The JAIC decided the car deck capacity of 2,000 tonnes of water would not have been enough to sink it, so it theorised from calculations how much more tonnes would be need to have done so.
That's correct procedure.
It then introduced the intuitive idea that the remaining volume of water must have seeped in via decks 4 and five, and the only way it could do this is via the windows.
No, this is an unfair characterization of the process. When the strictly evident sources of flooding do not support the observation of the time taken for the ship to founder, it is not only appropriate but required to look for other sources of flooding.
You seem to think the JAIC was remiss in failing to consider the hole later observed, or the prospect of any such hole. However there was no evidence of anything with which the ship could have collided, and no evidence that there was any collision damage observed on the surface. The window hypothesis requires no unreasonable claims or assumptions. The surface collision theory requires many, and is therefore rejected.
It assumes the windows smashed - as indeed they might have done...
An assumption so well supported by prior experience that it needs no further justification.
but it doesn't actually illustrate this by providing a model simulation based on the MS Estonia's own window specifications.
False. The dimensions of the opening were considered in a flooding model that shows the flood rate from those specific windows was in the vicinity of the portion of the average flood rate not supported by direct evidence. This is sufficient evidence to consider the window hypothesis probable.
Remember MS Jan Heweliusz...
No, you don't get to say one shipwreck is anomalous just because it's not like another shipwreck.
The scientific method as espoused by Karl Popper recommends one begins with the null hypothesis. That is, the null hypothesis here, is that the windows did not break and the excess water did not ingress via the windows of Decks 4 and 5.
No, that's not how science works in the field of forensic engineering investigations. Investigating happenstance events is not generally aided by classical methodology.
In any case, what are we to do with all the numerous conspiracy theories you've proposed as viable alternatives to JAIC's findings? Are they testable in the way you demand should have been done here? Were any of them tested to that same standard of proof you say is required? Are you not applying a double standard?
You made a huge deal out of Prof. Amdahl, who you allege computed the size of vessel that would be needed to poke Evertsson's hole in the hull. You seem to consider this dispositive science, although it proceeded exactly along the same lines as JAIC's reasoning regarding the Deck 4 windows—only with far less rigor. Prof. Amdahl proceeded under the assumption given to him by Evertsson that the hole was produced by a collision with a vessel on the surface. It is certainly possible to calculate the properties of such a hypothetical vessel, but there is no evidence any such vessel existed. Thus Prof. Amdahl's findings fail not at the computational level but at the conceptual level.
In contrast, JAIC's findings here are conceptually sound because there is ample evidence that wave action is sufficient to break the glass in ship windows, ample history of it having done so, and ample evidence that the windows in question were in a position to be subject to wave action.
You then set out to reject the null hypothesis by setting up a simulation using the same type of reinforced glass as used in the MS Estonia and under the same conditions as recreated in a laboratory, obviously, making everything proportional to fit. You then get your results completely objectively and analyse them later. Not decide that X tonnes of extra water was needed therefore your hypothesis is proven even though you didn't follow the scientific method.
No. Your imagination of what proper science must be in this case is not informed by education or experience. You're just making stuff up, pretending to be an expert, and trying to foist it on your betters.
Setting up a full-scale empirical test when it is known generally that waves have the ability to break ship window glass and have done so numerous times is costly and unnecessary. Further, the parameterization of such an experiment requires values that were not measured and cannot be known with certainty. Thus any marginal result would come down to a belief in a specific resolution of the uncertain values. Therefore it just kicks the can of unknowability down the road a bit and offers no probative value. Your belief that such an experiment is necessary seems predicated on your uninformed belief that ship window glass is generally too robust to be vulnerable in this way, and that the JAIC is remiss in not believing similarly. Your belief amounts to nothing more than your typical, "Because I say so," contrary to established fact.
You really need to learn to stay in your lane.