STATISTICAL EVALUATION OF THE FLEXURAL MOMENT CAPACITY OF BEAMS REINFORCED WITH FRP AND HYBRID FRP-STEEL REINFORCEMENT

Fiber reinforced polymer (FRP) bars are more commonly used as alternatives to conventional steel reinforcement because of their non-corrosive characteristics, especially in aggressive environments. Therefore, a large number of researchers have concentrated on studies involving the mechanical behavior of the flexural strength of concrete beams FSCB reinforced by FRP with regard to using FRP bars. In this paper, the flexural moment capacity of concrete beams reinforced with FRP bars and a combination of FRP and steel reinforcement are investigated. The authors carry out a statistical analysis using a database of 144 beam specimens, including 72 reinforced with FRP bars and 72 with hybrid FRP-steel reinforcement. The experimental flexural strength results were compared to theoretically predicted values obtained according to ACI 440.1R-2015, CSA S807-2012, and a hybrid reinforcement equation, in which the three models were analyzed with respect to their prediction accuracy and reliability. Mean, coefficient of variation (COV), standard deviation (SD), Mean Absolute Percentage Error (MAPE), Root-Mean-Square Error (RMSE), Pearson's r, and Nash -Sutcliffe efficiency index are some statistical metrics which were used to better compare the predictive performance of all the equations. The COV of the developed hybrid equations was 15.27 % and SD was 0.16. Contrarily, model ACI 440.1R-2015 presented such COV of 17.5% and SD = 0.17, whereas model CSA S807-2012 showed a COV of 17.11% and an SD = 0.15. It led to a 12% improvement in the prediction of error compared with both ACI 440.1R-2015 and CSA S807-2012, suggesting better precision and uniformity in predicting the bending resistance of reinforced concrete beams. These results highlight the accuracy of the predictive hybrid model, especially in the estimation of the most important parameters such as elastic modulus, FRP, concrete compressive strength, and reinforcement ratios