The SafeView solution – blog five

As I reach the end of my internship, I’ve been reflecting on where this project has taken me along the way. It comes back to two key questions. Could SafeView, a forward visibility model, be a useful tool for better understanding overtaking opportunities? And why are these opportunities important?

Highlights of the SafeView project

Initially, I undertook a literature review, which discussed why an adequate supply of overtaking opportunities is important to reduce driver frustration. The ability to overtake is largely influenced by traffic volume, composition (e.g. amount of heavy vehicles) and the available sight distance. Where overtaking opportunities cannot be provided, passing facilities should be considered, such as passing lanes or slow vehicle bays.

Looking into the Waka Kotahi Crash Analysis System (CAS) data reinforced these findings. Modelling overtaking and head on crashes showed that important features were road curvature and traffic volume. Overtaking crashes were shown to be more likely to occur on straight, flat road sections, which is where we expect overtaking to happen. Words like “failed”, “truck” and “impatience” were found in overtaking crash descriptions.

Mapping the CAS data highlighted locations where overtaking crashes were happening more frequently. Some of these road sections were tested using SafeView. Although sight distance was highly variable in some places, SafeView did an excellent job of determining where the potential overtaking opportunities were.

Next steps

In the future, I believe that SafeView could be adapted to model the level of risk for unsafe overtaking across a road section. The available sight distance should be taken into account, as well as the amount and type of traffic, existing passing opportunities and the locations of historical overtaking-related crashes. This could help us to understand whether a road section may need treatment.

Taking part in this project has been a fantastic opportunity for me to grow my project management skills, to better understand how I can use data science in the “real world”, and to work alongside a great group of people. It’s exciting to see how analytical technology might be used to make travel safer on our road to zero journey.

SafeView