Investigating the Accuracy and Validity of Iran's Concrete Bridge Management Method Using Data Mining Approach

Abstract

In this research, data mining algorithms were employed to validate Iran's bridge management model.  The results indicated that all three data mining methods predicted the worst possible scenario for bridge failure when the daily road traffic volume is less than 8611 vehicles per day. Concrete science experts have identified factors such as high daily traffic rate, high percentage of light heavy vehicles in traffic, heavy snowfall, and number of days with ice on the road per year as the most influential factors in the failure of concrete structures, particularly concrete bridges. Since all three analysis methods yielded almost identical results, it can be concluded that the traditional Iranian model is not valid.