Structural failure probability of a steel structure 1

[KAMAL HOSEN]

For my school assignment, I need to calculate the structural failure probability. i can use Staaad.pro. Can someone tell me or share some reffrence how can I do that by using staad.pro in an easy and simple way. ( this is an assignment for my class. there is an existing steel structure. I have analysis the structure by Staad.pro. Considered gravity load, seismic laod and wind load. all members are found safe. PMM ratio >1, also safe in deflection. My professor asked me to present the failure probability. )
Thanks in advanced

 

[Luceid]

This is a very broad question. Failure probability due to what?

 

[KAMAL HOSEN]

this is an assignment for my class. there is an existing steel structure. I have analysis the structure by Staad.pro. Considered gravity load, seismic laod and wind load. all members are found safe. PMM ratio < 1, also safe in deflection. My professor asked me to present the failure probability.

 

[3DDave]

To start you need historical data on the sorts of failures that are anticipated. Then you need a time frame to evaluate. There is a 100% chance every structure will fail if one waits long enough. Then you need to account for whether the item is subject to being maintained, which can add some failure modes while offsetting others.

So if you have a structure of 2 parts, each with a 50% chance of failure in 1 year, then you take the 50% chance of not failing for the first and multiply that by the 50% chance of not failing for the second to find that there is a 25% chance of not failing or a 75% chance of failing.

Smaller failure chances times far more items give the structural answers. Not only does each part have a failure chance, and each fastener, but the installation of the fastener is a failure source - good bolt badly installed.

But without historical failure data, you cannot start except to demonstrate what you would do if you did have the data.

 

[rb1957]

failure due to overload is one thing, maybe consider failure due to inservice issues (like maintenance not detecting corrosion or fatigue cracks) ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Luceid]

Or maybe failure based on the likelihood of a seismic event greater than what you have designed for?

 

[Luceid]

Same goes for wind

 

[KAMAL HOSEN]

A seismic event is okay. Can you send me any references , how to do that? I want an easy and sample method. as it's a school project, my teacher ask me to choose an easy method.

 

[rb1957]

then that seems to be more aligned with "how is the severity of design load cases determined ?" ... what is the probability that the design loads will be exceeded ?

or (another way to ask/answer the question) how does structural strength degrade over time (corrosion, maintenance issues) so that the structure can no longer support the code defined loads ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[SWComposites]

Kamal -
There is no example to provide because no one calculates the "probability" of structural failure for buildings (or most other things), because a) it is not required by code, b) there is no data to feed into a probability analysis, c) lawyers would crucify anyone dumb enough to try to justify a non-zero probability in a case where someone got injured or killed.
There are various "probabilities" baked into some design code factors, but getting to the background for those requires digging deep into research results and code committee documents; something that no practicing engineer does.

 

[driftLimiter]

Seems to me that the heart of this question is related to design methodology. In the US we use LRFD and ASD design approaches. The load combination factors, and strength adjustment factors are calibrated using a statistical approach to provide consistent levels of reliability.

OP could look into this further. Link

 

[rb1957]

that I think is an excellent simple answer ! 1 pix, a bunch of words ... lick o'paint !

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[KAMAL HOSEN]

@ Drifgtl imiter . I checked the PDF, I think yes, this is the one that I was looking for. DO you have any other examples or Ref. from there I can learn how to calculate these value.

 

[driftLimiter]

@kamal

Unfortunately I do not. As mentioned by others on this thread, this is something that researchers have developed over the years, and many engineers don't know the specifics of it (me included).

But I think if you research the history of LRFD for example you will be in the right subject matter.