Seminar - Gordon A. Fenton - Reliability-Based Geotechnical Design

Tuesday, April 19, 2005, 1.10 - 1.50 pm
Civil Engineering Lecture Room 3

Abstract

In an effort to harmonize with structural codes, geotechnical design codes around the world are beginning to migrate towards some form of reliability-based design (RBD). Significant steps in this direction can be found in, for example, Eurocode 7 (2003), Australian Standards AS 4678 (2002), AS 5100 (2004), AS 2159 (1995), NCHRP Report 507 (2002) and the Canadian Foundation Engineering Manual (1992). These RBD provisions are most often presented in the form of a Limit States Design, to define critical performance states, combined with load and resistance factors which are calibrated to achieve the target reliabilities associated with the various limit states. The use of load and resistance factors is generally referred to as Load and Resistance Factor Design (LRFD).

At this time, reliability-based design provisions in geotechnical codes are primarily based on calibration with older codes, and for good reasons one of which is that calibration is a useful way to incorporate years of practical experience into a code. Another is that soils and their mechanical behaviour are much more difficult to characterize probabilistically than are the other quality-controlled materials commonly used by engineers (e.g. concrete, steel, wood). However, in order to take advantage of the full power of a reliability-based design methodology, the design provisions should incorporate probabilistic methods of describing soil behaviour.

This talk traces the development of reliability-based design in geotechnical engineering, and how it has been implemented in a variety of design codes around the world. Then, some of the significant stumbling blocks to going beyond calibration from older codes and introducing some of the real advantages of a reliability-basis for design are pointed out. Finally, some of the possible ways in which reliability-based geotechnical design can be improved, and the headway being made towards these improvements, are explored in the context of common geotechnical design problems, such as slope stability, bearing capacity, and settlement.