MSP usage increases with maintenance clock, with plenty of engineering spaces you get by with very little for years.
It's now attractive to use excessive engineering spaces, offload MSP to new ships once things start to fall apart, and scrap the ship.
Probably never bothering with overhauls or in-orbit maintenance.
I agree somewhat. In orbit maintenance cost MSP by size, not by failure rate, and as such, is worse than engineering space. And kinda exclusive with each other. There is no point in reducing failure rate if you pay by size anyway, so ships staying in orbit of maintenance facilities do not need much engineering. On the other hand, ships with a lot of it should avoid orbiting because they'll cost you a flat rate despite their low failure rate. Having to avoid maintenance facilities when optimized for long missions seems weird to me.
MSP cost of maintenance when in orbit should maybe depend on ship failure rate? It wouldn't solve the maintenance storage issue, but it would make orbital maintenance less counter intuitive in that crafts designed for reliability should cost less to maintain, be it in orbit or in space.
To solve the maintenance storage issue, adding stacking penalties to engineering modules when above a certain rate compared to ship size would probably help. Using more engineering space should be design-costly and valuable mostly long term.
I disagree with you about overhaul though. If you do it early enough (before things start breaking too much), even a craft with lots of engineering could benefit from it.
EDIT: unrelated, about particle lances:
I feel being bigger should maybe increase range of the beam, rather than just damage. Right now, it's just a gun with twice the damage for twice the size, but doesn't really offer new tactical opportunities compared to particle beams. Being x1. 5 range x1. 5 damage could possibly be more interesting. Just my opinion.