VEHICLE PROBES AS REAL-TIME ATMS SOURCES OF DYNAMIC O-D AND TRAVEL TIME DATA

Abstract

The IVHS field operation studies, that are being carried out at present, have demonstrated that the communications technology, on-board electronics and user interface of in-vehicle route guidance systems are rapidly approaching a state of development which, while not completely mature, could certainly support the deployment of several types of practical systems. Lagging considerably behind in this development is the availability of adequate real-time data to support such systems, as the data that is provided from UTCS and FTMS sources have clearly been shown to be deficient in terms of providing reliable link travel time data (especially on surface streets), and estimating real-time O-D data for the entire driver population. The proposed solution, to the latter traffic engineering problem, is expected to be the deployment of sufficient numbers of vehicle probes to directly provide such data when drivers start their trips (O-D) and traverse the various links (link travel time). Unfortunately, in view of the fact that sparse vehicle probe data has also been shown to have a considerable error component associated with it, an assessment of exactly what level of market penetration is necessary to provide adequate probe samples has become critical. This paper provides a combination of analytical development and simulation results to answer this question. Both analytical methods are developed for a general network, and are illustrated in detail for a specific example