Skip to main content

About this book

The book presents the work of the RILEM Technical Committee 249-ISC. Addressing the effective application of new recommendations for non-destructive in situ strength assessment of concrete, it provides information about the different steps of the investigation and processing of test results, until the delivery of strength estimates, and includes tables giving the minimum required number of cores in a variety of situations as well as several examples of how the recommendations can be used in practice. The book explores a topic which is of major importance, i.e. the assessment of concrete compressive strength in existing structures. This property (both mean and standard deviation) is a key input in many cases, such as the reinforcement of structures, the safety checking, the extension of service life.

As the new RILEM recommendations imply a deep revision (and improvement) of field practice, the book is intended for managers of structures, structural engineers and specialists of NDT that have to answer these issues. More widely, it will benefit engineers and students who are interested in NDT and in the safety analysis of structures.

Table of Contents




Chapter 1. In-Situ Strength Assessment of Concrete: Detailed Guidelines

Guidelines describe the general process of in-situ compressive strength assessment. This process is divided into three main steps, data collection (using nondestructive testing and destructive testing), model identification and strength assessment. Three estimation quality levels (EQL) are defined depending on the targeted accuracy of strength assessment, based on three parameters, mean value of strength and standard deviation of strength on a test region and local value of strength. All the necessary definitions (test location, test reading, test region, test result, …) are given and the different stages of data collection, i.e. planning, NDT methods, cores (dimensions, conservation, location, testing, etc) are described. The identification of the conversion model is detailed and a specific attention is paid to the assessment of test result precision (TRP). For the identification of the model parameters, two options are considered either the development of a specific model or the calibration of a prior model. A specific option is also proposed, namely the bi-objective approach. Finally, the quantification of the errors of model fitting and strength prediction is described. The global methodology is synthetized in a flowchart.
Denys Breysse, Jean-Paul Balayssac, Maitham Alwash, Samuele Biondi, Leonardo Chiauzzi, David Corbett, Vincent Garnier, Arlindo Gonçalves, Michael Grantham, Oguz Gunes, Said Kenaï, Vincenza Anna Maria Luprano, Angelo Masi, Andrzej Moczko, Hicham Yousef Qasrawi, Xavier Romão, Zoubir Mehdi Sbartaï, André Valente Monteiro, Emilia Vasanelli

Chapter 2. How to Identify the Recommended Number of Cores?

The concrete strength assessment process is influenced by uncertainties at many levels, including random measurement errors, sampling uncertainty and identification of the conversion model parameters. Therefore, instead of estimating the true value of the concrete strength, it is preferable to say that the objective of the assessment process is to predict a strength value ranging at a tolerable distance from the true strength value. This implies a deep revision of the assessment paradigm, in which both the acceptable tolerance interval and the risk of a wrong assessment must be given at the very beginning of the investigation. A large series of simulations has been carried out in order to understand and quantify how, for a given tolerance on the strength estimation, the risk value varies as a function of the precision of measurements, the number of cores and the strength distribution. Empirical models have been identified from the simulation results. These models have been finally used to calculate how many cores are required in various situations, to achieve the accuracy corresponding to three different estimation quality levels. This chapter describes the principles of the simulation, and how their results were used in order to build a series of tables where the recommended number of cores is made available in a variety of situations.
Jean-Paul Balayssac, Denys Breysse, Maitham Alwash, Vincenza Anna Maria Luprano, Xavier Romão, Zoubir Mehdi Sbartaï

Chapter 3. Evaluation of Concrete Strength by Combined NDT Techniques: Practice, Possibilities and Recommendations

This chapter presents a summarized state of the art of the available methods (SonReb and new methodologies) for the combination of NDT measurements. This synthesis can be helpful for selecting NDT methods for the combination. The chapter also discusses in which condition it is convenient or not to apply the combination of methods. Several possible approaches are detailed in a synthetic way. SonReb method is the most popular method for combining NDT methods, likely UPV and rebound hammer. The use of more than two NDT techniques with the objective of improving the evaluation is presented. This approach of combination can be interesting for example if moisture content varies in the tested concrete or if the quality of techniques is equivalent and if they are complementary. The results of the literature have been presented and discussed for a possible use in the RILEM recommendations. As a first result and without additional computational work, the use of the different approaches is discussed regarding the test result precision of, the TRP, and the magnitude of the final conversion model error.
Zoubir Mehdi Sbartaï, Vincenza Anna Maria Luprano, Emilia Vasanelli

Chapter 4. Identification of Test Regions and Choice of Conversion Models

The main objective of test region (TR) identification is to define an efficient conversion model. The first part of the chapter aims a difficult question, the identification of test regions (TR) because each structure is specific and so it is impossible to give a unique methodology. Here, three different possibilities are proposed. The first one is based on synthetic data obtained on a continuous structure for which TR are identified by means of k-means clustering method. The second approach concerns a real building for which TR are determined by means of two different statistical methods based on the analysis of confidence interval and ANOVA. On three real case studies, the second part of the chapter compares the performances of two scenarios, either the consideration of several TRs and so a conversion model on each one, or the consideration of a unique TR with only one model. The efficiency of each scenario is quantified by the error on the estimation of both mean strength and local strength.
Jean-Paul Balayssac, Emilia Vasanelli, Vincenza Anna Maria Luprano, Said Kenai, Xavier Romão, Leonardo Chiauzzi, Angelo Masi, Zoubir Mehdi Sbartaï

Chapter 5. Identification and Processing of Outliers

When analyzing real data sets, observations different from the majority of the data are sometimes found. These observations are usually called outliers and can be defined as individual data values that are numerically distant from the rest of the sample, thus masking its probability distribution. Outliers require special attention because they can have a significant impact in the concrete strength estimation process and because they may signal the presence of a different concrete population that deserves a separate assessment. The two-step process involved in an outlier analysis (outlier identification and outlier handling) is presented, discussing several statistical methodologies that are available for its implementation. To illustrate the application of an outlier analysis, examples involving univariate and multivariate datasets are presented. Several statistical methodologies are implemented for outlier identification, while outlier handling is illustrated by using robust statistics, i.e. outlier accommodation approaches that reduce the effect of existing outliers on the outcomes of statistical analyses of the data.
Xavier Romão, Emilia Vasanelli



Chapter 6. How Investigators Can Assess Concrete Strength with On-site Non-destructive Tests and Lessons to Draw from a Benchmark

A benchmark is carried out in order to compare how 13 experts define and can carry out an NDT investigation program and derive strength values from NDT measurements. The benchmark is based on simulations, which reproduces a synthetic data set corresponding to a grid of twenty 3m-high columns defining the structure of a building made up of beams and columns. The experts must assess the mean and the standard deviation of compressive strength. Three levels of assessment are considered corresponding to different quantities of test results (destructive or non destructive) available for the experts. The comparison of the various strategies used by the experts and the analysis of results enables the identification of the most influential parameters that define an investigation approach and influence its efficiency and accuracy. A special emphasis is placed on the magnitude of the measurement error. A model of the investigation strategy is also proposed.
Denys Breysse, Jean-Paul Balayssac, Samuele Biondi, Adorjan Borosnyoi, Elena Candigliota, Leonardo Chiauzzi, Vincent Garnier, Michael Grantham, Oguz Gunes, Vincenza Anna Maria Luprano, Angelo Masi, Valerio Pfister, Zoubir Mehdi Sbartaï, Katalin Szilagyi

Chapter 7. How Investigators Can Answer More Complex Questions About Assess Concrete Strength and Lessons to Draw from a Benchmark

This benchmark aims to assess mean compressive strength at several scales and to identify the location and characteristics of possible weak areas in the structure. It concerns synthetic data simulated on a group of four concrete cylindrical structures of identical dimensions with different kinds of strength distribution, based on a real case study. After having received the test results corresponding to their request (non-destructive or destructive), all the experts have to analyze these data and assess the concrete properties and to localize possible weak areas. In addition, they have to define their assessment methodology, i.e. level of investigation, number, type and location of measurements. This study provides information about how the accuracy of the final estimates depend on choices done at the various steps of the assessment process, from the definition of the testing program to the final delivery of strength estimates.
Denys Breysse, Xavier Romão, Arlindo Gonçalves, Maitham Alwash, Jean Paul Balayssac, Samuele Biondi, Elena Candigliota, Leonardo Chiauzzi, David Corbett, Vincent Garnier, Michael Grantham, Oguz Gunes, Vincenza Anna Maria Luprano, Angelo Masi, Andrzej Moczko, Valerio Pfister, Katalin Szilagyi, André Valente Monteiro, Emilia Vasanelli

Chapter 8. Illustration of the Proposed Methodology Based on Synthetic Data

This chapter illustrates first how the recommendations proposed by RILEM TC 249-ISC can be applied in practice. All the steps of the investigation process described in the recommendations are followed and the obtained results are discussed. The second objective is to demonstrate the practical interest of some of the options promoted or recommended in the recommendations. The case study, based on synthetic data, is a four-storey reinforced concrete frame structure. To be representative of a real case study, the synthetic data involves different types of variability, i.e. between mixes (each storey corresponding to a different mix), between components of a given storey and within each component, due to the casting process. Additional variability comes from concrete moisture that can vary slightly around its mean value corresponding to a saturation degree assumed to be 80%.
Denys Breysse, Jean-Paul Balayssac, David Corbett, Xavier Romão

Chapter 9. Illustration of the Proposed Methodology Based on a Real Case-Study

This chapter aims to apply the recommendations of RILEM TC 249-ISC on a real structure, a four-level reinforced concrete framed structure. The flow chart describing the methodology proposed in the recommendations is applied and each stage is detailed. The proposed procedure is compared to the usual practices. The application of the new procedure proposed in the recommendations does not induce significant difficulties compared to usual ones. It is demonstrated that this new procedure can provide mean strength and local strength by saving a significant number of cores. The concrete variability is assessed with a higher reliability compared to usual practices. The unavailability of the test result precision (TRP) in the case study is discussed and its influence on the prediction error is quantified. The definition of test regions by using NDT methods is also proposed. The relevance of test region identification is analysed in relation with the identification of the conversion models. In particular, the possibility to identify either a unique conversion model for all the test regions or a specific model for each test region is discussed.
Angelo Masi, Denys Breysse, Hicham Yousef Qasrawi, David Corbett, Arlindo Gonçalves, Michael Grantham, Xavier Romão, André Valente Monteiro



Chapter 10. Statistics

This chapter gives definitions and basics on statistics, as mean standard deviation, trueness, accuracy, uncertainty … The relation of test result precision with the number of test readings is emphasized. Theoretical considerations about the minimal distance between two test readings to ensure their dependency are presented.
Vincent Garnier, Jean-Paul Balayssac, Zoubir Mehdi Sbartaï

Chapter 11. Model Identification and Calibration

This chapter provides additional information about the “identification of conversion model” step and the “strength estimation” step as defined in the flowchart summarizing the RILEM recommendation. The advantages and limits of the various options are illustrated by analyzing the results of synthetic simulations. Based on the developed synthetic database, a comparison of the performance of different possible univariate conversion models identified and bi-objective method using linear regression is presented and discussed. It is shown here that the issue of the “best conversion model” is secondary, and that the choice of the conversion model has only negligible effects on the final uncertainty of the final strength. In fact, prediction error is the only way to address correctly. It is also shown why the bi-objective approach must be privileged as it provides, without any additional cost, a better estimation of concrete variability, without reducing the performance regarding the assessment of mean strength and local strength.
Zoubir Mehdi Sbartaï, Maitham Alwash, Denys Breysse, Arlindo Gonçalves, Michael Grantham, Xavier Romão, Jean-Paul Balayssac

Chapter 12. For Those Who Want to Go Further

Further analyses and results are presented and discussed in order to provide additional information and justifications regarding the efficiency of the investigation strategy proposed by the Recommendations and these guidelines. These additional results are obtained from an extensive simulation study based on the case study described in Chap. 8 that accounts for the randomness of the assessment process and of the concrete properties. The presented results address the assessment of the Test Result Precision (TRP) level, the assessment of the mean strength and strength standard deviation of concrete, and look at the effect of uncertainty on the values of the conversion model parameters and the resulting prediction error. Furthermore, other analyses address the differences resulting from using different TRPs, from selecting different Estimation Quality Levels (EQL), from using different methods to select the location of cores to be extracted (i.e. predefined coring or conditional coring), from considering the bi-objective approach to determine the variability of concrete, and from combining two NDT techniques.
Xavier Romão, Denys Breysse, Jean-Paul Balayssac, David Corbett


Additional information