> No one really understood or had a handle on the production process.
That was the problem with the Bessemer converter. It took about 10,000 trials and advances in analytical chemistry to make that process work. Once the thing starts the chemical reaction that makes it go, all you can do is watch until something steel-like comes out. With the wrong inputs, you get crappy steel. Bessemer only got good steel because he accidentally had the right composition of inputs. He had to refund some customers who bought converters and got nothing useful out.
The Romans could have built a Bessemer converter, but it wouldn't have worked consistently.
It took a long time to get open hearth furnaces working. They're a collection of brickwork chambers and plumbing which can get up to 3000F and hold that temperature.
The exhaust gases heat brickwork which in turn heats incoming air. It takes some understanding of thermodynamics, and really good firebricks, to make such a furnace. Otherwise the furnace never gets hot enough to make steel. Much of the early history of iron and steel making is the story of trying to get something done with a furnace that isn't hot enough to do it right.