摘要
分析了站间距离、上下车乘客数、停车延误等因素与公交车行程时间存在的线性映射关系,推导并建立了多元线性回归(MLR)模型,利用大连市21路公交车的数据,对模型进行了标定和检验,并对线性模型的误差来源进行了详细的分析和探究.重点探讨了行驶车速-行程时间线性相关性、乘客数-停站时间线性相关性、延误和行程时间线性相关性,解释了MLR标定的系数的含义,并阐述了导致MLR模型产生误差的来源,分析了非线性因素的影响.研究结果表明多元MLR模型能够解释影响因素与行程时间之间的线性因果关系,可以用来估计和预测公交车行程时间.
The linear relationship between bus travel time are analyzed and the influencing factors such as stop distance, the number of passengers getting on or off and stop delay to deduce and build a multi-linear regres- sion （MLR） model which was calibrated and tested by using the data of Bus No. 21 in Dalian. The error source of the linear model is discussed in detail. The analysis is focused on the linear relationship between speed and travel time （ exclusing dwell time）, the number of pessengers and dwell time, delay and travel time. Thus the exact meaning of the MLR model coefficien can be explained, and the error source especially the effects of the nolinear factors while building the MLR model can be traced. The result shows that the MLR model can account for a portion of the linear relationship between the influencing factors and travel time, and it can be used to estimate or forecast bus travel time.
出处
《大连交通大学学报》
CAS
2015年第2期1-5,共5页
Journal of Dalian Jiaotong University
基金
辽宁省教育厅科学研究计划资助项目(L2013190)
关键词
智能交通
行程时间
多元线性回归
公共交通
预测
误差分析
intelligent transportation
travel time
multi-linear regression
public transport
forecasting
error analysis
