Safety on Heavily Trafficked Urban Motorways in Relation to Traffic State

Abstract

Motorways represent seven per cent of the urban arterial road network in Melbourne yet carry 40 per cent of the urban arterial road travel in terms of vehicle kilometres travelled and this percentage is growing. The number of casualty crashes on metropolitan Melbourne motorways has increased over the decade at a faster rate than on other urban roads in metropolitan Melbourne. Police crash reports more often attribute crash cause to traffic conditions and vehicle interactions rather than infrastructure. As urban motorways are generally built to the highest standards, a new way of looking at motorway safety is needed. This led to the formulation of a hypothesis that the dynamics of the traffic flow are a significant contributor to casualty crashes on urban motorways. To test this hypothesis, in-depth analysis was undertaken on metropolitan Melbourne motorways. Crash data was linked to traffic data including vehicle occupancy (a proxy measure for density), vehicle speed and flow. Occupancy was used to categorise the ‘traffic states’ ranging from free flow to flow breakdown (congestion). Applying a Chi Square Goodness of Fit Test to the linked showed a statistically significant association between traffic state and crashes, with a higher than expected crashes in the traffic states where flow breakdown is relatively certain or has occurred. The results of this analysis can be used to improve safety on urban motorways through the development of Intelligent Transport System strategies to keep the motorway operating at conditions that minimise flow breakdown risk.