Automated vehicle identification tag-matching algorithms for estimating vehicle travel times - Comparative assessment

The computational complexity associated with three candidate automated vehicle identification (AVI) tag-matching algorithms that could be used to obtain individual vehicle travel time data in real time is examined. These algorithms are suitable for application to a linear roadway facility using transponder tags that do not have programmable memory. Analytical expressions are derived to estimate the worst-case and average computational load associated with each algorithm. A simulation is conducted to test the validity of the assumptions made in these derivations and to perform a sensitivity analysis on several key system parameters, including the rate of flow of AVI-equipped vehicles, the mean travel time between tag reader stations, the coefficient of variation of travel time, and the proportion of vehicles that pass the upstream tag readers.