论文部分内容阅读
列车客座率作为列车运营效益的直接衡量指标和计算客票收入的重要基础,客座率的科学预测可为运输企业增开列车等运营决策提供重要参考。本文探讨利用BP神经网络方法研究列车客座率预测的适用性。首先分析列车客座利用率的12类影响因素,其中结合具体线路细化停站结构为n级,则影响因素变量为n+10个;其次以影响因素变量为自变量,列车客座率为因变量,生成多层感知器进行要素分析并预测客座率。结果表明MLP模型预测精度较高,可用于动车组列车开行方案稳定或小范围调整时期的列车客座率预测。
The train load factor as a direct measure of the operating efficiency of train and an important basis for calculating ticket revenue. The scientific prediction of load factor provides an important reference for transport enterprises to increase their operational decisions. This article explores the applicability of using BP neural network to study the prediction of train load factor. Firstly, 12 types of influential factors on the utilization of train passenger are analyzed. Among them, the number of influencing factors is n + 10 with the detailed structure of n-tier, and the influencing factors are independent variables and the passenger load factor is dependent variable , Generate multi-layer perceptrons for factor analysis and forecasting load factor. The results show that the MLP model has a high prediction accuracy and can be used to predict the passenger load factor during the stable or small adjustment period of the EMU train schedule.