In reliability analyses of the ultimate limit state design of a spread foundation, the probabil- istic modelling of the calculation model is often ignored. However, as part of any reliability analysis, it is important to consider the uncertainty in the calculation model as well as the uncertainties in the soil strength parameters and applied loads. This paper investigates the model uncertainty by applying a ran- dom variable model factor, M, to the calculation model and examining what level of variation this random variable would need to have to affect the reliability of a foundation design. This is carried out by increas- ing the coefficient of variation of M and observing the effect this has on the reliability index, and on the sensitivity factors, α, which represent the relative sensitivities of the basic random variables. A spread foundation subjected to different loads is examined at the ultimate limit state for drained and undrained conditions. This paper shows that a model factor to account for the model uncertainty is not required in the ultimate limit state design of a spread foundation since the uncertainties in the soil strength parame- ters or the loads in the case of an eccentrically loaded foundation are found to control the reliability of the designs.
[1]
A. M. Hasofer,et al.
Exact and Invariant Second-Moment Code Format
,
1974
.
[2]
Trevor L. L. Orr,et al.
Reliability of shallow foundations designed to Eurocode 7
,
2010
.
[3]
R. Rackwitz,et al.
Structural reliability under combined random load sequences
,
1978
.
[4]
M. Holicky,et al.
Designer's guide to EN 1990 : Eurocode: Basis of Structural Design
,
2002
.
[5]
R. Rackwitz,et al.
Reviewing probabilistic soils modelling
,
2000
.
[6]
Claudia Cherubini,et al.
Reliability evaluation of shallow foundation bearing capacity on c', Φ' soils
,
2000
.
[7]
Minoru Matsuo,et al.
Reliability analysis of shallow foundations in reference to design codes development
,
2000
.