My assumption was it would take a very sizable amount of venture capital to do it properly, with a lot of red ink early on. There are an immense number of restaurants, bakeries, etc founded in the US every year, so I suspect there's potentially some kind of big market for this - especially as the order-to-go business gets drastically bigger over the next decade.
I considered that the large unit space / building could have a dining area, where customers could sit down and could then select from numerous kitchens that are local to that space. That pushes it even closer to the mall food court. My reservation on that, is that some percentage of the restaurants will be producing subpar experiences in terms of quality. So a successful restaurant in the space, wants its customers to eat in a successful atmosphere, where customers as a whole are happy; if half the customers in the dining area are unhappy, the enjoyment by the other happier half will suffer as well (people pick up on such things, it rubs off). Then I considered sub-divided dining areas, so each rented kitchen gets its own limited sit-down space, it just might be a stretch though and adds to the overall cost burden in the system with minimum gain (ie, it's probably just easier to focus more narrowly on food delivery).
To deal with the vacancies issue, you'd only build these in higher population urban areas, to maximize the potential number of restaurants etc. that you can rent to (maximize the potential start-ups due to population). You'd also seek to under-build versus demand, preferring to always have a waiting list (recession times might even trash that best effort). I don't know if it would work, but, I'd seek to get a modest upfront payment allocation from the waiting list, that covers the risk on vacancy turn-over. By keeping the demand above the space availability, you can probably get the start-ups to front the first month of rent to get on the list (with some terms of refund). It wouldn't be something you could start with zero capital, the idea would be to get it into the $5k (solo bakery) to $25k range to go from moving into the space to the time it takes to get the first customer.