Why shipping breaks things that use never did
Transport loads, and why they differ from operating loads.
A product can be engineered thoroughly against every load its normal working life is expected to throw at it and still arrive at its destination cracked, loosened or dead on arrival, because the single journey by road, rail or air that delivers it can subject it to a shock and vibration environment considerably harsher, and differently shaped, than anything its actual operating life was ever designed against.
Why a transport journey can be the harshest load a product ever sees
A designer sizing a part against its working life almost always starts from how the product will actually be used, the forces it will carry, the speeds it will move at, the cycles it will repeat, and builds a load case around that picture of ordinary operation. A truck bed, a shipping container stacked several units deep, or an aircraft's cargo hold subjects the same product to a completely different kind of loading altogether, one that has nothing to do with how the product will eventually work and everything to do with how roughly it is being carried while it is not working at all. Road surfaces feed a more or less continuous, randomly varying vibration up through a vehicle's suspension and into whatever is strapped down in the back, potholes and expansion joints add sharp individual jolts on top of that continuous shaking, and a box handled by a forklift, a conveyor or a person in a hurry can be dropped, tipped or stacked under weight in ways no operator would ever treat the finished product once it is actually in service. None of this appears anywhere in a typical operating load specification, because none of it happens while the product is doing its job, and a part that comfortably clears every load case its working life will ever throw at it can still be the first thing to crack loose during the one trip needed to get it there. The trip itself is also, in an important sense, unrepeatable in a way ordinary operating loads are not: a product might be switched on and off thousands of times over its working life, giving a designer every opportunity to learn from a marginal component that starts failing early, but it is usually shipped exactly once, so a transport failure mode has no equivalent early-warning history building up behind it before the first unit that actually fails arrives broken at a customer's door.
The moving-van comparison
A china plate can sit stacked safely in the same kitchen cupboard for years without a single chip, opened and closed on hundreds of ordinary mornings with no incident at all, and then chip badly the one time it travels a short distance packed into the back of a moving van. The damage rarely comes from any single dramatic knock, it comes from the van's continuous low-level vibration over the whole journey slowly walking the stacked plates against each other, a tiny relative movement repeated thousands of times over the length of the drive until two edges that started out perfectly aligned have worked their way into contact hard enough to chip. The plate was never subjected to anything like this in the cupboard, where it sat still and undisturbed between uses, and the van ride is not simply a rougher version of ordinary cupboard life, it is a genuinely different kind of loading the plate's years of safe storage never once tested it against.
Why vibration during transport differs from vibration in service
A product's normal operating vibration, where it exists at all, is usually fairly narrow and predictable, a motor turning at a known speed, a fan spinning at a fixed rate, a mechanism cycling through the same repeated motion, all of it concentrated around one or two frequencies the designer already knows about and has likely designed around. Transport vibration is broadband rather than narrow, a genuinely random mix of energy spread across a wide range of frequencies simultaneously rather than concentrated at any single one, because a truck's engine, its tyres, the road surface and every joint and panel in the trailer are all contributing their own frequency content to the ride at once. Somewhere inside that broad spread of frequencies sits the product's own natural resonant frequency, the one at which a gentle push builds into a large swing the way pushing a swing at just the right moment does, and because transport vibration covers so much of the frequency range at once, it reliably finds and excites whatever resonance a product happens to have, amplifying a comparatively mild input into a locally severe one exactly at the frequency the product itself is least able to resist.
The one number worth remembering
A product's structure amplifies vibration most strongly at its own resonant frequency, and because road and air transport spread their vibration energy across a genuinely broad range of frequencies at once rather than concentrating it narrowly, a part that would never meet a matching frequency during its actual working life can still meet one, and get shaken correspondingly harder, purely by being loaded onto a vehicle for delivery.
What follows from this
A design that has only ever been checked against its own operating loads has been checked against the wrong load case for the specific hazard of getting from the factory to the customer, and packaging alone cannot always be relied on to close that gap, since padding reduces shock but does comparatively little against a sustained, broadband vibration that keeps finding a resonance no amount of soft foam has removed. The practical answer is to test a product, or at least its more fragile subassemblies, against a transport vibration and shock profile deliberately built to resemble what a real truck, aircraft hold or parcel network actually delivers, rather than assuming that a part strong enough for its working life is automatically strong enough for the far rougher and far less predictable ride that comes before that working life ever begins. Fasteners deserve particular attention in this light, since a joint that never sees a single load cycle during years of stationary service can still see thousands of small ones during one loud, vibrating truck journey, and a bolt that would have held indefinitely under its actual working load can arrive already loosened, or already fatigued, before the product has done a single hour of the job it was actually built for.