A comparison between "measured" and "calculated" values in geotechnics - An application to settlements

Computation models used in geotechnics are usually affected by error. Estimations of settlements of shallow foundations in cohesionless soils on the basis of the results of in situ tests, particularly the Standard Penetration Test, can be made by different methods. Each of the methods has been developed and/or calibrated according to specific considerations, to which can be added the specific typology of soils covered in the analysis. Only recently, efforts spent in attempts made to evaluate successfully, on the basis of data from observed settlements, the precision, accuracy and reliability of the many available computational models, have met some success in terms of acceptance by professional circles. This paper reports on statistical/probabilistic elaborations and discusses the correlations between measured and calculated settlements (or viceversa) and thereby presents circumstantial judgements about the positive aspects of each method as compared to the other methods considered in the analysis.