| dc.description.abstract | Suitability of traffic engineering facilities is critical to traffic safety. A safe transportation network must be designed with good traffic engineering. According to the statistical data, about 300,000 traffic accidents occur each year in Taiwan. How to reduce the chance of traffic accidents has been a topic of continuous concerned for central and local government. In the recent years, the central government enforce local governments to improve traffic environment and safety. This study will was focused on the major factors influence traffic accidents, the major factor included traffic signs, traffic marking, traffic signal, safety of pedestrians and bicycles, road width and lanes were evaluated in this study. Those data were from the survey of “Urban Road Maintenance Management and Pedestrian Accessible Environment Assessment Project” which was held by Construction and Planning Agency. There were two Artificial Intelligence (AI) methods included Gene Expression Programming (GEP) and Support Vector Regression (SVR) to explore the major influence factors of traffic accidents in this study. The result showed SVR analysis got better performance than GEP analysis and showed the most important major factor of traffic accidents is traffic marking. Hierarchical Cluster Analysis was adopted to confirm whether the traffic marking location, the distance between crosswalk lines and stop line, would be influence the chance of traffic accident. Based on the results of cluster analysis, and traffic marking would be influence the chance of traffic accident. Compared the others, the longer distance would obvious reduce the pedestrian traffic accidents. However, the longer distance it will also influence the traffic flow on the intersection. It was worthy to further studying. | en_US |