The greatest barrier to using fabs in FA is their initial cost, and by cost I mean the associated cost of the power generators necessary to run them. They're a major investment that almost never translates to as much economic leverage as upgrading MEXs, MStorage, RAS, RAS SCUs. Fabs have one strength in that they can be built where the player chooses, in this way they can be protected and as such merit their low hp and volatility. Lowering Fab cost to something like 2000e [down from 4000], 100m and moving them back to T1 would be interesting and much more reasonable. Remember that the cost of increasing your mass output by 1 then would be tied to the cost of 1 MFab + the cost of 8 T1 Pgens [which is 6000e, 600m]. So these changes are altering the real energy cost of fabricating 1 additional mass by considerably less than 50% of its 4000 cost, instead the energy cost of 1MFab + 8Pgen is going from 10000e to 8000e, or 20% less energy cost.
T3 MFabs are completely broken, they are less efficient than T2 Fabs in every way. Their production should be 18m [up from 12], their cost would change to 50000e [down from 65000], 1800m [down from 3000], and their consumption would be 2600 such that it is slightly more energy efficient than 18 lower tier fabs (which would make the same mass at 2700 energy consumption but cost considerably less energy upfront). With the buffs to mass adjacency this would make for a pretty solid late game Factory adjacency, not as good as a MEX but viable if you want additional mass efficiency.
Unfortunately a lot of these stats and balance are dictated by volatility and adjacency bonuses which in my opinion have caused more problems than the strategic depth they have added. For the uninitiated they're just another thing that 'is the way it is' and must be learned.