The effects of reinforcing bar diameter and concrete compressive strength on the bond behaviour of glass fibre reinforced polymer (GFRP) bars were investigated. The primary experimental parameters were the type of reinforcing bar (steel, sand-coated GFRP and spiral-wrapped GFRP), the concrete compressive strength (20, 40 and 60 MPa) and the bar diameter (19 and 25 mm). The experimental results revealed that as bar diameter increases, the bond strength of the GFRP bars decreases at a greater rate than that of the steel bars. The bond strength was found to decrease linearly with increasing bar diameter for the GFRP bars while, for the steel bars, the bond strength decreased with increasing bar diameter up to 19 mm and then became constant thereafter. For high-strength concrete, a lower reduction rate in bond strength was observed with increasing bar diameter. It is postulated that an increase in concrete compressive strength causes internal interface (fibre–resin) failure of GFRP bars, while an increased bar diameter results in external interface (resin–concrete) failure.

A novel approach and program are developed for deterioration and predictive modeling of concrete bridge decks based on nondestructive evaluation (NDE) data. Through an iterative process—combined with data processing, bridge deck segmentation, regression analysis, data integration, and deterioration and predictive modeling—the developed program aids estimates of the remaining service life of bridge decks. Data collected on an actual bridge deck during a period of five and half years are used to illustrate the operation and performance of the developed program. Based on evaluation of condition maps, condition indices, and deterioration curves developed for a range of input parameters, the proposed method quantifies progression of deterioration in bridge deck. By reviewing the predictive models, combined with segmentation of the bridge deck area, a more realistic and practical estimation of the deck’s remaining service life can be made. It is anticipated that the proposed method will provide objective and comprehensive evaluation and prediction of bridge deck condition based on data from multiple NDE technologies.