Load Truncation Approach for Development of Live-Load Factors for Bridge Rating

Abstract

Various local governments have developed state-specific vehicular live-load factors for bridge rating. However, a significant computational demand is often associated with such an effort. This is due to the large size of the weigh-in-motion (WIM) databases frequently used in the procedure. In this study, a method is proposed that can significantly reduce the computational cost of the analysis, while still maintaining reasonable accuracy. The proposed approach develops approximate live-load random-variable statistics by truncating the WIM database based on gross vehicle weight, and then a complete reliability analysis is conducted to develop new live-load factors that meet AASHTO-specified rating standards. Two WIM databases, one based on typically legal vehicles and another based on unusually heavy vehicles, are considered for evaluation. Results of the proposed approach are compared with an exact assessment as well as to a simplified method suggested by AASHTO. It was found that the proposed approach may provide very large reductions in computational cost with minimal loss of accuracy, whereas significant inaccuracies were found with the existing simplified approach.