Add the cost of re-training, software compatibility testing, a pilot program, etc. and those costs will blow out MASSIVELY.
Anyone in IT worth their salt knows that the software license cost is a tiny part of the TCO or cost to change. There are huge amounts of other costs involved and they are really hard to calculate. Switching platforms is a risk. Switching from XP to say, 7 is a big enough risk with big enough costs and there's a high level of application compatibility there. Switching to ChromeOS? Lol. Even if the software and hardware was FREE, it would still cost money. A lot. A very difficult to calculate number. Business decision makers do not like large, difficult to calculate $ values for risk. With good reason: being able to budget effectively goes out the window.
Also... it is a cost of doing business. We all have the same issues. If you're not going to be bloody careful to isolate it, you are running the gauntlet and need to do a risk assessment and come up with a contingency plan for when it all goes pear shaped. Once you've done the risk assessment, you make the call on what to do. That may be upgrade, it may be isolate until the equipment goes end of life.
Sitting on your hands and whining "waaah it is too expensive" is a cop out - not an action plan. You need an action plan.
You isolate it from the general users production network and the internet and move on. From the sounds of it, that device should be running an embedded OS and should be treated as such.
You no longer have support for bugs, etc. deal with it.
However, you have had better learn for next time that when you purchase a device worth 100k pounds there sure as shit better be some sort of support contract in place. Or you're going to end up in the same situation next time.
"Only the hypocrite is really rotten to the core." -- Hannah Arendt.