A customer wanted me to solve a particular problem and boss said it is impossible to do because of a particular fact. However, I found a publication which shows that this particular fact is possible to do, and it really wasn't too hard job to do. When I showed this publication to the boss, he said it must be wrong and wasn't willing to listen to details.
It seems like my boss does not want to know what other experts say. How do I handle this?
Well, the boss has decided so the issue is settled then. You have done your job by finding alternatives and suggesting them, his job is to decide what to do. If he chooses to ignore the expertise despite proof, then there is not much you can do.
I'd make sure to get his decision in writing (email) because if you are the one handling the user request there's a chance this could blow up and its good to have proof regarding who decided this.
Also, I'd say this is a pretty big red flag that the job you are at isn't the best; A boss that micromanages and makes decisions above his competence level AND sticks to that decision despite being proven wrong. All signs you might want to make sure the resume is up to date.
On the other hand; There is also the alternative that the boss knows something you don't and is basing his decision on that. Such as the implementation is too expensive, too risky, not enough people want this feature, the product is being discontinued, etc.
Edit: But if there are reasons such as those above and the boss chooses to not disclose them to an employee, that's also a red flag.
To give you an idea what might be going on in the boss's mind:
Results from publications are notoriously difficult to reproduce. You may well be right that it's possible, and the boss may well be wrong that it's impossible. But as a starting ground for coming up with a solution, a published paper often isn't much to go on.
Then, once you've got it working, you have to test it for robustness and discover its limits. Then you need to figure out how to communicate to the customer how it works and what the limitations are, and go through another round of debugging once the customer inevitably discovers a problem.
This not to say that your boss's chosen course of action is the best one or that their reasoning makes sense, just that it's understandable when someone isn't persuaded by one published finding. If you want to persuade your boss, you're going to have to make them understand the solution for themselves. Of course, if they won't listen to details, then you have a whole other problem.
If it's not too hard to do, then it's not too hard to go ahead and test whether the solution is any good - you have to validate through testing because something being in print does not necessarily make it accurate. Going ahead and testing may require the boss's cooperation, though. If the boss is a pretty forceful type who has made up their mind, trying to go through your boss may not be a good career move for you.
You could try having a colleague going through their own boss and test the solution, or you could escalate the attempt to the boss's boss - be very careful how to phrase to the boss's boss if your boss is a forceful type who takes it personally when subordinates try to go over their head.
As you can see, it's a fair amount of effort and initiative on your part. You are going to have to decide whether the problem is serious enough that it's worth it for you to apply this initiative and effort. If it's a trivial matter, I'd drop it for the time being.
If it's not a trivial and in fact, it's a serious matter, recall that the boss stuck their neck out and made the decision against going forward. Live with their decision, it's your boss's neck not yours. I had a CEO who was not above lying and lying pretty shamelessly when things went wrong. Immediately after he made a stupid and arbitrary decision, I would send him a follow up email restating what the CEO had decided and cc: ing those of my colleagues who were in on the action. Those emails kept him honest - just barely.
You must decide when you have done all you can and the battle is no longer yours to fight. In general, I don't obsess about saving people from themselves when they are hell bent on doing what they are doing.
