Ford’s Big Data Study May Predict Accident Black Spots

Ford claims it has managed to crack a major challenge for smart cities with a system that can predict where accidents will happen.

By employing big data technology, Ford says its system could, potentially, help cities identify locations which, if nothing is done, are most likely to be the scene of future traffic incidents. To help find the answer, Ford Smart Mobility, spent the last year recording a million kilometers of vehicle and driver behavior in and around London.

The company tracked vehicle journeys in the city and logged detailed driving data from motorist events such as braking, its severity, and where hazard warning lights were applied and used this to identify “near-misses”. Ford cross-referenced this information against existing accident reports and built an algorithm to determine the likelihood of where future incidents might occur.

This idea is one of the opportunities identified in the Ford City Data Report that was launched at the Financial Times Future of Transport conference in London. The report, using data that was obtained and analyzed with the consent of participants, took its findings from more than 15,000 days of vehicle use, from 160 connected vans in the city and delivered 500 million data points.

Jon Scott, project lead at City Data Solutions, Ford Smart Mobility, said: “We believe our insights have the potential to benefit millions of people. Even very small changes could make a big difference – maybe cutting back a tree that has obscured a road sign – whether in terms of traffic flow, road safety or efficiency,”

The report also investigated other opportunities, such as how scheduling delivery van journeys for earlier in the day, before peak times, could benefit all road users and how using journey data could help to identify the best locations for electric vehicle charging points.

— Paul Myles is a seasoned automotive journalist based in London. Follow him on Twitter @Paulmyles_



Leave a comment

Your email address will not be published. Required fields are marked *