基于指数平滑法优化马尔科夫模型的铁路客运量预测

Railway Passenger Traffic Volume Forecast Based on Optimized Markov Model Based on Exponential Smoothing

铁路客运量预测是编制客运发展规划、统筹客运设备资源配置、提高客运服务水平的关键。为提高铁路客运量预测精度,针对马尔科夫模型“无后效性”的特点,基于指数平滑法的基本原理和思想从“状态概率向量计算方法”和“预测值的马尔科夫修正公式”两方面对马尔科夫模型进行优化。通过霍尔特线性趋势指数平滑法预测上海市铁路客运量,运用基于指数平滑法优化的马尔科夫模型和传统马尔科夫模型对预测值进行修正以进行对比验证。结果表明,马尔科夫模型能够提高指数平滑法预测结果的精确度,且指数平滑法优化马尔科夫模型的预测修正值的平均误差较传统马尔科夫模型减小了0.738%,具有更好的适用性和精确性。

Railway passenger transport volume forecasting is the key to compiling passenger transport development plans,coordinating the allocation of passenger transport equipment resources,and improving passenger transport service levels.In order to improve the prediction accuracy of railway passenger traffic,in view of the"no aftereffect"characteristic of the Markov model,based on the basic principles and ideas of the exponential smoothing method,the Markov model is optimized from two aspects:the calculation method of the state probability vector and the Markov correction formula of the predicted value.Finally,the Holt linear trend exponential smoothing method is used to predict Shanghai railway passenger traffic,and the Markov model optimized by the exponential smoothing method and the traditional Markov model are used to modify the predicted values for comparison and verification.The results show that the Markov model can improve the accuracy of the prediction results of the exponential smoothing method,and the average error of the forecast correction value of the optimized Markov model of the exponential smoothing method is reduced by 0.738%compared with the traditional Markov model,which has better performance Applicability and accuracy.