基于Fisher聚类的公交客流量时间序列预测及对比
|
摘要
公交客流量预测是城市公共交通管理的基础,科学的客流量预测能够为公交系统管理和路线调整提供可靠依据。考虑到公交客流量的波动差异性以及预测的复杂性,首先利用Fisher算法对原始数据聚类,并依时段划分为六种类型;然后选择自回归差分移动平均模型以及季节性自回归差分移动平均模型两种方法开展公交客流量的预测,并以广州市公交客流量数据进行实证分析,最后计算两种模型的平均绝对误差和平均绝对百分比误差,对比分析基于聚类数据的两模型预测效果的优劣。结果发现:基于Fisher聚类数据,季节性自回归差分移动平均模型的预测效果较好,且比数据未聚类前对应模型预测的效果更优。
The prediction of bus passenger flow is the basis of urban public transport management.Scientific prediction of passenger flow can provide a reliable basis for the management of bus systems and route adjustment.Considering the dissimilarity of fluctuation and the complexity of prediction of bus passenger flow,the original data was firstly clustered by using Fisher's algorithm and was divided into six types according to time interval.Then,the autoregressive differential moving average model and seasonal autoregressive differential moving average model were selected to make predictions of public traffic passenger flow,and empirical analysis was conducted based on Guangzhou public traffic flow data.Finally,the average absolute errors and the average absolute percentage errors of the two models were calculated,and the prediction effect of the two models based on clustering data were compared and analyzed.The results show that based on the Fisher clustering data,the seasonal autoregressive differential moving average model has a better prediction effect,which is much better than that with data without clustering.
The prediction of bus passenger flow is the basis of urban public transport management.Scientific prediction of passenger flow can provide a reliable basis for the management of bus systems and route adjustment.Considering the dissimilarity of fluctuation and the complexity of prediction of bus passenger flow,the original data was firstly clustered by using Fisher's algorithm and was divided into six types according to time interval.Then,the autoregressive differential moving average model and seasonal autoregressive differential moving average model were selected to make predictions of public traffic passenger flow,and empirical analysis was conducted based on Guangzhou public traffic flow data.Finally,the average absolute errors and the average absolute percentage errors of the two models were calculated,and the prediction effect of the two models based on clustering data were compared and analyzed.The results show that based on the Fisher clustering data,the seasonal autoregressive differential moving average model has a better prediction effect,which is much better than that with data without clustering.
引文
[1]LIU X,HUAN X.Prediction of passenger flow at Sanya Airport based on combined methods[J].Data Science,2017,727:729-740.
[2]LIU X,HUANG X.Prediction for passenger flow at the airport based on different models[J].Parallel Architecture,Algorithm and Programming,2017,729:25-40.
[3]KUMAR S V,VANAJAKSHI L.Short-term traffic flow prediction using seasonal ARIMA model with limited input data[J].European Transport Research Review,2015,7(3):1-9.
[4]成诚,杜豫川,刘新.考虑节假日效应的交通枢纽客流量预测模型[J].交通运输系统工程与信息,2015,15(5):202-207.CHENG Cheng,DU Yuchuan,LIU Xin.A passenger volume prediction model of transportation hub considering holiday effects[J].Journal of Transportation Systems Engineering and Information Technology,2015,15(5):202-207.
[5]蔡昌俊,姚恩建,王梅英,等.基于乘积ARIMA模型的城市轨道交通进出站客流量预测[J].北京交通大学学报,2014,38(2):135-140.CAI Changjun,YAO Enjian,WANG Meiying,et al.Prediction of urban railway station’s entrance and exit passenger flow based on multiply ARIMA model[J].Journal of Beijing Jiaotong University,2014,38(2):135-140.
[6]王莹,韩宝明,张琦,等.基于SARIMA模型的北京地铁进站客流量预测[J].交通运输系统工程与信息,2015,15(6):205-211.WANG Ying,HAN Baoming,ZHANG Qi,et al.Forecasting of entering passenger flow volume in Beijing subway based on SARIMA Model[J].Journal of Transportation Systems Engineering and Information Technology,2015,15(6):205-211.
[7]马佳羽,韩兆洲.复杂季节时间序列模型研究[J].统计与决策,2017(6):27-30.MA Jiayu,HAN Zhaozhou.Time series data with complex seasonal periods studies[J].Statistics & Decision,2017(6):27-30.
[8]周辉宇.基于大数据规则挖掘的交通拥堵治理研究[J].统计与信息论坛,2017,32(5):96-101.ZHOU Huiyu.Study on traffic congestion based on big data rule mining[J].Statistics & Information Forum,2017,32(5):96-101.
[9]徐晓伟,杜一,周园春.基于多源出行数据的居民行为模式分析方法[J].计算机应用,2017,37(8):2362-2367.XU Xiaowei,DU Yi,ZHOU Yuanchun.Resident behavior model analysis method based on multisource travel data[J].Journal of Computer Applications,2017,37(8):2362-2367.
[10]闫伟,刘云岗,王桂华,等.基于数据挖掘的交通流预测模型[J].系统工程理论与实践,2010,30(7):1320-1325.YAN Wei,LIU Yungang,WANG Guihua,et al.Data mining using in a novel traffic flow forecasting model[J].Systems Engineering:Theory & Practice,2010,30(7):1320-1325.
[11]李林波,姜屿,王静,等.基于数据融合的公交客流规模测算方法[J].城市交通,2016,14(1):43-50.LI Linbo,JIANG Yu,WANG Jing,et al.Passenger volume estimation based on data fusion[J].Urban Transport of China,2016,14(1):43-50.
[12]刘自强,王效岳,白如江.基于时间序列模型的研究热点分析预测方法研究[J].情报理论与实践,2016,39(5):27-33.LIU Ziqiang,WANG Xiaoyue,BAI Rujiang.Analysis and forecasting method of research hotspots based on time series model[J].Information Studies:Theory & Application,2016,39(5):27-33.
[13]徐潇.基于客流预测的公交调度优化研究[D].郑州:郑州大学,2017.
[14]刘建军.几种时间序列模型在客流量预测上的比较[D].武汉:武汉邮电科学研究院,2015.
[15]乔占俊.基于季节调整的区域中长期负荷预测[J].统计与决策,2014(21):83-85.QIAO Zhanjun.Regional mid-and-long term load forecast based on seasonal adjustment[J].Statistics & Decision,2014(21):83-85.
[2]LIU X,HUANG X.Prediction for passenger flow at the airport based on different models[J].Parallel Architecture,Algorithm and Programming,2017,729:25-40.
[3]KUMAR S V,VANAJAKSHI L.Short-term traffic flow prediction using seasonal ARIMA model with limited input data[J].European Transport Research Review,2015,7(3):1-9.
[4]成诚,杜豫川,刘新.考虑节假日效应的交通枢纽客流量预测模型[J].交通运输系统工程与信息,2015,15(5):202-207.CHENG Cheng,DU Yuchuan,LIU Xin.A passenger volume prediction model of transportation hub considering holiday effects[J].Journal of Transportation Systems Engineering and Information Technology,2015,15(5):202-207.
[5]蔡昌俊,姚恩建,王梅英,等.基于乘积ARIMA模型的城市轨道交通进出站客流量预测[J].北京交通大学学报,2014,38(2):135-140.CAI Changjun,YAO Enjian,WANG Meiying,et al.Prediction of urban railway station’s entrance and exit passenger flow based on multiply ARIMA model[J].Journal of Beijing Jiaotong University,2014,38(2):135-140.
[6]王莹,韩宝明,张琦,等.基于SARIMA模型的北京地铁进站客流量预测[J].交通运输系统工程与信息,2015,15(6):205-211.WANG Ying,HAN Baoming,ZHANG Qi,et al.Forecasting of entering passenger flow volume in Beijing subway based on SARIMA Model[J].Journal of Transportation Systems Engineering and Information Technology,2015,15(6):205-211.
[7]马佳羽,韩兆洲.复杂季节时间序列模型研究[J].统计与决策,2017(6):27-30.MA Jiayu,HAN Zhaozhou.Time series data with complex seasonal periods studies[J].Statistics & Decision,2017(6):27-30.
[8]周辉宇.基于大数据规则挖掘的交通拥堵治理研究[J].统计与信息论坛,2017,32(5):96-101.ZHOU Huiyu.Study on traffic congestion based on big data rule mining[J].Statistics & Information Forum,2017,32(5):96-101.
[9]徐晓伟,杜一,周园春.基于多源出行数据的居民行为模式分析方法[J].计算机应用,2017,37(8):2362-2367.XU Xiaowei,DU Yi,ZHOU Yuanchun.Resident behavior model analysis method based on multisource travel data[J].Journal of Computer Applications,2017,37(8):2362-2367.
[10]闫伟,刘云岗,王桂华,等.基于数据挖掘的交通流预测模型[J].系统工程理论与实践,2010,30(7):1320-1325.YAN Wei,LIU Yungang,WANG Guihua,et al.Data mining using in a novel traffic flow forecasting model[J].Systems Engineering:Theory & Practice,2010,30(7):1320-1325.
[11]李林波,姜屿,王静,等.基于数据融合的公交客流规模测算方法[J].城市交通,2016,14(1):43-50.LI Linbo,JIANG Yu,WANG Jing,et al.Passenger volume estimation based on data fusion[J].Urban Transport of China,2016,14(1):43-50.
[12]刘自强,王效岳,白如江.基于时间序列模型的研究热点分析预测方法研究[J].情报理论与实践,2016,39(5):27-33.LIU Ziqiang,WANG Xiaoyue,BAI Rujiang.Analysis and forecasting method of research hotspots based on time series model[J].Information Studies:Theory & Application,2016,39(5):27-33.
[13]徐潇.基于客流预测的公交调度优化研究[D].郑州:郑州大学,2017.
[14]刘建军.几种时间序列模型在客流量预测上的比较[D].武汉:武汉邮电科学研究院,2015.
[15]乔占俊.基于季节调整的区域中长期负荷预测[J].统计与决策,2014(21):83-85.QIAO Zhanjun.Regional mid-and-long term load forecast based on seasonal adjustment[J].Statistics & Decision,2014(21):83-85.