Solving vehicle routing problems by maximum neuron model

In this paper, we propose a new clustering method for the first phase of a two-phase method of the vehicle routing problems (VRPs) and the traveling salesman problems (TSPs). In the first phase, the customers are grouped as several delivery areas for vehicle by using maximum neuron model. In the second phase, the TSPs for each areas are solved by using elastic net model proposed by Andrew et al. Conventional maximum neuron model proposed by Takefuji et al. is not suitable for these continuous problems. But by including a self-organization rule to this model, the solution quality is improved. Our simulation results show that maximum neuron model can achieve to obtain better solutions than other methods for some kinds of problems in VRPs and TSPs.