Research And Implementation Of Intelligent Logistics Vehicle Assembly And Distribution Platform Based On DGA Algorithm

Nowadays,the logistics industry is developing rapidly along with the new model of "Car Free Carrier" in logistics business.Although the existing distribution vehicle platform based on this business model solves the problem of matching goods with freight cars to a certain extent,only part of operations implement the expert system.The vast majority of operations that require decision-making processes are still done manually,which reflects that intelligent operation is not implemented.In this mode,the vehicle scheduling module is often unable to integrate existing freight information effectively and incapable of providing users with the best transportation routes,which results in a large amount of freight information being idle.The requirements of users for getting timely information and realizing cost minimization are not satisfied.As a result,logistics enterprises are in great need of providing users with the optimal transportation routes by constructing an intelligent distribution platform for logistics vehicles.This thesis first abstracts the scheduling problem into a typical NP problem,i.e.VRP(Vehicle Routing Problem).Then,an improved genetic algorithm is used to find the solution.Based on this algorithm,the intelligent vehicle scheduling module in the intelligent logistics vehicle distribution platform is designed and implemented.On the one hand,the platform uses the intelligent algorithm to solve the problems of auto-matching of vehicles and goods,and also to solve the problems of designing optimal routes.The algorithm can realize the automatic operation function and does not need administrators to operate,which realizes "artificial intelligence";On the other hand,the platform meets the needs of users to reduce transportation costs and save freight time by providing reasonable freight information.The main work of this thesis includes:(1)When using the genetic algorithm to solve VRP,problems like slow solution rate,low solution accuracy and local optimum are easy to occur.This thesis proposed an improved genetic algorithm—the multi-group hybrid improved vehicle scheduling algorithm(Dispatch Genetic Algorithm).The algorithm first uses natural numbercoding to reduce the coding length,and then uses three co-evolutionary strategies to optimize the initial population,and finally uses the simulated annealing mechanism to jump out of the local optimal solution.The results of the simulation tests show that the algorithm has higher searching ability and accuracy when solving large-scale VRP.It is proved that the algorithm averts the deficiency of the genetic algorithm when solving VRP.(2)Based on the original logistics platform,the intelligent logistics vehicle collection and distribution platform is designed and implemented with the intelligent vehicle dispatch module as the core.The user management module and the financial management module are regarded as the auxiliaries.The overall architecture of the platform adopts the B/S architecture model.At the same time,the SAAS model is employed to provide cloud computing services to customers.On the one hand,the platform's intelligent vehicle scheduling module uses the DGA to calculate the optimal path to achieve automated vehicle deployment capabilities.On the other hand,the financial management module and the price model of the user management module can provide customers with more comprehensive information and meet different freight requirements of customers.