Theory About what Kodak Is Doing
Let's assume, for the moment, that the article cited in the OP is fact. Don't know that it is or not, but let's pretend that it is. I have a theory about what Kodak is up to and I'd like to hear what some of the rest of you might think. Let me lay out my thoughts, and this is specific to color films:
First, some observations on my part:
1) Kodak has put their recent R&D work (such as it is) into color negative products. Ektar, Portra, Vision 3 and maybe even consumer C-41 films derived from the above.
2) Films outside the CN arena have been reliably discontinued, probably as the master rolls have run out. Examples: Kodachrome, Ektachrome and so on.
3) Color negative use IS declining but will probably settle out at some relatively stable level. Maybe we haven't hit bottom yet or maybe we have. IDK. But I am assuming there is or will be a residual market worth serving.
4) We've all noticed over the last few years that the newest offerings (Portra, Vision 3, Ektar) have definite similarities. Kodak seems to present Vision3 as the root from which the others are derived.
5) It also seems that the latest offerings are very versatile. (Well, maybe not Ektar) Good results over a wide range of situations, push to higher speeds with good results, etc.
6) In the OP the speaker is cited as saying that new technology will allow Kodak to make any sort of film (I am assuming current offerings here) at no particular minimum quantity, as long as the roll is 54 inches wide, i.e. coated on the same line on the same machine.
So, then, this is what I think they are doing:
Kodak has, or is working on, a standardized color negative emulsion that they can use to make ANY AND ALL of the current products. There are slow speed, medium speed and fast speed silver components which can be selectively blended to yield an ISO of whatever you want: 100, 250, 400, 800 and so on. To make any particular film stock, you then tweak the particular dyes and sensitizers that are added to the emulsion. All this mixing is done under computer control just ahead of the coating line AT PRODUCTION TIME. A master roll is then coated with whatever product mix they desire in a specified sequence and then cut up accordingly.
So let's say, for example, we would need (in 35mm) 5000 rolls of Ektar, 10000 of Portra 160, 7500 of Portra 400 and 50,000 feet of 5207 in 35mm for a given week's production. (Realize I am just making up numbers for an example as I don't know the specifics involved.) A coating run might go like this, then, assuming a continuous supply of film base at the beginning of the line: 4000 feet are coated with 5207, then under computer control, the mix is changed and the next 950 fee are coated for Portra 160. Then the mix changes again and the next 700 feet are coated for Portra 400. Finally the mix is changed for Ektar and the final 500 feet of the run are coated with that.
The coating line doesn't stop. As the film comes off the line it is cut between the various products and made into smaller master rolls of each. These smaller masters then are slit and packaged in the traditional manner. Or maybe the the coated film is fed continuously through the slitting and packaging operation with that part of the process automatically changed on demand as well.
This production methodology could be scaled for production on a weekly basis, or maybe, daily or maybe even hourly. The idea being that they keep the line running by producing only the product mix needed over a relatively short period of time. What's being coated by the machine is a mix of very closely related variations based on the same basic components. It allows enourmous flexibility. If we don't need any Ektar this week, we don't coat any, but we have already shipped the stock we made and don't have master rolls (INVENTORY and therefore $$$$) languishing on the shelf aging and potentially going to waste. You make it and ship it in one operation exactly at the time it is needed. Nothing sits around in the warehouse consuming space and inventory dollars going bad.
If you needed a special film, as long as it is made from the same basic components, you run it through the same process by creating a special program for the computer control and coat exactly the amount you need. No leftover and no waste. Or at least minimal waste. You can coat 100 feet of it once a year and the impact is just about nil because we have moved from a batch process to a customizable, continuous process using standaized components.
Obviously this is a simplistic example but I think it fits the "facts" - in the sense of "facts" that are floating around here. If I were given their constraints these days, that's how I'd do it. And based on the direction the manufacturing consultants my employer has hired over the last few years have steered our workplace, if Kodak had anyone similar helping their reorganization, and I am sure they do, I'd be willing to bet this is the direction they are being driven.
Can B/W films fit directly into the process on the same line? IDK but B/W can likely be produced in the same manner.
I might be wrong but it seems a lot of people in these forums don't really understand about modern lean production methods. I'd think the days of batch production at Kodak and making up big master rolls to sit on the shelf till they get slit are long gone. If they arent long gone then no wonder Kodak is in the predicament it's in because they'd have a sh%%load of money tied up in persihable and perishing inventory which eventually would have to be written off. 
