Guide to non-destructive concrete strength assessment: Homogeneity tests and sampling plans

Abstract

Non-Destructive testing (NDT) of concrete can be used in conjunction with destructive techniques (e.g., coring) for an efficient assessment of concrete strength in existing structures. A major drawback that hinders the efficient use of test measurements is the limited description specified by the relevant standards regarding how the overall investigation program should be carried out. This study discusses two elements of the investigation program: homogeneity of different concrete zones and the locations for core extraction. Different homogeneity measures have been investigated to identify heterogeneous concrete zones using extensive statistical analysis of synthetic datasets. The considered homogeneity measures were: t-test, Kolmogorov-Smirnov (KS) test, earth mover's distance, and Cramér-von Mises distance. The results of the analysis indicated that the earth mover's distance and Cramér-von Mises distance outperform the traditional t-test in the identification of heterogenous concrete zones, and hence, recommended to be used for concrete assessment. Furthermore, different sampling plans of core locations, namely, similarity sampling, cluster sampling, proportional stratified sampling, and post-sampling are investigated through the generation of risk curves as indicators of the estimation quality of each sampling plan. It was found that a reduction in the probability of inaccurate concrete strength estimation is feasible. As such, the utilization of sampling plans in concrete assessment can enhance the reliability of the assessment.