sorry but the answer is "it all depends (on how you design)" ...
I am personally suspicious of welds and would take some convincing that they're the way to go.
I prefer discrete fasteners ... rivets and screws/bolts; rivets should be arranged so that they're carrying the load in shear, bolts are preferred for tension loads.
Many years ago there was a Fellow named Harry Miller, who designed & built race cars and racing engines.
(fascinating man, well worth a "google" if you are not familiar)
I was in an automotive museum once, and there was a Miller race car on display, from the late twenties. One of the curators happened to be nearby, and I struck up a conversation (being a bit of a Fan-boy) and in talking, I learned that Miller distrusted welding for his frames. They were assembled by laboriously Lapping the elements together, and designing the joints to be retained by reamed & carefully lapped in place taper pins, fitted 90 degrees to one another. The theory being this would remove all relative motion between the parts. All of this is hand work, of course. And everything on the car was "finished", no raw castings, or sharp edges. I asked if the museum, or restorer had accomplished this detailing, and was told "that's how he built it, aesthetics was as important as performance & robustness to Miller."
Older Boeing aircraft [707 720 KC-135 fuselages] utilized spot-welding [extensively] to fasten stiffeners & doublers to skins; and [in some cases] skin-panel to skin panel [lap-splices]. Spot [and seam] resistance welding is elegantly simple for fastening thin skins together without holes and fasteners. Materials used in these acft were 2024-T3/-T42 and 7075-T6 clad sheet metal. Historically SWs held-up fairly well... until fay-surface corrosion caused swelling between skins which overstressed [cracked/tore] the spot-welds in tension. Primary problem with SW'ed assys is that all parts MUST be bare/clean for the SW equipment to produce good quality "weld-spots". After assy, the the assy was dunked in "alodine", then drained/rinsed, then dried, then primed, then edges were fillet-sealed to minimize moisture intrusion. These SW'ed stiffened-skin panels were then hand riveted into position on the frames and longerons. Most skin panel assy lasted +30-years and +80,000-Hrs before corrosion caused [noted] failures... so the "system" was relatively successful [707s and 720s]. NOTE: USAF KC-135s have flown relatively few hours... but the fleet is are now averaging over +45 years old... so corrosion is a huge problem all over the acft.
NOTE.
Spot welds have relatively low shear strength so it takes a lot of tightly-spaced SWs to fasten parts together tightly.
NOTE.
These SW skin panel assys were then rivetd into position. Often, the outer layer [skin] was too thin for the flush-rivet tension-head countersinks; however that didn't stop them from being countersunk!!! The designs allowed the csks thru multiple sheet layers... so many skins were actually knife-edged at the Csk... and relied on the sub-layers to transmit rivet shear loads. Also many of the knife edged csks eventually became moisture intrusion points as the rivets loosened.
Riveting of stiffeners and doublers** to skins/frames [automated assy practices] produces high quality joints with corrosion protective finishes applied to all parts before assy (instead of applied after assy). Where tensile or shear stresses are higher than rivets can take, then lock bolts [swaged-collar, or Hi-Loks, etc] are used.
** NOTE.
Most doublers have been eliminated by the use of thicker skins that are chem-milled extensively.
NOTE.
Fillet and fay-sealant is used "sparingly" in modern assy [except at lap joints, etc]. The modern [precision] fastening produces joints tight enough for pressurization and fluid-tightness; and the pre-applied inorganic and organic coatings (including fastener finishes) provide high corrosion resistance. Good cost-value to minimize use of sealant as much as possible!!! OH Yeah... and wet sealant can mess-up assys, hand-tools, clothing and automated equipment something fierce.
NOTE.
In some acft the skins, doublers and stiffeners are now being adhesive bonded together... which is a whole different discussion.
Sooo... it all comes back-to [to Quote RB1957] "it depends".
Regards, Wil Taylor
Trust - But Verify!
We believe to be true what we prefer to be true.
For those who believe, no proof is required; for those who cannot believe, no proof is possible.
