Prediction of Accidents Using Artificial Neural Networks in North Cyprus

Abstract

When compared to global traffic catastrophes, North Cyprus road system ranks among the safest up to the current fiscal quarter of 2025. Considering an approximate estimate of about 1.3 million deaths and millions of non-fatal injuries which are caused by road accidents each year, making them a serious public health concern, it is challenging to anticipate and avoid traffic accidents on new road segments due to a multitude of subjective and objective factors. With minimal data sets at disposal, artificial neural networks (ANN) have proven to be successful in forecasting traffic accidents. Therefore, it is crucial to properly evaluate the pertinent hazards. A methodology for creating an accident risk prediction model that can be utilized as a tool for infrastructure management decision-making is presented in this study. These models use easily determined objective factors, such as road length, terrain type, road width, average daily traffic volume, and speed limit The approach makes it possible to handle the supplied data appropriately, looks at how it can be utilized to create models using artificial neural networks (ANNs), and establishes a methodical ANN optimization procedure to find the best ANN model architecture. The approach is applied using accident count data as a result, it is essential to accurately assess the relevant risks. This paper presents an approach for developing an accident risk prediction model that can be used as a tool for infrastructure management decision-making. The method offers a systematic ANN optimization process to determine the optimal ANN model architecture, making it feasible to handle the provided data effectively, and examines how it may be used to develop models using artificial neural networks (ANNs). The method is used with data on the number of accidents on North Cyprus road