Reducing the risk of accidents is always top of mind for organizations with fleets of vehicles and employees in the field.

A chemicals company that owns and operates a large trucking fleet wanted to do exactly that: develop a smarter, more automated way to keep their fleet of drivers—and those they share the road with—safe.

Existing methodologies for linking historical data to fleet safety had been helpful in identifying high-risk routes and drivers, but they wanted to develop an algorithmic solution that more precisely measures risk by incorporating factors like weather patterns, routes, vehicle specificity, driving patterns, and payload transport risk. And with the proliferation of on-board computers (OBC) and sensors, the data they needed to build a next-generation solution was at their fingertips.

With an Analytics Starter Pack from Topcoder, the chemicals company developed an algorithm that delivered a precision improvement of >40%, as compared to the baseline algorithm, which had a precision scoring of just <5%. During a two-week data science competition on Topcoder, more than 1,000 algorithmic solutions were submitted by over 100 top data scientists around the world. Solutions were scored against a new OBC-based trip risk prediction metric that measured the precision of each submission.

Now armed with a faster, more accurate method of predicting risk, the company can better protect lives as well as cargo and vehicles.

Registrants
187
Competitors
101
Submissions
1080
Durations
2 Weeks