Non-destructive assessment of both mean strength and variability of concrete: A new bi-objective approach

• Bi-objective approach shows high efficiency in capturing the concrete strength variability.
• Regression approach has limited ability in capturing the concrete strength variability.
• Calibration approaches cannot be used to estimate the concrete strength variability.
• Bi-objective, regression and calibration approaches can efficiently assess the mean strength.

Using non-destructive techniques (NDT) like rebound hammer in combination with destructive techniques (DT) like core test is a common practice. Two approaches are widely used to produce a model for assessing the concrete strength. The first approach consists in fitting a specific model between NDT measurements and cores using the regression analysis. The second approach uses a prior model which is calibrated according to measured core strengths. The EN 13791 and ACI standards require a large number of cores to estimate mean concrete strength and concrete strength variability and consequently to calculate the characteristic strength value which depends on these two inputs. In this work, we propose a new approach for identifying the models based on NDT and DT tests in order to capture both mean strength and concrete strength variability. This approach is first illustrated by synthetic simulations which are a good way to study a problem having many degrees of freedom. The proposed approach is then tested on a real data set. In both cases, it is confirmed that the common approaches are able to estimate the mean strength but they fail, even with a large number of cores, to accurately estimate the concrete variability and hence the characteristic strength. Reversely, the new approach shows its high efficiency in capturing the concrete variability (in addition to the mean strength) with a number of cores lower than that prescribed by the standards.