基于时间序列的铁路客流量预测及票额优化配置  

Railway Passenger Flow Forecast and Ticket Allocation Optimization Based on Time Series

作　　者：牛坤 龙慧云[1] 于雪涛[3,4] 刘满义 NIU Kun;LONG Hui-yun;YU Xue-tao;LIU Man-yi(College of Computer Science and Technology,Guizhou University,Guiyang 550000,China;State Key Laboratory of Public Big Data,Guizhou University,Guiyang 550000,China;School of Traffic and Transportation,Shijiazhuang Tiedao University,Shijiazhuang 050043,China;Key Laboratory of Traffic Safety and Control of Hebei Province,Shijiazhuang 050043,China)

机构地区：[1]贵州大学计算机科学与技术学院,贵阳550000 [2]贵州大学公共大数据国家重点实验室,贵阳550000 [3]石家庄铁道大学交通运输学院,石家庄050043 [4]河北省交通安全与控制重点实验室,石家庄050043

出　　处：《科学技术与工程》2020年第24期9937-9942,共6页Science Technology and Engineering

基　　金：贵州省科技计划(黔科合重大专项字[2018]3007,黔科合重大专项字[2018]3001,黔科合支撑[2018]2162)。

摘　　要：铁路客流量受多因素影响,其时序特征明显,因此,基于平稳时间序列构建客流数据预测模型及区间票额分配模型,有利于掌握客流动态变化,改善铁路运营压力。实现特征数据抽取系统开发,进行累加、循环、筛选算法等数据预处理;运用多因子方差分析评价多种因素的显著相关性影响,通过自回归移动平均模型(auto-regressive moving average model,ARMA)进行短时旅客客流量预测及其优化建模,结果表明:该模型拟合效果良好,预测精度高。此外,基于线性规划实现客座率最大化的区间票额分配优化模型,并通过真实数据验证其方案是可行、有效的,研究结果对于指导铁路票额分配具有较好的参考价值。Railway passenger flow is affected by many factors,and its timing characteristics are obvious.Therefore,based on the stationary time series,passenger flow data prediction model and interval ticket distribution model were constructed.It was beneficial to grasp the dynamic changes of passenger flow and improved the railway operation pressure.The characteristic data extraction system was realized,and preprocesses data such as accumulation,cycle and screening algorithm was conducted then.The multi-factor analysis of variance was used to evaluate the significant correlation effect of various factors.The short-time passenger flow was predicted by auto-regressive moving average(ARMA)model,and the optimization model was constructed.The results show that this model has good fi-tting effect and high prediction accuracy.In addition,the optimization model of interval ticket allocation based on linear programming to maximize passenger load factor,and it is verified by real data,the scheme is feasible and effective,which has a good reference value for guiding railway ticket allocation.

关 键 词：时间序列 建模预测 优化 客流量 票额分配 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]