Affiliation:
1. Rutgers Infrastructure Monitoring and Evaluation (RIME) Group, Department of Civil and Environmental Engineering, Rutgers, The State University of New Jersey, Piscataway, NJ
2. HNTB, New York, NY
Abstract
Truck load spectra based on weigh-in-motion (WIM) measurements have been utilized in developing site-specific live load models to predict the maximum load effects on bridges. Conventional load extrapolation has been utilized to develop the AASHTO load-and-resistance factor design (LRFD) Bridge Design Specifications, while few studies have evaluated the accuracy of the load extrapolation techniques with actual data. The current AASHTO Manual for Bridge Evaluation (MBE) utilizes the top 5% of the load effects to extrapolate the 5-year maximum load effects for load rating. However, in the cases of high truck volume, the predicted 5-year maximum load effects using AASHTO MBE are significantly lower than the observed value because of the selection of the upper tail. Therefore, the choice of the upper tail size needs further validation. This paper proposes a modification to the conventional live load extrapolation method. Firstly, more accurate maximum load values for different return periods are determined through simulation and validated using 7 years of continuous data. Then, the values from conventional live load extrapolations using different upper tail sizes are obtained and compared with the simulation values. The optimal upper tail size is determined when the minimum error is yielded. The findings suggest that using a specific number of trucks for the upper tail yields greater accuracy compared to a percentage-based approach. Specifically, the recommended range is between 3,000 to 5,000 trucks, with an optimal number of 3,600. This paper concludes with recommendations to the AASHTO MBE to enhance the accuracy of live load extrapolation.