【摘要】 针对物流配送过程,传统路径优化算法对交通拥堵、天气状况、环境因素不敏感,导致车辆在物流配送中效率低下、意外状况多的问题,提出基于深度学习的物流配送路径优化算法。首先构建基于自编码网络的模型,依据样本数据对模型进行训练,预测路段代价值ω,然后与城市干道网络相结合,建立带权重交通网络。最后,通过与禁忌搜索物流配送路径优化算法对比实验,该算法在实际配送中配送速度、物流成本与经济效益明显优于禁忌搜索路径优化算法。因此,该算法是物流配送路径优化的一种有效方法。更多还原

【Abstract】 Aiming at the problem of logistics distribution process, traditional path optimization algorithm is not sensitive to traffic congestion, weather conditions and environmental factors, which leads to the problem of low efficiency and unexpected situation in logistics distribution, pro-poses a logistics optimization algorithm based on depth learning. Firstly, constructs a model based on self-coding network. The model is trained according to the sample data, and predicts the expected value of the road segment. Then, the weighted traffic network is estab-lished by combining with the urban trunk network. Finally, through the comparison experiment with the tabu search logistics route optimization algorithm, the distribution speed, logistics cost and economic benefit of the algorithm are better than that of tabu search. Therefore, the algorithm is an effective way to optimize the logistics distribution path.更多还原