Multi-dimensional periodical analysis of traffic accidents in the central districts of Diyarbakir

Highways constitute a problem that affects the society both mentally, mentally and physically due to its intensive use due to many reasons. This chronic problem that needs to be solved globally is "TRAFFIC ACCIDENTS". In the last decade, approximately 12 million traffic accidents have occurred in Türkiye, resulting in the death of 62,762 people and the injury of 3 million 24,470 people. In this thesis, accidents occurring in Diyarbakır, a developed and highly populated province located in the Southeast of Turkey, will be evaluated. The data on accidents occurring in the four central districts of Diyarbakır, namely Kayapınar, Bağlar, Yenişehir and Sur, cover the period between 2013 and 2023, obtained from the General Directorate of Security. Within the scope of the study, traffic accidents, each of which has different components, have been analyzed in detail under four headings: temporal, structural, type of occurrence and driver characteristics. The results show that the peak times of accident occurrence reflect the routine life reality of the society. The results of the road structural analysis show that accidents are more likely to occur on road sections that drivers perceive as safe, convenient and comfortable for them, with different levels of accidents according to each road structure element. In terms of the type of occurrence, it is evaluated that accidents with different impact and collision situations caused by speed, attention and not driving in accordance with environmental conditions and traffic rules are the majority. It is an important result that accidents occur more frequently at points where road safety elements, which have an important share in preventing accidents, are absent and inappropriate. The result of the analysis of the data, in which one third of the drivers involved in the accident were recorded as faultless, reflects a tragicomic situation. It is understood that there is a male predominance in the gender of the drivers who cause and deal with accidents, and that experienced drivers in the 25-64 age group are more prone to accidents. It is also revealed in the light of the analyzed data that the level of drunk driving remains low. In addition, trend analysis on total fatalities and injuries and correlation analysis between accident parameters were carried out. Trend analyses did not reveal a clear pattern for fatalities, but clearly showed trends and differences in injuries, especially seasonal patterns. Pearson's correlation analysis shows that there is a high inverse correlation between the type of accident and the number of vehicles involved, and a positive correlation between the legal speed limit and the presence of a pedestrian path, and between the presence of a safety lane and the presence of a lane line. However, for fatalities, a high positive correlation was found between the legal speed limit and the type of accident.