Improved Particle Swarm Algorithm to Solve the Vehicle Routing Problem

Vehicle routing problem is a NP hard problem. To solve the premature convergence problem of the particle swarm optimization, an improved particle swarm optimization method was proposed. In the first place, introducing the neighborhood topology, defining two new concepts lepton and hadron. Lepton are particles within the scope of neighborhood, which have weak interaction between each other, so they update speed and position according to the individual extreme value. Hardon are the local optimal particles, which through collide to produce strong reaction, so set the local optimal particles collision with the global optimal particle to update speed and position. Finally, when the algorithm step to stagnation, we take particle decay to increase the population diversity. Simulated experiments show that stagnation behavior of basic particle swarm optimization are avoided in improved particle swarm algorithm and a good ability of searching better solution in process of solving vehicle routing problem.