Criteria for comparing computed and observed results in geotechnical engineering

A number of criteria for comparing computed and observed results in geotechnical engineering are investigated. These criteria, which are based on different types of statistical analyses, involve the use of regression models, the use of the root mean square error and standard error of the bias as significant indexes, and the use of the ranking index and ranking distance, which are related to the accuracy and precision of the calculation model. The application of these criteria is examined by investigating their use for comparing computed and observed total and relative settlements of pile foundations. On the basis of this investigation, the authors conclude that the ranking index and ranking distance offer the best criteria for comparing computed and observed results in geotechnical engineering.