LRFD Load Calibration for State of Michigan Trunkline Bridges

This report presents the process and results of a research effort to calibrate the live load factor for the load and resistance factor (LRFD) design of bridges on Michigan’s trunkline roadways. Initially, the AASHTO LRFD Bridge Code was reviewed which included investigation into the design code’s background documentation (NCHRP Report 368). Weigh-in-motion (WIM) data were procured for more than 100 million trucks at sensor locations throughout the state of Michigan, including those gathered by other researchers earlier. The WIM data were divided by functional classification and numerically run over influence lines for 72 different critical load effects present on 20 randomly selected bridges, and then projected to create the statistical distribution of the 75-year maximum . Several projection techniques were investigated for comparison. Projection using the Gumbel approach presented herein was found to be the most theoretically accurate for the data set. However, taking into account the practical approach used in the calibration of the AASHTO LRFD Bridge Code, a more empirical and consistent approach was selected for application. Based on the findings presented herein and those of the Phase I portion of this study, the live load factor was calibrated using an approach that was as consistent as possible with that used for the AASHTO LRFD Bridge Code calibration. A reliability index β of 3.5 was used as the structural safety target in both calibrations. Based on the calibration results herein, it is recommended that the live load factor should be increased by a factor of 1.2 for the Metro Region in Michigan to cover observed heavy truck loads. For other regions in the state, this additional factor is not needed. The cost impact of this recommended change was also studied and documented in this report, and was estimated at a 4.5% cost increase for the Metro Region only.