Highway Travel Time Estimation With Data Fusion

This monograph presents a simple,innovative approach for the measurement and short-term prediction of highwaytravel times based on the fusion of inductive loop detector and toll ticketdata. The methodology is generic and not technologically captive, allowing itto be easily generalized for other equivalent types of data.  The bookshows how Bayesian analysis can be used to obtain fused estimates that are morereliable than the original inputs, overcoming some of the drawbacks oftravel-time estimations based on unique data sources. The developed methodologyadds value and obtains the maximum (in terms of travel time estimation) fromthe available data, without recurrent and costly requirements for additionaldata. The application of the algorithms to empirical testing in the AP-7 tollhighway in Barcelona proves that it is possible to develop an accuratereal-time, travel-time information system on closed-toll highways with theexisting surveillance equipment, suggesting that highway operators might providetheir customers with such an added value with little additional investment intechnology.