- Publication no: ABC-KNP005-11
- Published: 31 October 2011
More sustainable bridge infrastructure can be achieved by eliminating costly and unnecessary replacement or overly zealous upgrading of ageing bridges. Estimates of current safety of bridges or their remaining service life are critically dependent on correct probabilistic modeling of vehicle and axle masses, and their temporal and spatial distribution. The limit states commonly critical for the reassessment of bridges are the Ultimate Limit State (ULS) for both local components and overall spans, and the Fatigue Limit State, or remaining fatigue life, for local components of metal bridges where individual trucks or wagons generate significant cycles of stress. Weigh-in-motion (WIM) instrumentation provides vital data for building probabilistic load effect models for highway bridges. The objective is to use the estimated values of load effects (from measured vehicle and axle masses) and estimated values of strength (from observed condition and measured material properties and dimensions in situ) to arrive at a reliability as good as that implicit in new design using assumed values of load and resistance. The first step is to establish the integrity of the WIM data, checking for drift from the time of calibration, identifying the permit vehicles and validating any exceptional events. The next is to simulate the loading at the bridge site – often some distance from the WIM site – for multiple presence of heavy vehicles in all lanes. ULS assessment requires application of extreme value statistics to the data. A methodology is presented.
Three examples will be presented. The first is an elderly flat slab bridge with short spans in Victoria. The second is a suite of wrought iron truss bridges from northern New South Wales. The last is the multiple lane, heavily trafficked long span West Gate bridge Bridge, where the critical load case for longitudinal bending of the steel box girder is a traffic jam. The raises the question of acceptable risk and underscores the need for good practice based upon engineering principles, experimental data and experience.