Real-time Estimation of Travel times on Arterial Roads – the Arrb Travel Time Model

Provision of travel time has been identified as a key road user support service which assists road authorities to improve network operation strategies and implement control measures. Travel times are commonly estimated for freeways where they are displayed to road users via variable message signs (VMS) but have proved far more difficult to calculate for arterial roads where traffic flow is interrupted. ARRB Group (ARRB) has developed the ARRB Travel Time Model (ATTM) to calculate real-time travel time estimations at link level (between two signalised intersections) using real-time System Monitoring (SM) data retrieved from SCATS. Link travel times are then summed to provide travel time estimates at segment levels (several links) and at route levels (several segments). The ATTM calculates travel time as the sum of cruise time, uniform delay and overflow delay. Initially developed in 2004 for Melbourne’s Princes Highway East, the model was extended to produce travel times for buses in dedicated bus lanes with signal priority along Melbourne’s Hoddle Street. In 2007 the ATTM was deployed for the Western Australian Reid and Tonkin highways and work is currently underway to extend this ATTM for the Auslink transport corridor from Kewdale to the Port of Fremantle. This paper describes and reviews the ARRB Travel Time Model for estimating arterial travel times for general traffic. The paper finds that the ARRB Travel Time Model is robust and has improved in its accuracy since its initial deployment (within 30% at link level, 10% at segment level and 5-10% at route level). To enhance the system, it is recommended that the system’s configuration be automatically established through interrogation of corporate data sets, additional detectors be provided at slip lanes and SCATS Strategic Approaches be allocated for all approach traffic lanes at all intersections.