Predictive Coach Pilot Program Evaluation

Predictive Coach proved to be a valuable asset in a transit setting with a fleet already firmly focused on safety. Fleets interested in continually improving safety reap significant benefits from Predictive Coach.

Executive Summary

Driver behavior is the primary contributor in the majority of crashes. One technology that reduces risky behaviors is Onboard Safety Monitoring (OSM) systems. However, data suggest that OSM systems alone are insufficient in lasting change. Lasting behavioral change results from using individualized driver training like Predictive Coach.

Project Objective

The project objective was to evaluate Predictive Coach’s driver training method to reduce risky driving behavior in a municipal setting with a fleet already utilizing dash camera video monitoring.

Project Method

Keolis, an international private transportation fleet with over 63,000 employees, participated in this study at the Keolis terminal in Florida over 13 weeks. The Predictive Coach program was implemented with 59 drivers and two vehicle types, buses md light vehicles.

Analyses focused on two assessments. The first analysis goal was to compare the mean rates of the targeted risky driving behaviors prior to and after the targeted Predictive Coach intervention. The second analysis goal was to examine response generalization to Predictive Coach training.

Predictive Coach lmproved Safety Despite Already Having Cameras lmplemented in Fleet

Results showed the rate of overall risky driving per 1,000 miles in buses was 31 % lower du ring the Predictive Coach program than before the program began, despite Keolis having effective safety programs and a dash camera video monitoring safety solution already in place. The Predictive Coach program offers behavioral reinforcement targeted training, thus effectively addressing specific driver behaviors.

Conclusions

The Predictive Coach program resulted in statistically significant reductions in the rate of most risky driving behaviors tracked. Once the Predictive Coach algorithms were activated, transit bus hard cornering was reduced by 19% and speeding was reduced by 63% when compared to baseline rates across drivers. Additionally, the results showed that all drivers, not just drivers that were assigned a training, reduced their risky driving. Furthermore, not part of the study, risky behavior immediately began to rise once Predictive Coach was disengaged.

During the baseline, the mean rate of overall exceptions was 19.31 per 1000 Vehicle Mi les Traveled (VMT). Du ring the Predictive Coach intervention, the mean rate of overall exceptions was 13.83 per 1000 VMT. This represents a 31 % reduction in overall exceptions.

Many fleets struggle with the challenges of managing telematics data, driver training and fleet management. Telematics without management oversight present inherent liability via willful negligence. Predictive Coach helps combat willful negligence by reducing the workload fleet managers face day to day by automatically assigning behavior-based driver training. Predictive Coach leads the transportation industry with patented technology which saves lives by changing driving behavior and reducing fleet management workload.

Summary

The objective of this study was to evaluate the ability of Predictive Coach’s driver training method to reduce risky driving behavior with Keolis, a well-established private transit fleet already using a dash camera video monitoring safety solution. The Keolis fleet was already safe, with a long history of participating in leading
video-based telematics programs. With this in mind, any positive results from the Predictive Coach program would have been impressive. However, a statistically significant reduction of 63% in excessive speeding events demonstrates the value of Predictive Coach. Even safe fleets can become safer with the Predictive Coach program.

MENU