Improved Particle Swarm Optimization Algorithm For Vehicle Routing Problem Research

The vehicle routing problem in logistics, combinatorial optimizationand management field are influenced by many wide attention of scholars,many practical theory method has been using to the actual cases, thePeople's Daily life production have played a major influence. The vehiclerouting problem as the core of the logistics distribution, and peopleproduction life. Therefore, developed a reasonable vehicle routingalgorithm, reasonable distribution of goods, in order to obtain low costinvestment, efficient service returns, not only is a challenge also has acertain scientific research value.This paper, based on the current research status of vehicle routingproblems, the main of the vehicle routing model and various algorithm arereviewed. The model of vehicle routing problem from simple staticvehicle routing problem (VRP, vehicle routing problem), vehicle routingproblem with time Windows (VRPTW, vehicle routing problem with timewindow), take delivery of the vehicle routing problem (VRPPD, capacityvehicle routing problem with time window) model, dynamic, random anddevelopment to the more complex fuzzy vehicle routing problem, solvingalgorithm from the initial also exact algorithm development to heuristicalgorithm, intelligent optimization algorithm, etc. In view of the currentvehicle routing problem is a hot issue VRPTW model, this paper mainly through the traditional Particle Swarm Optimization algorithm (PSO,Particle Swarm Optimization) was improved, and used for solving thevehicle routing problem with time Windows. Particle swarm optimizationalgorithm is a kind of effective optimum algorithm, this algorithm basedon the thought of bird flock foraging produces in the process of the path,the direction and speed of the best route, because the problem with anatural VRP combination, therefore, to be able to use its convenient tosolve VRP problem.The advantages of the algorithm is algorithm few parametersinvolved (mainly speed and direction), and the solution quality is highercharacteristic. However, and almost all of the heuristic algorithm, thetraditional particle swarm optimization algorithm on the global searchability is insufficient, easy to cause the algorithm into the local optimumin advance. Therefore, this paper puts forward the improved particleswarm optimization algorithm for traditional particle swarm optimizationalgorithm in the following measures:1, Reconstruction of the velocity formula and position the formula oftraditional PSO algorithm;2, Particle expression and particles of the assessment of the degreeof health.3, Introducing the son and strengthen exchange swap sequence ofglobal space search. 4, Introducing variation mechanism improvement population quality.The article will improve with time solving algorithm is used forwindow and load limit vehicle routing problem, more close to the real lifeof the logistics management problem, ask the maximum satisfy customerservice demand, and meet all the constraints, under the premise of lookingfor the optimal route; Particles in the optimization process play animportant role, for it directly determine the next generation of populationhealth level of quality, so the article into the evaluation mechanism toensure the continuation of the excellent individual; For the solution andorder exchange often bring the result that expect is less than, through theexchange of the son, and get the new swap sequence solution sequence,access to new combinatorial optimization. Finally, through theSolomo₁00standard test data set to the improvement of the particleswarm optimization algorithm simulation experiment, the experimentalresults show that the improved algorithm in the search for the optimalpath path process, and other similar algorithm, the test data set theiteration times less, and running more short time. Therefore, the paperputs forward the improved algorithm is effective.In this paper, according to the reality and the field research hot spot,to a different vehicle routing problem model the description of the system,and combined with improved particle swarm optimization algorithm isused to make the vehicle routing problem has a new solution, for modern life in the transportation industry, logistics and distribution in the fieldssuch as the optimal vehicle routing scheme of the planning and designprovides reference and reference...