Predicting an auto accident's severity
Southern Methodist University said its model can also help uncover accident hotspots and the reasons behind them.
Researchers at Southern Methodist University (SMU) have developed a predictability model that factors in certain variables, such as drunk driving and speeding, to determine how severe an accident would be.
As part of the program, the university established an interactive online system to enable the insurance industry to leverage the tool. SMU reported the model can also help uncover accident hotspots and the reason behind them.
Details can be as precise as estimate the degree of injury to the driver and the car, based on a range of factors, including road conditions, when someone is driving, the state they are traveling in and eight other factors, SMU reported. In addition to injuries, from minor to fatal, the model also projects potential property damage with a wreck involving one or more vehicles as well as medical costs.
For example, the model predicts that if a sober 25-year-old man, driving 80 mph (in a 70 mph zone) in a 2012 Ford Focus on a dry Dallas road at 2 pm is in an accident involving two cars, the chances that he dies are 2%. However, if alcohol is involved, but all other variables remain static, the chances of death increase to 10%.
“This can hopefully influence drivers’ behavior positively and reduce crashes by making drivers more aware of dangerous driving habits,” Tony Ng, an SMU statistical science professor and one of the co-creators of the model, said in a press release. “The model can be considered a useful educational tool to make the general public aware of the risk and avoid those poor driving behaviors like speeding and drunk driving.”
However, Ng cautioned that the model is “only as good as the data.” For its trials, SMU used a representative sample of police-reported car crashes of all types reported to the National Highway Transportation Safety Administration. The data was pulled from the U.S. Department of Transportation General Estimates system, which has known measurement errors, the university reported.
Related: