In terms of design time -- if it's deflection controlled, includes manufactured products, or interfaces a lot with geotech, ASD wins.
If it's concrete, LRFD (or both).
In terms of material -- that depends on your code and loading. In steel, LRFD is considered slightly more economical for LL/DL < 3 (gravity), but slightly less economical for wind lateral designs.
ASD is easier to follow, but LRFD generally gets you more out of your structure. I don't know what you are working on but ASD is starting to not be accepted for new designs. Make sure you check the local or state building codes to see if ASD is still allowed.
Most everything I do is controlled by deflection - so I stick with ASD (except for concrete). I make more money that way as it is quicker to check deflection first.
If I was doing a large steel structure, it would probably makes sense for me to sharpen my pencil a bit and use strength design.
Depends on all the things that Lo and XR brought up from an engineering perspective, as well as the legal/regulatory issues klaus and Rick brought up.
Serviceability is still a concern regardless of which method you use. I have rarely (possibly never?) had to increase a member size to meet LRFD after checking it for deflection. But when I've done LRFD first, I frequently have to increase member sizes to control deflection. If you check ASD first, you rarely have to increase for deflection. Of course, this can be handled by controlling your workflow. Design for deflection first and then check those sizes for your required strength by your preferred method afterward and you'll be okay.
Lo brings up a good point about the Dead Load:Live Load ratio. There is a graph out there somewhere that shows efficiency of each system. Pay attention to that when you do your initial loading analysis, and pick a strength method from there to get the most of it.
One thing to keep in mind - for steel designed per AISC, ASD and LRFD are both Strength Design methods. ASD is Allowable Strength Design, not allowable stress (at least not anymore). The underlying method for determining the nominal strength for the material is exactly the same in ASD and LRFD. The only difference is the statistical manipulation of material deficiencies, poor workmanship, etc. and higher than anticipated loading. LRFD considers them separately - different types of loading will vary differently and the material quality control is generally high at AISC certified facilities; ASD lumps it into a single safety factor.
When I'm designing steel beams, I quickly analyze the beam for both factored moments and shears (ULS of LRFD) and also determine my required moment of inertia using my service loads (SLS). Then pick my beam appropriately. I will agree with the above that deflection governs many a design in steel.
Veer07 - it depends on what you are designing. I'm a bridge guy, and our beams seldom controlled by deflection. In the bridge world we also take advantage of composite action. I have demonstrated several side by side comparisons where ASD requires much more steel, on average 25%-30% than the same design using LRFD. However, in the bridge world where most projects are tied to federal funding, you don't have a choice and have to use LRFD because it's mandated by the feds.
There used to be a "Beam Calculator" program floating around in cyberspace that was AISC based in either ASD or LRFD. You can design a beam in about 3 minutes. I checked it and it is reliable. Google AISC Beam Calculator.
