基于手机数据的地铁线网清分模型验证与优化

如何高效准确地验证并优化地铁线网票款清分清算模型是地铁行业一直以来的研究重点。地铁票款清分清算模型大多是基于概率分布的出行路径选择模型。该类模型应用出行时间对应选择概率的正态分布得到乘客对不同出行路径的选择概率,结合不同出行路径上各个线路运营商的运营里程比,即可计算得到所谓的"清分因子"。为验证现有清分模型的准确性,提出使用手机数据重建乘客出行轨迹的方法。通过在地铁线网换乘节点安装无线探测装置,获取乘客智能手机唯一的MAC地址,将乘客出现的时间、地点信息实时发送到数据处理后台,按照时间先后排列捕获乘客出现的地点,以此重建出行轨迹。目前只有开启Wi Fi功能的智能手机可以被捕获。通过这一方法,不仅可以验证地铁票款清分清算模型,还可以进一步对模型参数进行修正。本研究在杭州市地铁线网中进行实验,通过重建当地乘客实际出行路径,对票款清分清算模型进行验证和修正。

Verifying and Optimizing Metro Fare Clearing Model Based on Cell Phone Data

How to verify and optimize the metro revenue clearing model efficiently and accurately is a research focus metro operations.Usually,metro revenue clearing models are based on probability distributions.This model uses the route choice probabilities,which is computed based on normal distribution of travel time,and operating mileage proportions of each metro line operator to calculate the revenue clearing proportions for all the operators of the metro network.To verify the accuracy of the revenue clearing proportions,we propose a travel route reconstruction approach based on cell phone data acquisition technique.With wireless AP (access point) sensors installed at transfer stations,the unique MAC (medium access control) address of the smart phone with with function turned on will be recorded and transmitted to a data analysis platform.After matching the MAC address information WiFi time and location,the smart phone user's travel route can be reconstructed.Then,the parameters in fare clearing model can be verified and optimized according to the travel route choice probabilities.We implement the proposed method in Hangzhou metro network and further modify the Hangzhou metro revenue clearing model.