Predictive Analytics helps Dupré see accident reduction

Jan. 6, 2010
Dupré Logistics was the first company in the nation to partner with Fleet Risk Advisors to implement Advanced Predictive Analytics, a modeling system that aggregates data and presents potential accidents before they happen to the Dupré management team. As a result, accidents have been reduced by 67% since 2004

Dupré Logistics was the first company in the nation to partner with Fleet Risk Advisors to implement Advanced Predictive Analytics, a modeling system that aggregates data and presents potential accidents before they happen to the Dupré management team. As a result, accidents have been reduced by 67% since 2004.

For Dupré, this investment in the people in their trucks and on the road has been well worth it. A driver identified in the bottom tier, who goes through the targeted risk management training, counseling, schedule adjustment, or programs suggested by management, is 50% less likely to have an accident within the next month.

The primary business objective of the Predictive Analytics implementation has been to combine historical data with the current month’s operational data to predict the next month’s safety performance. In order to accomplish this, computers capture data of a driver in more than 400 data elements. Data aggregation and pattern recognition technology is used to identify risk signatures of drivers, vehicles, and schedules. The program allows Dupré to devise new risk management and performance monitoring strategies with the goal of simultaneously reducing operational costs and mitigating risks.

At the end of each month, the predictive models rank current drivers against a statistical and demographic profile of Dupré’s theoretical best and safest driver. Each tier represents one-third of the current driver population. This predictive segmentation allows Dupré to focus its training and risk mitigation efforts on a small subset of its drivers, rather than the entire driver team.

Based on system results, Dupre’s management team can identify which driver will be more likely to have an accident in the next month. Those in the bottom tier of the predictive system then are counseled by the Dupré management team to adjust risk factors identified. Dupré has designed a suite of targeted risk management and performance monitoring strategies including: focused driver training programs; new safety field representative training; terminal and management scorecards; driver scorecards and enhanced incentive programs; alerts from in-vehicle technology; and dispatch training. Dupré also has implemented a modified recruiting and retention program.

Related