You are here

Bayesian Methods for Prediction of Deterioration in Concrete Bridge Decks Using Visual Inspection Data

Bridges are key components of transportation infrastructure systems that are exposed to various uncertain environmental stressors and loading conditions. The proper functionality of these structures is critical for local and regional economic prosperity along with the assurance of public safety. Aging and deterioration are primary concerns regarding the performance bridges degrading to deficient or obsolete states. According to a recent ASCE infrastructure report (ASCE 2013b), it is estimated that every day more than 200 million trips in 102 largest metropolitan regions in the US are taken across deficient bridges. As the demands on public funds increase, it is becoming even more critical to determine and implement optimal decisions aimed at maintaining and improving public infrastructure. Bridge deck deterioration forecasts provide the necessary inputs in support of such decisions. More accurate forecasts are expected to lead to more effective decision. The ability to use bridge inspection data to update the parameters of bridge deck deterioration models on an ongoing basis following each inspection season will lead to more accurate forecasts. The enhanced accuracy of predictions of future states of bridges will lead to more effective maintenance decisions that results in increased reliability of assets, enhanced ride comfort, and less interruptions and time delays for passing traffic. Moreover, the quantification of the value of such updates with respect to the state-of-the practice will allow researchers and state agencies to better allocate the research efforts and investments vis-à-vis the various aspects of deterioration modeling and parameter estimation efforts.


Funding Source: NEXTRANS: USDOT Region V Regional University Transportation Center