The Research On Microscopic Evaluation System Of Road Traffic Safety

Traffic accidents cause the serious threat to life and property of road users. Social stability and economic development are imperiled too. Road traffic safety is an important issue which concerns people's livelihood and social harmony. For road users, the main concerned aspects are traffic accidents (safety) and speed (rapid convenience). Based on the view of engineering applicability, the paper has conducted the microscopic evaluation of highway traffic safety, and the highway is in motion and does business. The paper has deeply explored the theories such as the occurrence mechanism of traffic accident, data preprocessing technique, traffic safety prediction, speed damage theory, etc. The microscopic evaluation system of highway traffic safety is structured. In the microscopic evaluation system, the improved identified method of accident-prone section, the accident prediction model and the amendatory prediction formulas of running speed are constructed based on the common methods and the road practical situation.Based on the theoretical analysis, data preprocessing technique is introduced into the microscopic evaluation system. The concept and importance of data quality control is emphasized too. The work of data processing and variable selection has effectively achieved in the following traffic safety evaluation. Based on the study of the accident-prone section characteristics and common identified methods, the principal component -Q clustering analysis method has been researched, which uses as the identified method of accident-prone section. Based on the traffic predictive theory and common predictive methods, the accident prediction method of R clustering analysis -multiple linear regression has been analysed. Accident prediction model is tested and the prediction error is analysed. In the test of design consistency, the line connection of adjacent section and the line shape of section have been conducted safety test through using the indicator of running speed. The predictive formulas of running speed have been re-deduced and appropriately revised, so the running speed of section becomes more reasonable.Finally, based on the actual road condition and design indexes, combined with softwares of SPSS and MATLAB, the paper has initially treated the research data in the practical engineering example. Using the relevant methods in the paper, the highway has been conducted the identification of accident-prone section, the construction and analysis of accident predictive model, the test of design consistency and section safety evaluation. The evaluated effects are compared with common evaluated methods. According to the evaluated result, the traffic safety protective measures are proposed. Through example analysis, the relevant computing method and model are reasonable and practical, which provide a new idea and useful reference to microscopic evaluation of road traffic safety.