Method for estimating effect of traffic volume and speed on pedestrian safety for residential streets

When local governments find that the demand for traffic-calming measures exceeds their available resources, it may be necessary to select the most needful or most promising projects. Since pedestrian-vehicle collisions tend to be rare on local streets, it will usually not be possible to use accident counts to identify high-hazard locations. A deterministic model of a collision between a pedestrian and a vehicle is described and then some of the model's variables are allowed to be random. The probability of a collision can then be evaluated as the probability of obtaining a set of variable values leading to a collision. Both parametric and nonparametric methods are described and illustrated, including an example in which estimates of the collision probability and the probability distribution of vehicle collision speeds are used to identify promising sites for traffic calming. The methods can be used for any location where an analyst has vehicle spot speed and traffic count data and is able to specify reasonable values for the other model variables.