高速公路通行费预测研究及应用

发布时间：2018-01-14 07:09

本文关键词：高速公路通行费预测研究及应用　出处：《长安大学》2015年硕士论文　论文类型：学位论文

【摘要】：随着我国高速公路的快速发展,高速公路通行费预测逐渐成为高速公路运营管理研究中的核心问题之一。良好的通行费预测管理方式是高速公路收费政策正常实施的保障,同时也是高速公路路网建设和发展的重要环节。高速公路通行费预测不仅体现了高速公路收费站的运营水平和管理现状,同时也影响着高速公路征费费率调整、通行费监管以及运营权转让等工作,能为高速公路的决策者在宏观政策的制定和实施中提供重要的决策依据。本文分析了高速公路收费站通行费数据特征,阐述了现有高速公路通行费预测理论和思路,并针对其不足,以时间序列理论和灰色理论为基础,探索和研究了高速公路收费站通行费短期预测和中长期预测的预测方法。首先,文中研究了ARIMA预测模型在高速公路收费站通行费预测中的应用,预测结果表明ARIMA预测模型结合通行费数据特征修正后,能够满足高速公路收费站通行费短期预测要求。其次,针对ARIMA预测模型在高速公路收费站通行费中长期预测中的局限性,建立优化的灰色GM(1,1)预测模型,并通过最优加权系数的方法构建新的组合预测模型,预测结果表明相对于单项预测模型,组合预测模型在高速公路收费站通行费预测中具有更高精度,拟合效果更好,能够展现通行费的发展规律。最后,基于高速公路收费站通行费预测研究中的预测模型,采用Java、Javascript、FineReport报表技术等完成了高速公路收费站通行费预测管理系统的设计与实现。研究结果表明,相对以往的预测方法,结合高速公路收费站通行费数据特征的ARIMA预测模型和组合预测模型的预测精度得到了明显的提高,预测误差低于5%,满足了高速公路通行费预测需求。最后将构建的预测模型应用于高速公路收费站通行费预测管理系统的设计和实现中,使高速公路运营管理部门收费管理工作更加规范、高效,也验证了各类预测模型在高速公路收费站通行费预测中的可行性和实用性。
[Abstract]:With the rapid development of highway in China, the prediction of expressway toll has gradually become one of the core issues of the operation and management of expressway toll management. Good prediction mode is the normal implementation of highway toll policy, but also is an important part of highway construction and development. The prediction of highway tolls not only reflects the operation level of the highway toll station and the management of the status quo, but also affects the highway toll levy rate adjustment, regulation and operation rights transfer, can provide an important decision basis for highway decision-makers in the formulation and implementation of macroeconomic policy. This paper analyzes the highway toll toll data characteristics, this paper expounds the existing highway tolls and prediction theory ideas, and to overcome the shortcomings in time series theory and grey theory, exploration The prediction method of cable on the highway toll toll prediction and long-term prediction. Firstly, this paper studies the application of ARIMA model in highway toll station in the prediction, the predicted results show that combining the ARIMA prediction model of traffic correction fee data characteristics, can meet the requirements of the toll highway toll prediction station. Secondly, aiming at the limitations of ARIMA forecast the long term forecast of tolls on the highway toll station in the optimization model, establish the grey GM (1,1) prediction model, and through the method of the optimal weighting coefficients to construct a new combination forecasting model, the prediction results show that compared with the single forecasting model, combination forecasting model has higher accuracy in the toll station highway toll prediction, the fitting effect is better, the development of law can show tolls. Finally, the highway toll station traffic prediction based on cost Study on the prediction model, using Java, Javascript, FineReport report technology to complete the highway toll toll prediction management system design and implementation. The results of the study show that, relative to the previous prediction method, the prediction accuracy of the highway toll station ARIMA prediction model and combined forecasting model of toll data characteristics were improved obviously the prediction error is less than 5%, to meet the demand forecast of highway tolls. Application of the final predictive model will build on the highway toll station toll prediction and management system, so that the charge management of expressway management departments more standardized, efficient, also verified the feasibility and practicability of prediction model in highway toll station the charge in the prediction.

【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491

【相似文献】