Study of Urban Taxi-Related Accident Analysis Using the Multiple Logistic Regression and Artificial Neural Network Models

In this research, factors affecting the severity of property damage only (pdo) and injury/fatal accidents were examined using taxi-related accident data from march 2015 to march 2021 in urban sites of rasht city. the multiple logistic regression and artificial neural network (ann) were applied to recognize the most influential variables on the severity of accidents. results indicated that the multiple logistic regression in the backward stepwise method had a prediction accuracy of 88.54% and r2 value of 0.871. moreover, the regression analysis revealed that the wet surface condition, night without sufficient light, rainy weather, kia pride taxi and lack of attentions increased the severity of accidents, respectively. the most important result of the logit model was the significant role of environmental factors, including slippery road surface, unfavorable weather as well as poor lighting condition, and also indicated the dominant role of poor quality of vehicles along with human factors in increasing the severity of accidents. comparing the correct percentage of prediction in the multiple logistic regression and ann model, the results showed that ann model performed better so that the prediction accuracy of ann was 95.8%.