Road Safety Risk Identification Driven by Abnormal Driving Behavior Data

With the development of vehicle network technology, transportation companies have generated GPS-based driver behavior data for operating vehicles, providing potential research on and road safety risk perception. Research perception at present stage generally involves fuzziness subjectivity, degree is difficult to quantify, big data-based has not yet been used practically. To address this problem, paper proposes a method quantifying based hierarchical analysis fuzzy comprehensive evaluation under drive abnormal driving data, constructs scientific reasonable urban identification model, explores important factors affecting safety. The results certain reference effect traffic management departments accurately locate dangerous sections in network.