On-line Performance Measurement Models for Urban Arterial Networks

This paper addresses the problem of real-time estimation of urban arterial network performance. A number of analytical time-dependent models have been formulated and implemented for the on-line estimation of arterial link and system travel times. The proposed models provide alternative ways of incorporating the contribution of various network and operational characteristics to the travel time. The model inputs are volume and occupancy data from loop detectors, and signal control parameters. The models were tested on several real-life networks representing a wide range of configurations, demand patterns and control features. The results indicate that simple models appear promising in accurate measurement of link travel times. More complex models accounting for the randomness of delays at intersections improve the network-wide travel time estimates at the expense of data requirements and computational efficiency.