New Technology Predicts Drivers' Behaviors

Fatalities in truck wrecks are high. The sheer size and mass of the vehicles are doubly dangerous if not operated safely. The most often found causes of truck accidents include distracted driving and sick or tired drivers. While they may be the most common causes, they are not the only causes.

Keeping motorists safe on the roadways entails addressing all causes, common and otherwise. That hasn’t always been an easy task. New technology has been developed to allow researchers to study other reasons and causes behind truck accidents.  

Truck owners and fleet managers can analyze data using predictive analytics; a collection of information on a specific driver’s past driving behaviors.  Using a variety of safety models, experts can evaluate the data to predict future driving behavior. 

The safety models are designed to show if a driver is susceptible to specific driving behaviors like hard breaking, distracted driving, speeding, fatigue and other issues. The program gathers information and then makes predictions based on the past behavior of the driver but also common sense assumptions.

FleetRisk Advisors is a company that provides these analytics. Their vice president and general manager, Vikas Jain, relays a story of a fleet whose star driver showed at risk for an accident in one of their tests.  The fleet dismissed the prediction as a glitch since the driver had an impeccable record. 

Unfortunately the driver was involved in a terrible wreck only days later. The test revealed information about financial and emotion stressors the driver was experiencing and used it along with other information to predict the immediate eventuality of an accident.

The tests are used only as a warning system for the fleet manager. Using this predictive technique as a tool for better communication, fleet managers are actually able to deter accidents. Once a test identifies someone who is at risk for a specific driving violation, it’s up to the fleet manager to decide how to take the next step.

Jain says that data has shown that drivers who were marked at risk and received remediation were far less likely to commit driving violations than the rest of the driving pool. There are other benefits as well. Some models are designed to predict which drivers are prone to leave or file workman’s comp. They can also be designed to aid fleet owners in deciding who to hire.

The tests analyze three years of data and include over 30 behavior predictors like how much money the driver makes, how many text messages were sent from inside the truck, financial and emotional predictors. 

Models vary from fleet to fleet and only have about 20% of the same predictors. Many of the predictors are surprising to the layperson. It isn’t always easy to see how they can predict a person’s driving behavior. 

The age of technology has seeped into every aspect of modern life and trucking is no different. The onset of GPS systems and other onboard technologies have helped driver issues to an extent. However, this predictive analysis technique is the first of its kind that allows for the hindrance of future issues.