What is meant by "performance-based"? 2

[DTS419]

IRC 104.11 says "...Compliance with the specific performance-based provisions of the International Codes shall be an alternative to the specific requirements of this code."

What is meant by 'performance-based' and what are those provisions in the Internation Codes?

Does this mean based on empirical testing, or does this mean that designing something to ACI 318 would be considered performance base since it is not prescriptive?

 

[JoelTXCive]

My understanding is that 'performance based' means that if a material or product provider can prove to the EOR's satisfaction that their product meets the intent of the applicable building code provision; then it would be deemed acceptable.

I think that empirical testing would be acceptable, but also other means of proving that a material or product achieves the desired intent in the building code.

Another way of saying it is: "The code says you need to do X, but it is acceptable to do Y if it can be proven that Y will yield the same result"

 

[rb1957]

I think it's a little different. One of our standards changed from being very helpful and defining a load requirement to "performance based" where now they say "the applicant will demonstrate that the aircraft is safe ...". ie the codes change from a specific requirement to one based on the performance of the structure.



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

 

[SE2607]

My understanding is that 'performance based' means that if a material or product provider can prove to the EOR's satisfaction that their product meets the intent of the applicable building code provision; then it would be deemed acceptable.

I think that is "Alternative" Design.

Based on my cursory reading of ASCE-41, I believe performance design is to ensure ductile failure if/when subjected to a design level earthquake. Currently, I believe most applications are for existing buildings which considers a shorter life than a new building, thus a potentially lower seismic load. The design procedure divides structural elements in the lateral force resisting system into two categories: Displacement Controlled Elements and Force Controlled Elements. The Displacement Controlled Elements are expected to get into the yield range before failure and are items such as shear walls and moment frames. Force Controlled Elements are items which are expected to experience brittle failure and are items such as concrete anchors and bolts loaded in shear. The load delivered to these DCEs are the lower of the capacity of the DCEs can deliver to it, or the design load based on AISC-7.

To me, the takeaway is to not overdesign the DCEs. My experience is with wood framed structures, so I will use that as an example. Typically, when I have a line of segmented shear walls of various lengths, I take the total load and divide it by the total length of the shear walls and have the same spec for all of the segments. I know some engineers will even bump up that load to account for eccentricities and other "unknowns". For performance based design, that would be over designing the shorter shear panels and under designing the longer shear panels because the load would be distributed based on their relative stiffnesses. As far as the hold downs go, based on ASCE-7, I do a net uplift calculation based on the load tributary to each shear panel and design based on that load. With ASCE-41, the anchorage would have to be designed such that the concrete anchor remains elastic when subjected to the capacity of the DCE (i.e., shear wall). As you can probably guess, doing an ASCE-41 analysis on a structure based on ASCE-7, most of the anchors would require retrofitting.

I welcome any corrections as I just skimmed ASCE-41.

 

[human909]

To put succinctly:

Performance-based design is a design that might not directly follow the specific code requirements, but can be shown to perform equally or better than the code requirements in achieving the relevant goal.


(Whereas meeting the code generally automatically satisfies. Performance-based design needs to be proven to satisfy, either by engineering, testing or other satisfactory means.)

 

[DTS419]

(Whereas meeting the code generally automatically satisfies. Performance-based design needs to be proven to satisfy, either by engineering, testing or other satisfactory means.)

So would designing something to ACI 318 or ASDS count as a performance-design by engineering even if the design is not tested?

 

[canwesteng]

In the context of residential codes (IRC) I believe so. IRC provides prescriptive requirements for design, you can demonstrate you meet the same level of reliability by using relevant design codes.

 

[JoshPlumSE]

Performance-based design is a design that might not directly follow the specific code requirements, but can be shown to perform equally or better than the code requirements in achieving the relevant goal.

I think Human909's answer is the most accurate one. So, he gets a star from me.

I might elaborate on it a little though as it concerns civil / structural engineering:
Design following the code can be described like baking following a recipe. Use the ingredients in the way the recipe dictates and you will end up cooking an acceptably safe structure. If the structure doesn't perform well, the baker will point to the recipe and say, "the fault is in the recipe, you shouldn't blame me."

However, a good baker, who understands all the ingredients and how they react with each other can create his own recipe for a particular purpose and expect it lead to a sufficiently safe structure. But, if the structure doesn't perform well, the baker must take the blame because he didn't use a recipe.

Performance based design allows for a more "1st principles" kind of design where you do not follow specific code provisions. But, where you evaluate an existing building for safety using more advanced criteria and analysis such that the building can be proven safe. Sometimes this means retrofitting the building to behave with more ductility to prevent anticipated non-ductile failure modes. Or, it may merely demonstrate that the ductility provided by the existing building is sufficiently safe for occupation without upgrade.