چكيده به لاتين
Earthquakes have long been recognized as one of the most significant and impactful natural disasters. When an earthquake occurs, transportation networks play a critical role in evacuation, relief, and emergency activities. Therefore, assessing their performance in the face of earthquakes, i.e., evaluating their seismic resilience, is crucial. Given that machine learning models, including neural networks, have a significant role in predictions today, the aim of this study is to develop a machine learning-based model for evaluating the seismic resilience of transportation networks. This research presents a case study on the city of Istanbul, Turkey. Based on data obtained from Istanbul, which includes data related to roads, buildings, and the seismic characteristics of the area, regression neural networks GRNN , RBF and MLP were developed. In these models, in addition to considering damage to road structures, the debris from collapsed buildings was also examined. The results show that the developed GRNN model has an accuracy of 78%, which is higher compared to the other mentioned networks. Furthermore, by evaluating the developed GRNN model, it is determined that the RMSE values and the Fit curve, determined using test data, show satisfactory results compared to other models.