Fusion of travel time data in Niayesh tunnel using Bayesian inference

As data collection costs decrease, transportation systems have shifted from systems requiring data to systems requiring data analysis. Since the accuracy of these data varies with the sources of data collection, acquiring higher accuracy data from a combination of multiple sources is the main challenge of working with such data. Data fusion is a very efficient mechanism can interconnect data from different sources to increase the accuracy of data in line with the purpose of the study.The main goal of this article is to get the most accurate travel time possible from multiple sources.Among the data fusion methods are the Kalman filter, Bayesian inference, artificial neural networks and Dumpster-Scheffer theory,from which the Bayesian inference is used and its results are investigated.It is proposed that by combining different data sources with different temporal and spatial coverage, the most accurate travel time with maximum spatial and temporal coverage would be achieved.The Niayesh tunnel in Tehran was selected as a case study,where extensive equipment for intelligent transportation systems is installed. After determining the accuracy of each source with floating car data, the fused travel time of one day in which multiple sources data were available was calculated.In this study, considering the possibility of simultaneous access to multiple data sources at the same location,the following source, Google travel time data, Bluetooth travel time data and Inductive loop detectors were fused. The improved travel time can increase the accuracy of travel time costs in transportation planning, information on variable message signs and routing software.