Research And Development Of Optimization On Vehicle Routing Problems In Dense Logistics Alliance

Optimization on vehicle routing problem in dense logistics alliance is kind of new and more complex problem in reality comparing with traditional vehicle routing problems, and this paper proposes a new improved ant colony algorithm to solve it based on lots of research on vehicle routing modules and swarm intelligence.Swarm intelligence algorithm is heuristic searching algorithm, which has good performance on NP-Hard and combinational optimization problems, especially on ones with large scale or high complexity based on lots of research work. Now it abstracts many academicians and experts to continuously make research, improvement and optimization on it. As representative of swarm intelligence algorithm, ant colony algorithm is a simulated evolutionary algorithm, which can be commonly used and offer very good robustness, but it also has some shortcomings, such as taking longer computing time and easy to fall into local best. To improve this, a new algorithm, optimal parallel ant colony algorithm with particle swarm features, is proposed in this paper, with introduction of particle character, crossover and mutation mechanism.Besides, this paper designs architecture of dense logistics alliance platform and implements vehicle scheduling module in code level. Then it applies its proposed algorithm and some other algorithms to the case of alliance vehicle routing problem, and the result shows that new algorithm gives a better solution in decreasing computing time and avoiding early maturing phenomenon than the others.