Methodology for uncertainty-based inspection planning of concrete bridge decks using mechanistic models of crack formation and propagation
Date
2015
Authors
Sanders, Patrick, author
Atadero, Rebecca, advisor
Ozbek, Mehmet, advisor
Senior, Bolivar, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
The bridge inspection program in the United States started in 1971 following the collapse of the Silver Bridge in Ohio. Since then, bridges have generally been inspected every two years. This uniform inspection interval for all bridges is not the most efficient system for conducting inspections because many bridges do not require inspections this frequently, and leads to unnecessary use of inspection resources on bridges that do not require them. Efforts toward changing the bridge inspection program to a risk-based program are being made. Such a program would allow for bridge inspection timing to be based on each particular bridge’s need, and each bridge would then be inspected for the components that present the highest risk to the structure. The research presented herein focuses on using the uncertainty in the current condition of concrete bridge decks to plan inspection timing, and to plan inspection type based on limited resources. A mathematical program called Concrete Deck Cracking Probability Model (CDCPM) is written in MATLAB to model the uncertainty in the occurrence of transverse cracking and delamination in the deck. Through literature review, mechanistic models of the processes that affect cracking are determined and implemented in the program. Using Monte Carlo simulations, the uncertainty in the occurrence of cracking and delamination is analyzed based on the input parameter uncertainty. The effect of different climates and mechanistic models on the prediction of cracking is explored using CDCPM. This model is then applied to a bridge in Larimer County, Colorado to see how the results compare to actual inspection results, and then the model is used to plan a future delamination inspection on the bridge. Another scenario is investigated where CDCPM is used together with uncertainty forecast plots to allocate inspection resources to two bridge decks. CDCPM revealed that the uncertainty in the current bridge condition is heavily influenced by the mechanistic models chosen for the analysis, as well as some of the parameters used in the models. The surface chloride content has a large effect on the delamination of the concrete deck, while the relative humidity and ambient temperature for a given location can affect the probability of transverse cracking. Analysis of the Larimer County bridge inspection reports showed that the model can be good indicator of future damage to the deck. The model provides unique insight into inspection planning using uncertainty. Using the model output along with uncertainty forecast plots of different inspection methods, the inspection manager is provided with a useful planning tool that gives information on the effect of inspection methods on bridge condition assessment. Suggestions for implementing the bridge inspection plan for agency use is also provided.
Description
Rights Access
Subject
bridge inspection
bridge deck modeling