New Heuristic Algorithms For Traveling Salesman Problem And Its Derivational Problem

Traveling salesman problem is one of the hottest topics in the combinatorial optimization problem of operational research area,and it has both theoretical and practical significance to study this problem.In addition,heuristic algorithms can solve traveling salesman problem effectively,and thus heuristic algorithms are also research hotspots in recent years.Traveling salesman problem can be solved more efficiently and accurately through improving traditional heuristic algorithms.Moreover,as an extension of traveling salesman problem,inbound logistic problem is reality-based problem,and can be solved to support and guide the future regional planning.Firstly,this paper gave a brief introduction on the backgrounds of traveling salesman problems and its respective heuristic algorithms and their academic significance,introduced the background and significance of inbound logistic problem,reviewed the research status of these problems and algorithms at home and broad,and summarized shortcomings of the research status and the innovation points of this paper.Then,based on the shortcomings,this paper made an introduction for traveling salesman problem,inbound logistic problem,and the respective heuristic algorithms.Then,a hybrid artificial bee colony algorithm based on simulated annealing and tabu search algorithm was proposed to solve traveling salesman problem.And the crossover operation of genetic algorithm and dynamic neighborhood 2-opt were creatively introduced into this algorithm as local search process in employee bee phase and in onlooker bee phase.Through running benchmark instances in TSPLIB,the performance of this algorithm was tested.And compared to colony-intelligence based algorithms,this algorithm was proved to have a better performance.Moreover,according to national physical situation,a fact-based m-n inbound logistic problem,derive from traveling salesman problem,was proposed,the nonlinear programming model was also established,and solved by a heuristic algorithm.And the result demonstrated that reasonable programming for m-n inbound logistics can improve the loading rate efficiently,decrease the number of cars,and thus decrease logistic cost.Finally,summarized this paper,and the future research direction was given.