Submitter
Can Arkali, Analytic Science Senior Director, Scores Analytic Development
Company
FICO
Industry
Analytic Technology
Business Challenge
Despite advancements in safety technology, road fatalities are on the rise and results continue to deteriorate. Per the NHTSA, “human choices,” such as driving while intoxicated or distracted, account for 94% of serious accidents. The value of data and analytics is when it can provide insight and predictive foresight. In the case of the FICO® Safe Driving Score, this is the identification and reinforcement of safe driving behaviors. For commercial fleets, the goal is to reduce preventable accidents and associated repair, downtime, liability costs and most importantly, to get their drivers home safely each night.
Analytics Solution
The FICO® Safe Driving Score leverages telematics-based driving data to predict the likelihood of future driving incidents. It provides a consistent and objective measure of driver risk and safety based on driving behavior characteristics including acceleration, braking, cornering, speeding and cellphone distraction. Designed to be used within a smartphone application, it provides a driver’s risk score after each valid trip and over time. The higher the score, the more likely a driver is to display safe driving behaviors in the future. Researching and evaluating telematics data for several years, FICO analyzed depersonalized data covering multiple countries, millions of road miles, and roughly 20 million trips.
Impact
The safe-driving industry trend is to build confidence through active coaching. The FICO® Safe Driving Score’s first application is within Mentor® by eDriving Fleet, a closed-loop, smartphone app. This platform fuels the FICO® Safe Driving Score and provides a “playlist” of engaging interactive training interventions customized for each driver. As with all FICO® Scores, there is significant transparency and insight that is shared with opt-in drivers and their supervisors. Sensitive to data privacy concerns, the program is only used with the express consent of the driver. Confirming the efficiency of the model, those who complete the closed-loop training consistently received the highest scores (lowest risk).