Officinae
A couple of weeks ago I was asked about officinae and their relationship to a mint. For those who have no idea what "officinae" are (let's hope you know at least about mints!) these are the working offices within a minting facility. In the ancient Roman empire a single mint might employ thousands of different workers in each of these coin-making factories. During the late Roman period it is, as far as I'm aware, a uniquely Roman feature that coin engravers were often ordered to note not just the mintmark but the "assembly team" as well on each coin going out the door. Specifically, the question was whether these officinae were spread out across several buildings or were they all under the same roof. Without the gift of a crystal ball I can't claim an absolutely authoritative answer but I can throw an educated guess or two.
Let's think for a moment why there was a need for officina markings in the first place, and by extension mint marks as a whole.
For a quick review let's go back to the early years of the late Roman age when the use of mintmarking really goes into the mainstream. Under Emperor Diocletian a radical monetary reform takes place which can best be summarized as a "scientific" approach to deal with inflation by means of trimming the costs associated with producing all this money. Since at this point there was no appreciable amount of silver to be sucked out of them anymore, and reducing the weight of the modules was clearly another deadend solution, the only meaningful recourse left in making cheaper coins was to streamline the process as much as possible.
To this end the effort expended in rendering individualized portraits was phased out and reverse designs pared down to a minimum. The initially diverse number of portrait styles was quickly scaled down to the standard "laureate head right". The military designs featuring quadrigae and emperor-shaking-hands-with-another-guy were ditched when found to be too complex to scale up for superproduction. Engraving techniques from one corner of the empire to the next became less distinct as a larger number of less-skilled engravers could be trained using standardized teaching methods then quickly put to work.
Emphasis was diverted to the singular purpose of maximizing output while retaining high quality control. That was the game plan. But the homogenizing effect created logistical problems for treasury accountants who now couldn't easily keep tabs on quota fulfillment and the transport/disbursement of so many millions of coins. It is also likely that other bean counters needed their own efficient models for back office operations, track currency distribution and tax assessment to minimize the economy imbalances that triggered civic riots and birthed usurpers. Parenthetically, let's also keep in mind that hyperinflation presents a fiat economy with an intractable problem: for whatever reason you just couldn't get away with printing more zeroes on coins the way a modern bankrupt country can do on notes. The cost of each and every coin had to by necessity be less than its face value.
It is doubtful in the extreme that Rome at this time could operate under prolonged budget deficits. There was, of course, no ancient International Monetary Fund ready with bailout money like today. It was simply sink or swim with hard currency. So, in order for them to get the reins on production and ensure maximum efficiency the only workable solution was to mark the coins with a finer level of manufacturing detail than simply noting the origin city allowed for. Enter the officinae. Now that this aim is understood it's only logical that the factories pumping out coins would operate in such a way as to minimize costs. Having physically separate officinae doesn't serve this purpose in any way.
I speculate that the mints themselves were likely garrison compounds with a guarded perimeter and several buildings of specialized functions like a repository for inbound metal, administrative offices, sleeping quarters, mess halls, etc. Most importantly, I think the core areas devoted to die making and coining were not the dungeon-like bunkers we tend to think of but rather made use of a semi-open plan that allowed for lots of daylight for the engravers and coin workers to do their job as well as facilitate ventilation of toxic fumes from furnaces and the chemical baths to process raw scrap metal into finished flans. Internally, it's likely that they were assembly lines with shared components and toiling close by to allow for the quick reassignment of work teams to regulate output as needed.
Well, that's my theory anyway :-)
Ras