I guess an easy way to think of it would be material cost alone:
You can chose to pay ground freight or overnight airmail. Obviously shipping concrete and steel parts airmail is going to be prohibitively more expensive than freight. But you are paying for the speed through the extra costs. There’s a story of a man hired to build a bridge out in California. He was told he would get some insanely large bonus if he finished ahead of schedule and passed all the inspections. He hired helicopter crews to air haul huge sections of pre-assembled bridge sections cross country at exorbitant rates to finish for the maximum bonus (it was prorated.)
He still made a profit, but the costs were crazy.
Also, keep in mind that one crane is usually shared for many jobs as are specific crews, and managers etc. If you are doubling the speed of production on one job, that detracts from their abilities to do their other jobs (because they are spending more than twice the time on the rush job.)
For instance, once the sections of a boat are laid in place by the crane, you really don’t need that crane any more until that section is anchored, wired, plumbed, welded and riveted in place so that crane can go to the next boat and the next. But at double time pace, that crane’s going to have less time to make the rounds which means you would need another crane to pick up the slack.
In the biz we have something called float time. The maximum time we expect a section of the critical path (that would be the series of events that if delayed, delay the entire project and if sped up can speed up the entire project) and the minimum time we expect a section of the project to be completed. During the “float” (or the difference between the two) we schedule those workers not immediately being used for that part of the project to other projects so we can maximize their efficiency.
By rushing the order you are taking out all the float time from the critical path (this usually results in the creation of new paths or additional critical paths) and you are taking those resources away from other projects which could also create more critical paths or new critical paths there.