Safety factor of 1.2 to 1.5 is well recognized in many codes. In general there is no guarantee in Seed and Idriss technique as their technique is dated and failed to address many recent findings. So safety factor of 1.5 sounds sensible. A less value might be warranted in the case of assessment based on effective stress analysis.

Thanks a lot for your answer. It would be very useful if you can recall a reference for a certain deterministic approach that states the factor of safety,

Also, it will be very helpful if there is a reference relating this factor of safety to the deign levels of earthquakes (operating-based OBE and maximum-credible MCE)?

In my view, our understanding about liquefaction is not mature enough to find its way to codes of practice. For example, traditionally, it was believed that clean sands are the most susceptible soil type. Recent studies, however, proved that in fact silty sands are more susceptible compared to clean ones...!

Most codes have no provision regarding a safety factor below which liquefaction would definitely occur.

Building codes are quite weak in this regard (they jumped to another section when it comes to liquefaction or made some general obvious comments...!)

Bridge codes have some implicit provisions. For example you can find this in FHWA (Sec. 3.2.4.1):

"The ratio of CRR to CSR should be greater than 1 to preclude the development of liquefaction. Eeven when the ratio of CRR to CSR is as high as 1.5, increase in pore water pressure in the soil can occur."

Again you can see that no exact value was stipulated by the code.

As you know Japan is very advanced in liquefaction. So you can find a rather explicit procedure in OCDI (Sec. 13.2). However, the introduced procedure in OCDI differs with that of Seed and Idriss.

So current codes accept engineering judgment and you can use a safety factor of more than 1 in practice. Most people use a value of 1.2 to 1.5, depending on soil layers, seismicity of the region, and intended performance of the structure (importance of the structure).

FHWA: Seismic retrofitting manual for highway structures: part 1-bridges, 2006, Federal highway administration.

OCDI: Technical standards and commentaries for port and harbour facilities in Japan, 2002, by the overseas coastal area development institute of Japan.

Many thanks. There is always a debate whether to go to a factor of safety of 1 or to increase it to say 1.5. The costs of ground improvement works to reach a certain safety factor is huge and they substantially increase with the increase of safety factor. In fact one may opt to go for a factor of safety of 2 if you want to minimize the settlement associated with liquefaction.The currently adopted two-levels earthquake design is another issue that really need to be considered while assessing the safety factors. I would like to explore if there is a certain reference that gives the factor of safety with reference to OBE and MCE.

The most common modes of failure for embankment dams generally fall within the following categories (Foster et al. 2000):

• Overtopping 48%

• Piping 46%

• Slides 5.5%

• Earthquake liquefaction 1.5%

As such, liquefaction is a real contributor to dam failures. However, the trigger for the event is an unusual or extreme event (depending where the dam is situated. it is situated. In Canada, for deterministic dam safety analyses, the Factor of Safety used is selected on the basis of the likelihood of occurrence of the initiating event. For example, under normal operating conditions a factor of safety of 1.5 is selected. For more unusual and extreme events lower Factors of Safety are accepted. In this way, the Factors of Safety are used as a course proxy for risk.

The 2007 CDA Dam Safety Guidelines require the factor of safety for a dam subjected to earthquake loading conditions to be greater than 1.0 during and subsequent to the event. This recognizes that the event is infrequent and is an instantaneous load (albeit it a number of aftershocks) such that, following the event, there would be time to undertake the remedial measures needed to bring the factor of safety for the structure back up to 1.5 (under the normal loading conditionsdefine factors of safety for the assessment of embankment dams under earthquake loading conditions as listed below

The 2011 Ontario Dam Safety Guidelines (attached) offer similar guideance

Many thanks

I also agree with all the previous answers. As for the factor of safety, it is already implied in the selection of the design earthquake, this is why 1 is usually good if you have selected your earthquake ground motion sollicitation properly. For dams and buildings in Canada, from the CNBC 2010 and the CDA, the recurrence of the earthquake is 2% in 50 years of exceeding (1 in 2475 years) which is usually a very strong earthquake.

If you look in the CDA safety guidelines 2007, table 6.3, the seismic FoS in pseudo-static is also 1.0 and post-earthquake is 1.2-1.3.

Regards,

For buildings we usually use MCE of 10% in 50 years (return period = 475 years) which is much less than Dams (return period = 2475 years),

May I query your opinions on how this would affect the factor of safety?

In the Dam Safety Industry the Factor of Safety is independent of the selected return period of the initiating event (i.e., an earthquake or flood event).

The return period of the event is a proxy for the probability of occurrence. It is selected on the basis of an assessment of the consequences that might occur in the event that the initiating event caused a dam failure. For example, in situations where the failure of a particular dam could result in excessive loss of life (say in excess of 100 persons) or excessive economic or environmental losses are possible, the 1:10,000 year event is selected. In situations where the failure of the dam causes negligible environmental or third party losses, an event as frequent as 1:100 might be selected.

In this way, the dam safety industry attempts to incorporate risk awareness in the Standards Based Design process. Given that the selected return period (the proxy for probability of occurrence) varies depending on the potential consequences, the idea is that the risks are roughly equivalent. Therefore, the same level of safety (FOS) is necessary.

From my (short) experience, I would say that the CDA does take into consideration the impact in loss of life or in environmental or economical impact, but not the CEHQ in Quebec. Its always 1 : 2475 years for all dams, regardless of the impact. This is safe but rather conservative for very small dams with litle or no impact and can have a great influence on the rehabilitation cost. As it was rightfully explained by M. Donnelly, the same level of safety is necessary.

Thanks a lot. Do you recall similar references or codes for buildings (under OBE and MCE)?

1.35

Geotechnical Engineering