ABSTRACT
Section D.2 in ISO2394 Annex D focuses on the evaluation of the coefficient of variation (COV) for a single design soil parameter. This was the main focus for Chapter 3. However, it is more common to conduct a variety of laboratory and field tests, some of them in close proximity, in a site investigation. Data derived from two adjacent boreholes/soundings at comparable depths typically are correlated. Section D.3 in ISO2394 Annex D describes the advantage of adopting a multivariate distribution to capture these correlations in a systematic way. This is the main focus of the current chapter. The practical significance of doing this is that the COV of one soil parameter (or a group of parameters) is reduced when information on a second parameter (or a group of parameters) is made available. The reduction of the COV in the presence of multiple soil tests can translate directly to design savings through RBD. This direct link between the quality/quantity of site investigation and design savings cannot be addressed systematically in our traditional factor of safety approach.
