This has been repeated so often that it's almost a cliché at this point. At the end of the day, all personality typing systems have the same flaw: they're artificially constructed categorizations based on self-reported data.
OCEAN is also "pseudoscientific" in that sense; who's to say that the right way to boil down someone's character is to measure openness, conscientiousness, extroversion, agreeableness, and neuroticism? What do these terms even mean? What questions could you ask to accurately assess them? Not to mention that there are hundreds of other terms that could've been chosen, those precise five are somewhat arbitrary. People will also get different results everytime they take a Big Five test, because that's just the nature of asking people questions about themselves.
None of that means that it can't be a useful tool. OCEAN tends to be good for population-level studies and has the advantage of placing people on a spectrum, instead of into discrete boxes, just like how MBTI tends to be good for understanding how people make decisions, and the Enneagram tends to be good for understanding what motivates people. The question isn't whether one system is "better" or "more scientific," it's "Which screwdriver do I need for this screw?"