Aaron huff is senior editor of Commercial Carrier Journal. E-mail firstname.lastname@example.org or call (801) 754-4296.
One of the presentations at the recent Qualcomm Vision 2008 users conference in San Diego discussed what would be the “next big thing” in the industry. When C.R. England, the largest refrigerated carrier in the nation, showcased how it uses predictive modeling to improve safety, the “next big thing” suddenly became very real.
Carriers use various programs, initiatives and technologies to improve safety – and for a good reason. There are no silver bullets. Fleets must consider many different and unique factors that contribute to accidents and other types of risk.
The goal of most safety managers, however, is the same: target their efforts at drivers with the greatest risk, and manage those efforts effectively, says Chad England, vice president of safety, training and recruiting for the Salt Lake City-based company.
In 2005, England started an information technology project to target the company’s safety efforts and resources toward drivers with the highest risk. The plan was to create profiles of driver safety performance. England submitted the requirements of the project for approval, but it proved to be too complex for the company’s internal resources.
In early 2007, England approached FleetRisk Advisors (www.fleetriskadvisors.com), a company that provides technology-based risk management and loss control services for the transportation industry. The technology and modeling techniques developed by FleetRisk were much more comprehensive than the profiling idea England says he originally had considered.
When FleetRisk visited C.R. England’s headquarters, analysts extracted three years of data and more than 1,800 separate data elements from multiple databases at C.R. England; this data then was assembled into a C.R. England-specific relational database to facilitate modeling and analytics.
When FleetRisk returned with its first risk-based model, England was skeptical – but all of the results could be verified. “We found data that you wouldn’t think would affect accidents,” he says. “We found stuff we could not have possibly found on our own.”
C.R. England worked with FleetRisk to develop a unique Predictive Performance Services (PPS) program. The PPS considers historical and current driver performance data to score and rank drivers by their likelihood of having a preventable accident in the next 28 days, England says.
The program incorporates fatigue scores based on a system from Circadian Technologies that considers position histories and logbook information. The program also incorporates real-time data from Qualcomm’s Critical Events Recorder (CER), a software system that captures safety and performance data from C.R. England’s trucks.
To properly manage the information from the PPS program on a daily basis, C.R. England built an AS400-based application it calls the Safety Workbench. Safety and driver managers use the application to track driver scores for performance and fatigue; the performance and fatigue scores are refreshed every week, but fatigue soon will be a daily score, England says. The application provides additional data that explains what predictors are influencing the driver score for the current week, England says; it also keeps a record of the countermeasure(s) taken for drivers that are identified as high risk. Soon, C.R. England will integrate the daily fatigue scores into its operational screens. Additional enhancements will include scoring loads with a risk factor to help load planners match drivers to loads.
The results from the PPS program speak to its success. Different pieces of the PPS program were put into place in 2007, starting in April. An analysis by FleetRisk showed that by December 2007 year over year, accidents per driver had decreased 22.31 percent and accidents per million miles dropped by 11.86 percent as C.R. England grew by 9.66 percent more drivers and 10.69 percent more miles.
“Eliminating just one accident is worth far more than the cost of the program,” England says.