摘要
根据时间序列原理将单列车的客运需求划分为具有趋势规律的确定需求和具有分布规律的随机需求两部分,通过差分自回归移动平均模型得到确定性需求的表达式和随机性需求的分布函数,进而构成单列车的客流量预测模型;将该预测模型嵌入随机票额分配模型中,通过求解客流量的期望值而将随机票额分配问题转化为确定票额分配问题,实现随机需求与随机票额分配有效结合的单列车票额分配,从而避免由于客运需求预测结果误差较大而使票额分配结果不能满足实际客运需求。通过算例验证了方法的可行性。
According to the principle of time series,passenger transport demand for single train was divided into two parts,namely,the trend-oriented determinate demand and the distribution-oriented stochastic demand.The expression for determinate demand and the distribution function for stochastic demand were obtained by autoregressive integrated moving average（ARIMA）model,and then the passenger flow forecast model for single train was constructed,which was embedded into the stochastic seat allotment model.The problem of stochastic seat allotment was transformed into the determined one by solving the expected value of passenger flow.The seat allotment for single train was realized by effectively integrating the stochastic demand with the stochastic seat allotment.Accordingly,that the seat allotment didn＇t meet the actual passenger transport demand caused by the big error in demand forecast could be avoided.Numerical example has validated the feasibility of the method.
出处
《中国铁道科学》
EI
CAS
CSCD
北大核心
2015年第2期96-102,共7页
China Railway Science
基金
中央高校基本科研业务费专项资金资助(2014JBZ008)
北京交通大学轨道交通控制与安全国家重点实验室自主课题(RCS2013ZZ001
RCS2009ZT009)
关键词
铁路客票
票额分配
客流预测
随机需求
时间序列
Railway passenger ticket Seat allotment Passenger flow forecast Stochastic demand Time series
