Speaking as someone who worked on the behaviour of metals under strain, the problem for is that "hrs" are merely an approximation of the life of the component.
What really determines the life is "seconds at x kg loading (and perhaps at temperature T deg mas well)" ,called the strain history, and that will give a finite life to identical components - in aluminium anyway. steel gets complicated by the existence of a yield point.
now fudge for manufacturing differences and defects and corrosion, apply a rule of thumb or Two, do some testing of ultimate loads and voila! We convince ourselves that there is a 99.95 percent chance that the component will last "x" thousand hours.
Obviously in the case of the Dromader either the assumptions were wrong or the "allowable "strain history of the components was exceeded.