LRFD Philosopy of Sheet Piles

[VoyageofDiscovery]

This is a continuation of thread255-142962.

It appears the implemention of LRFD for these types of structures is very premature.

see pubs.nrc-cnrc.gc.ca/rp/rppdf/t96-107.pdf

The whole point of this type of change is to reflect the increase in knowledge in our field. The limit state philosophy here leads to more conservative designs. The authors suggest a new approach is required to make it more in line with WSD.

Comments?

 

[jdonville]

VOD,

At IFCEE09 in Orlando, Jerry DiMaggio, late of FHWA, agreed that better calibration of the LRFD factors for many geotechnical problems related to earth retention and slope stability (could be argued they both belong to a meta-set of general slope stability) needs to be undertaken.

Being currently involved with the design of cantilevered and anchored earth retention systems, I do note that the LRFD approach can be very conservative with respect to traditional ASD or WSD practice in terms of an equivalent factor of safety.

However, it has to be recognized that there is no direct correlation between a traditional factor of safety and (the LRFD concept of) reliability.

For example, how reliable is a system with a global FS of 1.25? What is the likelihood that the system will fail this limit state (probability of failure in 100 years)? The answer cannot be directly computed from the "lumped" FS.

Jeff

 

[born2drill]

Does anyone else here see the irony that the LRFD methods need to be "better calibrated" so that they give us the same results that we've been getting for years from ASD designs?

 

[miecz]

Irony, good word. I have to laugh when I listen to all the talk about a "more rational approach" that is then fudged to give the same design as the less rational approach. You either have confidence in your approach, or you don't.

 

[BigH]

Aw - but it gives the professors and all something to "research" . . .

 

[miecz]

I guess I never got it. So what problem was LRFD meant to solve exactly?

a.) Low safety of structures being designed.
b.) High cost of structures being designed.
c.) Irrational approach of structures being designed.
d.) other???

 

[kslee1000]

With the major leap in converting structural steel design from ASD to LRFD (a long journey), I guess everything is geared to be LRFDed, we may one day see Rock Mechanics been under its cover as well. Well, as an engineer, just more factors to play with, with less understanding on the origins. No big deal.

 

[craigmcg]

My biggest problem with designing sheet pile walls with ASD is that the concrete structures they are frequently attached to are designed with LRFD load combinations. Therefore, structures designed using both materials need to be modelled using two different sets of load combinations just to get the actual results. Furthermore, geotechnical data input into LRFD models does not directly compute because an ultimate stress value should be used instead. Geotechnical engineers still work solely on ASD however so there leaves a big gap that needs to be closed between the structural and geotechnical fields.

 

[jdonville]

Several recent papers have proposed that the LRFD factors - especially those dealing with strength - should depend on the variability of the input parameters (sample size, standard deviation, coefficient of variation) or the throughness of the investigation.

The lack of transparency into how resistance or load factors are arrived at is also an issue to be dealt with. However, I expect that the situation is improving and will continue to improve. For example, I am currently working on the design of several retaining structures that will support traffic. The AASHTO LRFD provision dealing with lateral surcharge due to traffic loading does, in the accompanying commentary, cite the published research and briefly describe the methodology used to develop the guidance.

Don't throw the baby out with the bathwater, folks.

Jeff