The National Private Truck Council's recent annual conference in Indianapolis featured a range of educational workshops covering everything from safety compliance to driver wellness as well as an exhibit boasting over 100 industry suppliers.
A forward-looking session outlined a computerized model for use within a fleet to predict driver turnover that is currently under development.
Dean Croke, chief product architect & senior vp of FleetRisk Advisors, which is developing the model, explained that it will identify “all attributes of drivers who stay with a company for a specified period of time and who have a productive and safe track record.”
At this point, the model is based on two years of historical electronic data extracted from multiple data systems at a FleetRisk client, bulk hauler and asset-based 3PL Dupre' Transport of Lafayette, LA.
Croke said that during the period studied, Dupre incurred a nearly 20% turnover rate. The data points extracted covered a range of attributes, everything from marital status to number of cash advances per month, even tenure at previous jobs.
The data was crunched through pattern recognition software, allowing FleetRisk to identify the “precise combination of data elements that contribute to optimal driver retention,” he explained.
Croke said the goal is to develop a “target model” of drivers who stay with the company for over two years and are safe and productive employees, too.
Once the model is built at a given fleet, which can take up to six months, it would then be simple to plug in ongoing data to continuously track turnover risk, he added.
Croke expects the model will be used to increase both the average driver's tenure with a fleet and reduce the carrier's accident rate through the hiring and retention of safer drivers.