Traffic Risk Assessment Based on Warning Data

To address the issues of insufficient danger excavation and long data collection period in traditional traffic risk assessment methods, this paper proposes a method based on driver’s improper driving behavior abnormal vehicle state warning data. Meanwhile, analyses built environment’s impact using spatial econometric model. Firstly, system with relative incidence (eye closure, yawn, looking away) (rapid acceleration, rapid deceleration, lane departure) warnings as indicators is constructed. Then, responsibility weights each type were determined entropy weight method. The classification thresholds Gaussian Mixture Model algorithm. Finally, model was used to quantify environment factors characterized by Point Interest (POI) regional risk, results class dependent variable. bus Zhenjiang, Jiangsu Province, are employed an example for validation. geographic cell 1 km × scale applied basic unit. show that optimal threshold road levels I II 1.92, accuracy rate 79.3%; III 0.75, 83.4%. number residential areas, mixing degree, stops significantly positively correlated transit risk. study provide references developing customized accident prevention measures appropriate setting urban supporting facilities.