation steps,and the variable neighborhood descent algorithm is combined to improve the genetic algorithm’s ability in local search and avoid the phenomenon of precocious convergence.Finally,an example is given to test the proposed model and algorithm.The traffic congestion,impassability and customer cancellations are simulated.The pre-planning stage and the real-time adjusted route are compared in the calculation.Then the performance of the hybrid heuristic algorithm is analyzed and compared with the traditional genetic algorithm.The results of the hybrid heuristic algorithm are better than the traditional genetic algorithm under different customer sizes.Sensitivity analysis of the number of unmanned vehicles is also carried out in this paper to provide a reference for the distribution center to select the appropriate number of unmanned vehicles scientifically.