Research On The Vehicle Routing Problem Based On Swarm Intelligence Optimization

The vehicle routing problem is a classic combinatorial optimization problem. It hasnot only got the extensive and in-depth research, but also made many achievements. Inrecent years, with the development of intelligent optimization algorithm, this problemhas been caught the attention again. This paper mainly discusses two kinds of swarmintelligence optimization algorithms to solve the vehicle routing problem.On the analysis of the vehicle routing problem, this paper proposes an ant colonyalgorithm. Mainly through encoding method, the separation and combination of the pathand fitness function, this problem is solved. By analyzing the shortage of the ACO, anew ACO with inversion operator is presented. This operator ensures the populationdiversity and improves the optimization performance of the algorithm. The experimentaldata show that those two algorithms are effective and the inversion operator plays itsrole.By analyzing the optimization mechanism of biogeography optimization algorithm,a hybrid algorithm is proposed. In order to overcome prematurity of ACO, the hybridalgorithm incorporates the search performance of ACO and utilization performance ofBBO. And on this basis, an adaptive parameter strategy is proposed to ensure that thetwo mechanisms can make full cooperation. The results show the effectiveness of theproposed hybrid algorithm.After a detailed analysis of the hybrid algorithm, a new algorithm is proposed thatcan solve continuous problems. This algorithm is based on differential evolution andbiogeography-based optimization. By combining adaptive parameter and ’pbest’mutation strategy, the problem of optimal control is solved. Experimental results showthat without loss original ability the proposed algorithm can get a satisfactory solution,and it has higher efficiency in terms of search accuracy and speed.