Impacts of traffic calming measures on speeds on urban roads

The main objective of this research was to improve the understanding of drivers’ behaviour while negotiating traffic calming measures, through the study of the impacts of these measures on speeds.
A case study was conducted in the City of York focusing on speed humps (flat-topped and round-topped), speed cushions and chicanes implemented in sequence. Data collection was conducted at three calibration and three validation sites with vehicles’ passing times simultaneously recorded at 16 points along the links. From these data a speed profile for each individual vehicle could be derived. The influence of various combinations of traffic calming measures on speeds of unimpeded cars and vans was evaluated through those speed profiles, and through acceleration profiles deduced from speed profiles.
The investigation of the hypotheses established from the analyses of speed and acceleration profiles gave insights into specific issues such as the consistency of crossing speeds, the additive effect of subsequent measures, the acceleration and deceleration rates associated with individual measures and travel times along calmed links. The knowledge acquired from these analyses was applied to the formulation of a speed profile model.
To describe drivers’ behaviour along traffic calmed links an empirical model was developed using multiple regression analysis techniques based on data collected at the calibration sites. Speeds along calmed links were described as a function of the input speed, the type of measure and the distance between measures. The speed profile model was shown to be a good representation for the data from the calibration sites. It efficiently predicted speeds of unimpeded vehicles over a given combination of traffic calming measures in sequence. The validation process also indicated that the model provided a good representation of the observed profiles at these sites, with the exception of the prediction of the effects of the chicanes on speeds. This type of
measure was shown to produce diverse impacts on speeds which depended on the detailed design.
While the model is a useful design tool, recommendations have been made for further enhancement to it.