Probabilistic bridge deck condition analysis using ground penetrating radar (GPR)

Ground Penetrating Radar (GPR) systems provide quickly and non-destructively a wealth of data about important features characterizing the internal condition of many civil engineering structures. Knowledge of the location and extent of internal defects can improve the ability to predict the remaining service life of bridge decks providing considerable economic benefits through the selection of appropriate maintenance and rehabilitation actions.; It is the goal of this study to improve upon the current practice of using and interpreting GPR data and to advance a probabilistic methodology to facilitate an evaluation, verification, and validation of the accuracy and reliability of GPR measures on bridge decks. Specific objectives of the research are to: (a) develop a procedure to accurately identify and quantify the area of deterioration; (b) verify and validate the developed procedure through laboratory experiments on bridge deck models; and (c) incorporate the findings of the experimental study and additional findings from real-world (field) applications into a new methodology that improves the current practice of bridge deck condition assessment.; Data from the field and laboratory applications of GPR contributed significantly in achieving a fundamental understanding of material and structural behavior, and in the development of analytical models of quantities associated with the accuracy of GPR measures, the reliability of GPR data interpretations, the repeatability of GPR survey data, and the spatial variability (in the vertical direction) of the radar reflection amplitudes. Several case studies were conducted to verify the developed probabilistic methodology, and the findings are presented and discussed.; Important findings of this study include: (a) the developed statistical methodology enables a appropriate interpretation of the GPR data and, thus, contributes to an acceptance of the GPR technique for bridge deck condition assessments; (b) the presented methodology provides GPR users with a systematic way to accurately quantify the extent of bridge deck deterioration; (c) the results of the experimental study, when incorporated into field data, dramatically improved our understanding of the variability of rebar depth inside bridge decks. It is concluded that this research represents a significant contribution to improving the current practice of assessing the condition of bridge decks.