Probability-based evaluation of vehicular bridge load using weigh-in-motion data

Abstract

The Load and Resistance Factored Design (LRFD) method has been implemented for designing bridges in Indonesia for more than 25 years. LRFD treats load and strength variables as random variables with specific safety factors for different types of load and strength variables. The nominal loads, load factors, reduction factors, and other criteria from the bridge design code can be determined to meet the reliability criteria. Statistical data from weigh-in-motion (WIM) vehicular load measurements that were taken on North Java Highway Cikampek–Pamanukan, West Java (2011), were used in this study as statistical load variables. A 25 m simple span bridge with reinforced concrete T-girders was used as the model for structural analysis due to the WIM measured and nominal vehicular load based on the Indonesian bridge loading code RSNI T-02-2005, with the applied bending moment of the girders as the output. The distribution fitting result of the applied bending moment due to the WIM measured vehicular loads was lognormal. The maximum bending moment due to the nominal vehicular load from RSNI T-02-2005 is 842.45 kN-m and has a probability of exceedance of 5x10-5. It can be concluded from this study that the bridge designed using the standard loading from RSNI T-02-2005 was safe, since it has a reliability index (β) of 4.68, higher than the target reliability, ranging from 3.50 or 3.72.