TY - JOUR AB - © 2018 Weiwei Qi et al. Drivers' mistakes may cause some traffic accidents, and such accidents can be avoided if prompt advice could be given to drivers. So, how to detect driving risk is the key factor. Firstly, the selected parameters of vehicle movement are reaction time, acceleration, initial speed, final speed, and velocity difference. The ANOVA results show that the velocity difference is not significant in different driving states, and the other four parameters can be used as input variables of neural network models in deceleration zone of expressway, which have fifteen different combinations. Then, the detection model results indicate that the prediction accuracy rate of testing set is up to 86.4%. An interesting finding is that the number of input variables is positively correlated with the prediction accuracy rate. By applying the method, the dangerous state of vehicles could be released through mobile internet as well as drivers' start of risky behaviors, such as fatigue driving, drunk driving, speeding driving, and distracted driving. Numerical analyses have been conducted to determine the conditions required for implementing this detection method. Furthermore, the empirical results of the present study have important implications for the reduction of crashes. AU - Qi, W AU - Wang, Z AU - Tang, R AU - Wang, L DA - 2018/01/01 DO - 10.1155/2018/8014385 JO - Journal of Advanced Transportation PY - 2018/01/01 TI - Driving Risk Detection Model of Deceleration Zone in Expressway Based on Generalized Regression Neural Network VL - 2018 Y1 - 2018/01/01 Y2 - 2024/03/29 ER -