Statistical Analysis for Study of the Effect of Dark Clothing Color of Female Pedestrians on the Severity of Accident Using Machine Learning Methods

The color and brightness of pedestrian clothing are among the factors that could increase severity their accidents due to lack visibility, especially at night. Today, as most Iranian females tend wear hijab or dark clothing, necessity investigating female influenced by is important. Many studies have been performed analyze accidents, but a study has not yet conducted determine effect pedestrians on accidents. Therefore, in this study, 12 independent variables affecting such color, age, accident time, day, weather condition, education, action, crossing facilities, permit, job, road classification, fault status were studied. Frequency analysis, Friedman test (FT), Factor Analysis (FA) methods, well modeling methods Multiple Logistic Regression (MLR) Artificial Neural Networks (ANNs) using Multilayer Perceptron (MLP) Radius Basis Function (RBF), used. Results indicated had significant influence chador increased probability MLP model better prediction percentage than rest, accuracy which was 94.6%. Finally, safety solutions presented according results reduce safety.