［1］ Abkowitz M，Engelstein I． Empirical methods for improving transitscheduling ［C］． World Conference on Transport Research: Researchfor Policies in a Changing World． Hambwrg，Germany，1983: 884-885．
[2]
［2］ Chien S，Ding Y Q，Chien H W． Dynamic bus arrival time predictionwith artificial neural networks ［J］． Journal of TransportationEngineering，2002，128(5): 429-438．
Sun Ximei，Yang Zhaosheng． The model of real-time dynamic traveltime prediction between urban public transportation stop［J］． Journalof Jilin University: Engineering and Technology，2002，32(1):47-51．
Hang Mingsheng，Yang Xiaoguang，Peng Guoxiong． Study of predictedtravel time in urban expressway based on Kalman filter［J］．Journal of Tongji University，2002，30(9): 1068-1072．
Zhang Xiongxi． Application of wireless senor network to intelligenttransportation system of city public bus［J］． Journal of ChongqingJiaotong University: Natural Science，2009，28(6): 1064-1066．
［8］ Chien S，Ding Yuqing． Applications of Artificial Neural Networks inPredicting of Transit Arrival Times［C］/ /Annual Conference of ITSAmerica Proceedings． Washington，D． C． : TRB，1999: 859-870．
[12]
［9］ Kalaputapu R，Demetsky M J． Modeling schedule deviations of busesusing automatic vehicle-location data and artificial neural networks［J］． Transportation Research Record，1995，1497: 44-52．
