A method for creating bicycle-specific safety performance functions or cyclist safety performance functions like that used for motor vehicles in the United States Highway Safety Manual and application of that method to motorist-cyclist collisions at intersections in Boulder, Colorado is the focus of this interesting research paper, titled “Bicyclist safety performance functions for a U.S. city” and published in the journal, Accident Analysis and Prevention.
Before I go on, I have one small change to make to the language used in respect of this paper and that is the term bicyclist is often used in the United States to describe a cyclist. As this blog is Australian centric I have taken the liberty of using the term “cyclist” instead. My apologies to our American bicyclists and readers.
Cyclist Safety Performance
The authors in their paper argue, and I as an academic I am inclined to agree, that if we have a better understanding the fundamental relationship of traffic volume (read number of cyclists) to motorist-cyclist collisions it will help lay the groundwork for future research and allow road safety experts and traffic planners to investigate the impact of specific infrastructure, speed, and other potential factors that may have an impact on cyclists’ safety. Something I would hope is pretty important in today’s world.
It is often argued in forums and the like that the more cyclists on a road, the lower risk to cyclists, a simple matter of visibility or awareness. This paper goes someway to providing rigorous evidence in support of this view.
In this study intersections were the chosen as the unit of analysis since over two-thirds of the motorist-cyclist collisions in the Boulder datasets used occurred at intersections or were intersection related. Australian studies such as Lindsay (2013) also support the view that intersections are a significant point of motorist-cyclist crashes so again it has a point of interest for Australian readers.
Whilst the full paper should be review carefully for an in-depth understanding of the findings, in summary the authors found that the Boulder city models created in the paper illustrated the following essential points:
- Motorist-cyclist collisions at signalised intersections are significantly related to the annual average daily traffic count and annual average daily cyclist count;
- Motorist-cyclist collisions at signalised intersections increase non-linearly with increasing bicyclist and motorist volumes;
- Collisions per cyclist decrease with increasing cyclist volumes; and
- The models indicate that intersections with fewer than 200 annual average daily cyclists have substantially higher collisions per cyclist.
- Intersections with higher bicycle volumes tend to have fewer collisions per bicyclist.
- The safest intersections for cycling predicted by the model are those with high bicycle volumes and low motor-vehicle traffic.
Interestingly the authors went on to suggest that
though no designated bicycle boulevards were included in the model, roads with low motorist traffic and high bicycle volumes, such as bicycle boulevards may minimize risk to cyclists.
This is one aspect that further needs exploring as does the concept of cyclist’s “super-highways” or transport routes similar to what we see developed for motorised traffic.
The full reference for the paper is as follows:
Nordback, K., Marshall, W. E., & Janson, B. N. (2014). Bicyclist safety performance functions for a U.S. city. Accident Analysis & Prevention, 65, 114-122.