Traffic-Based Condition Assessment and Fatigue-Life Predictions for a Highway Bridge

Most civil infrastructure today was built during the second half of 20^th century and is nearing the end of its service life. With increasing infrastructure demand that cannot be met with current resources for new supply, replacement of all aging infrastructure is an unsustainable solution. Asset management decisions such as repair, improvement and extension of existing infrastructure are enhanced through monitoring since most structures have much reserve capacity. Recent advances in sensing and computing technology enables multiple-model data interpretation for structural identification and capacity prediction. In this paper, the application of error-domain model falsification for fatigue-life prediction is presented when data from a weigh-in-motion station and strain gauges is available. The application of this methodology is illustrated on a highway steel-concrete composite bridge having four spans with a total length of 219.8m. It is concluded that a reduction in uncertainty is justified and this leads to longer predictions of remaining fatigue life. This methodology enables condition assessment and fatigue-life prediction without interrupting the functionality of the bridge.