Numerical methods are increasingly used to predict ultimate limit state stresses in underground structures. Considerable experience has been gained in the UK of design and construction of SCL tunnels in soft ground, particularly under London. The urban setting means that extensive monitoring is used to ensure the safety of existing structures, and therefore a wealth of monitoring data has been gathered and much of it has been published as case studies.

Verifying that the results of a numerical analysis are reasonable is a necessity, and these case studies of previous experience are commonly used in one of two ways in design. Firstly, the results of a numerical analysis may be directly compared to case studies of previous tunnels constructed in similar material at similar depth and with a similar construction method. This will provide a check that the results of the design analyses are reasonable. Secondly, a backanalysis of a case study may be used to validate the method of analysis itself. As long as the case study is similar to the tunnel being designed, this is a reasonable assumption but there are pitfalls to this method.