Robustness analysis of on-site assessment for concrete strength based on Monte-Carlo simulation

The strength assessment of concrete materials constitutes a critical component in the performance evaluation of existing reinforced concrete (RC) structures. Owing to the high variability of concrete materials, a reliable assessment typically requires comprehensive judgment based on a limited number of core strength measurements combined with rebound test results. However, for concrete of different strengths, key parameters in existing evaluation methods, such as the number of cores and the selection process of coring locations, still lack reliable guidance. Additionally, the criteria for selecting conversion models that determine the relationship between concrete strengths and rebound values remain unclear. In this study, a dataset containing 270 pairs of rebound values and core strength measurements was established based on six 16-story RC frame-shear wall residential buildings in a residential complex in Beijing. Using the Monte-Carlo method, this paper analyzes the impact of the number of cores, selection process of coring locations, conversion models, and other factors on the accuracy of existing concrete strength evaluation methods at different strength levels. The results indicate a significant negative correlation between concrete strength and model accuracy. Furthermore, adopting conditional coring and increasing the number of cores contribute to improving the robustness of the model. This research provides practical guidance for the performance inspection of existing structures in engineering practice.