Analysis of the accuracy of in-situ concrete characteristic compressive strength assessment in real structures using destructive and non-destructive testing methods

•The new standard version procedures (EN 13791-2019) are applied in a real case study.•Effect of the improved procedures on in-situ concrete strength assessment is analysed.•In-situ strength assessment accuracy is analysed using Monte-Carlo simulation process.•In-situ characteristic strength assessment is compared with of the old standard version (EN 13791-2007).•The statistical procedures of in-situ characteristic strength seem to be conservative. Estimation of in-situ concrete compressive strength in existing structures is considered as a crucial step in assessing structural capacity. Destructive testing (DT) by coring in conjunction with non-destructive testing (NDT) offers an interesting method for concrete strength assessment in existing constructions. The non-destructive assessment of in-situ strength requires a correlation between NDT measurements and the core compressive strengths. Therefore, the accurate quality of the in-situ strength assessment is a fundamental point. The NDT techniques such as Rebound Hammer (RH) and Ultrasonic Pulse Velocity (UPV) are widely used separately or in combination to estimate the in-situ strength. A series of on-site data (205 data triplets) of non-destructive testing (RH and UPV) and coring have been carried out on structural elements (columns and beams) in an existing building to assess the in-situ concrete strength according to the new standard version procedures (EN 13791-2019) with a comparison to its old version 2007. The analysis highlighted in a real on-site situation the positive effect of the new procedures in the recent version (2019) of EN 13791 on strength assessment, namely: the selection process of coring positions, the required number of cores, and the identification method of the conversion model. However, the statistical procedures proposed in the new version for calculating the in-situ characteristic strength appear to underestimate the results and seem to be excessively conservative as compared to the 2007 version.