居民公交出行链重复性量化分析及其出行规律研究

作为城市交通的枢纽，公共交通系统承载了大量的居民出行。自动数据采集系统收集的IC卡数据包含了大量的乘客出行信息，通过这些数据可分析居民公交出行规律，进而优化公交服务。引入信息熵及熵率对居民公交出行链重复性进行量化，研究了基于量化指标分析居民公交出行规律的方法。通过出行地点状态标定，将乘客的出行链转化为离散的出行序列；利用信息熵和熵率对出行序列进行量化分析，得到出行重复性与量化指标的关系，即出行序列的信息熵越大，熵率越小，该乘客出行重复性越高，出行规律越强。基于重复性量化处理，以石家庄公交智能卡乘客出行数据为例，分别从群体和个人这2个方面对公交乘客的出行规律进行分析。结果表明，出行链重复性量化指标可以对出行规律的强弱进行直观判断。当乘客出行规律不明显，但信息熵高于样本均值2.53 bits、熵率低于样本均值1.13 bits/事件时，可通过进一步分析挖掘出乘客潜在出行规律。

A Quantitative Analysis on Repeatability of Residents' Bus Trip Chain and Travel Regularity

As a hub system of urban transportation, public transportation carries a large number of residents'travel.The IC card data collected by the automatic data collection system contains a large number of passenger-travel information, which can be analyzed to optimize the public transport service.The information entropy and entropy rate are introduced to quantify the repeatability of the trip chain of public transport, with the method of analyzing the law of public transport travel based on the quantitative index studied.The trip chain of passengers is transformed into a discrete travel sequence through the state calibration of the travel place.Information entropy and the entropy rate are used to quantitatively analyze the travel sequence, thus obtaining the relationship between travel repeatability and quantitative index.In other words, the higher the information entropy of the travel sequence, the lower the entropy rate, the higher the passengers'travel repeatability, and the stronger the travel rule.Based on the repeatability of quantitative processing, the work takes the travel data of smart card passengers in Shijiazhuang as a case study to analyze the travel rules of bus passengers from group and individual.The results show that the quantification index of trip-chain repeatability can intuitively judge the strength of travel rules.When the information entropy is higher than the sample mean(2.53 bits)and the entropy rate is lower than the sample mean(1.13 bits/event)with the unobvious travel rules of passengers, the potential travel rules of passengers can be mined through further analysis.