摘要:AbstractIn the recent years, traffic incidents sharply increase in Turkey. Although traffic regulations strictly applied in Middle East Technical University Campus (METU), Ankara, Turkey, this increasing trend also present in the campus. To prevent traffic incidents, it is crucial to understand where and how they take place, the relations among the incidents, and the locations. The aim of this study is to make a traffic pattern analysis concerning time and locations within METU Campus using spatial data analysis tools. The study aims at answering whether there are any locations which show clustering in the distribution of traffic incidents and there are any changes in the distribution of traffic events for different seasons, days and time periods. In the existing system the events are recorded by gendarmerie and there is no tabular or visual database for the events. Concerning such a need METU Campus is selected to create a traffic incident database and analyze the pattern of incidents. The Traffic incidents are analyzed with both point and areal data analysis for determining whether there is regularity or clustering in the data. The incidents on road are used in the kernel density estimation with network distances along the road network. Nearest Neighbor Distance and the K-function exploration methods have been applied to determine the spatial relation in terms of distance between the incident locations. The results of different methods are compared with each other in order to identify the hot spot zones on roads. The results showed that there are clustering in the traffic incidents and the number of incidents changes by season, day and week. The results could be used as a tool for the management of the traffic system within the METU Campus for improving the traffic safety.