Discovering the causes of an accident may help authorities prevent future crashes, but after-the-fact information about driver behavior comes too late to help some crash victims. Until recently, the only way to prevent truck accidents was to try to gather information from traffic incidents that had already occurred. Even the latest on board, “real-time” electronic devices only provide driver data that could be used preventatively later.
Risk advisors for truck fleets have come up with a new technology, which takes recorded information, like driver speed and other behaviors, and looks for patterns. Data is analyzed and sifted through a risk model that can then predict how safe a driver is likely to be on the road in the future.
The models analyze three years of driver information from obvious behaviors like speeding to shift times, the number of communications while on the road and pay. More than 30 predictors match the behavior with possible stresses. The combined data creates a probable driver safety rating.
The general manager of the company offering the new technology says predictive analytics monitors behaviors of all truck drivers, not just the ones with clear safe or unsafe driving histories. Individual drivers’ behaviors often fluctuate due to stress, some of which was rooted in conditions outside work.
Predictive analytics allows companies to see problems before accidents happen. Managers then have time to talk with truck drivers about stresses that are affecting their work. The preventive measure is often welcome relief to drivers undergoing outside pressures.
The company’s customer feedback indicates trucking companies that use predictive analytics have fewer accidents than businesses that do not.
Source: Trucking Info, “Using analytics for proactive predictions,” Jim Beach, July 23, 2012