"steel design" is a fairly broad category, I'm sure there are many instances of steel design where LRFD would not be appropriate/applicable. First example that comes to mind is if you're designing lugs or lifting gear with ASME BTH-1.

CANPRO (Structural), Thanks for response. May I ask has ASME BTH-1 clearly banded use of LRFD method for lifting gears or lugs design?

The ASME BTH-1 is based off of ASD checks, and therefore that is why you would not use LRFD to design stuff to that standard.

LRFD is only going to be applicable to the design of building structures or structures similar to buildings. Lifting lugs, pressure vessels, vehicle chassis, and other distinctly non-building structures would be an inappropriate application of Load and Resistance Factor Design. At least as spelled out in AISC 360. I imagine the general principles could be applied to other areas, but you would have to take a long hard look at the statistics used to derive the various factors, the level of reliability required, applicable failure modes, etc.

The pressure vessel and tank industries, possibly the piping industries, are allowable-stress, generally.

I'd say that anything that is a "serviceability" issue tends to be better dealt with using ASD..... deflection, vibration, etc. But, I don't think there is anything that prohibits the use of LRFD for the basic design and then service level loading for the serviceability limits.

As boiled down as I can think of: LRFD (in a sense) puts more emphasis on the 'uncertainty' of the applied loads, and more trust in the strength of materials and quality of workmanship, etc (as we would like to imagine in modern construction). ASD generally says that we know, more or less, what the applied loads are going to be, and discounts the strength of materials and quality of workmanship. For example, LRFD puts more 'safety factor' on live loads than it does for dead loads. This is generally applicable when designing buildings as there are both dead and live loads. When designing a lifting lug or something, there is pretty much just 'a load' being applied. So the LRFD load factors are not quite applicable to such a system as they would be in a larger structure - "is my lifting lug carrying a live load or a dead load", vs. "I have this load I need to support, lets take some capacity off the material side of this (i.e., safety factor). Hopefully that gives a better idea of why LRFD might not be appropriate for that type of component based design.

I suspect over time, LRFD will be used for lifting lugs, pressure vessels, etc. It has a more overall uniform treatment of stresses and forces. For specialised items, serviceabilty issues will be included, and this can be reflected in stresses, etc. In Canada, we've 'messed things up a bit'. LRFD is used with the same load factors for Concrete and Steel. They've fudged with the material property factors to give similar results that the old reinforced concrete load factors of LL=1.7 and DL=1.5 had, rather than deal with the serviceability. This is an added level of complexity that has been 'fudged' away by modifying the mpf. With the old ultimate strength design for concrete, you were pretty certain that the calculated ultimate strength was pretty close to test results... not so sure anymore.

Dik