Dealing with Error Recovery in Traffic Flow Prediction Using Bayesian Networks Based on License Plate Scanning Data

This paper deals with the error recovery problem when scanned license plate data are used to predict traffic flows. The aim is to reduce the effects of errors owing to lost plates or mistaken transcription, to improve estimation results. To this end, a method is given and discussed for traffic flow prediction using plate scanning data and taking into account possible errors in plate number recognition. The proposed method uses Bayesian networks because this is an efficient tool for introducing the plate scan error flow as a variable in the model and mending the mistakes in the scan pattern. Several examples are used to illustrate the proposed model. Finally, some conclusions are included.

[1]  Johan Wideberg,et al.  Deriving origin destination data from a mobile phone network , 2007 .

[2]  Enrique Castillo,et al.  Trip matrix and path flow reconstruction and estimation based on plate scanning and link observations , 2008 .

[3]  Martin L. Hazelton,et al.  Estimation of origin-destination matrices from link flows on uncongested networks , 2000 .

[4]  M. Bell.  THE ESTIMATION OF ORIGIN-DESTINATION MATRICES BY CONSTRAINED GENERALISED LEAST SQUARES , 1991 .

[5]  Francisco G. Benitez,et al.  Review of traffic data estimations extracted from cellular networks , 2008 .

[6]  E. Cascetta.  Estimation of trip matrices from traffic counts and survey data: A generalized least squares estimator , 1984 .

[7]  H. Spiess.  A MAXIMUM LIKELIHOOD MODEL FOR ESTIMATING ORIGIN-DESTINATION MATRICES , 1987 .

[8]  E. Cascetta,et al.  A MODIFIED LOGIT ROUTE CHOICE MODEL OVERCOMING PATH OVERLAPPING PROBLEMS. SPECIFICATION AND SOME CALIBRATION RESULTS FOR INTERURBAN NETWORKS , 1996 .

[9]  L. Bianco,et al.  Combinatorial aspects of the sensor location problem , 2006, Ann. Oper. Res..

[10]  Monica Gentili,et al.  Locating Active Sensors on Traffic Networks , 2005, Ann. Oper. Res..

[11]  Fred Godtliebsen,et al.  Edge-Preserving Smoothers for Image Processing: Rejoinder , 1998 .

[12]  M. Maher.  INFERENCES ON TRIP MATRICES FROM OBSERVATIONS ON LINK VOLUMES: A BAYESIAN STATISTICAL APPROACH , 1983 .

[13]  Bruce Hellinga,et al.  A Statistical Analysis of the reliability of using RGS Vehicle probes as estimators of Dynamic O-d departure rates , 1994, J. Intell. Transp. Syst..

[14]  Enrique F. Castillo,et al.  Traffic Estimation and Optimal Counting Location Without Path Enumeration Using Bayesian Networks , 2008, Comput. Aided Civ. Infrastructure Eng..

[15]  Michael A. West,et al.  Bayesian Inference on Network Traffic Using Link Count Data , 1998 .

[16]  Shiliang Sun,et al.  A bayesian network approach to traffic flow forecasting , 2006, IEEE Transactions on Intelligent Transportation Systems.

[17]  Enrique Castillo,et al.  Optimal traffic plate scanning location for OD trip matrix and route estimation in road networks , 2010 .

[18]  Santos Sánchez-Cambronero,et al.  Predicting traffic flow using Bayesian networks , 2008 .

[19]  S. Bekhor,et al.  Route Choice Models Used in the Stochastic User Equilibrium Problem: A Review , 2004 .

[20]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[21]  Hai Yang,et al.  Optimal traffic counting locations for origin–destination matrix estimation , 1998 .