Seems like there’s two very different issues: first, if pilots are actually showing up to work drunk, that’s bad, period.
But separately: pilots who are self-administering tests in the morning that show presumably minuscule concentrations of alcohol, which nevertheless are non-zero, exceeding the strict zero-tolerance policy, and then grounding themselves over it, because of an extremely strict workplace culture. That’s an example of unintended consequences and overly-strict misapplication of a well-intentioned rule. If Japan Airlines wants their pilots to never drink alcohol, ever, they could make that rule. But it would either drive pilots away entirely, or worse, drive the drinking underground or even into the cockpit. That’s why rules exist at other airlines, allowing pilots to drink in their free time, just not within 8 hours of a flight or with a BAC of over 0.02, for example.
But if an organization imposes a rule that theoretically allows a behavior, but effectively makes it very easy to “fail”, by adopting “zero tolerance” metrics like 12+ hour exclusions or 0.000% BAC limits… of course that’s going to cause disruption.