Shrink Swell Index Correlation – Does It Matter Which Method You Use?

Alex Petty

There have been a number of studies completed on the use of different correlations of a variety of soil properties with shrink swell index. They have established some relatively good correlations using linear regression methods. In this presentation, these correlations are taken further to explore the use of multiple regression and simple machine learning algorithms to try and establish more accurate relationships in order to predict the shrink swell index. 

The use of the shrink swell index calculated from various methods and studies, including this one, is then explored with regard to the anticipated characteristic surface movement (ys) and associated Site Classification in accordance with AS2870-2011. The results of four different indicative plasticity properties are studied together with the effect of differing clay content with the same plasticity properties. These results and methods are discussed and thoughts on the applicability of the various methods for different soil compositions is presented. Final comments and discussion will be had on what the results mean in practical terms for geotechnical engineering applications.

About the speaker

Engineers Australia members participating in AGS technical sessions can record attendance on their personal CPD logs. Members should refer to Engineers Australia CPD policy for details on CPD types, requirements and auditing guidelines.