基于混合神经网络的车辆路径模型研究

为有效解决现代物流配送中的车辆路径问题,发挥BP神经网络在解决分类问题和Hopfield神经网络在解决组合优化问题中的优势,依据"分而治之"策略提出了基于混合神经网络的优化模型。通过BP神经网络对一个配送中心范围内的多个配送点进行区域划分,在各子区域内使用Hopfield神经网络求得最优配送路径,从而得到质量较高的解和较快的收敛速度。基于Matlab的仿真实验结果表明,与传统的爬山算法、遗传算法相比,该模型能够获得性能更好的全局最优解。

Research on vehicle routing model based on hybrid neural networks

To solve vehicle routing problem in the modern logistics distribution effectively,and play the respective advantages of BP neural network and Hopfield neural network in classification and combinatorial optimization,according to divide-and-conquer strategy,an optimization model based on hybrid neural networks is proposed.After the regions of distribution points within a distribution center are classified by BP neural network,Hopfield neural network is used to seek an optimal path within each region so that a...