This was always my experience in electronics - you, as the development engineer, set up small-scale production (going from prototypes to EV or engineering validation articles, which are basically the final product produced by these manual/slow methods), and then manufacturing engineers work with you to transfer your processes to an automated assembly line making who knows how many units a minute.
It's not unusual to produce a special prototype in the small/experimental floor; I mean, Apple probably goes through hundreds (or whatever) of prototypes and test articles before releasing their next phone, and they are certainly not stopping their million-per-day manufacturing plants to make them. However, it does show impressive commitment to the customer on the part of Fujitsu.
Also, electronics is more cyclical than cars (3-year-old model may be completely different and useless compared to the current one), so it is common to produce everything in batches rather than "the Toyota way" with just-in-time. In other words, if a laptop model is sold for 2 years, the whole run may have been made in 6 months, and then you have that manufacturing equipment sitting idle while the next prototype is worked on, so you may have equipment and engineers to spare. Only a few of the highest-volume factories (e.g. Apple again, I assume) run at full capacity all the time.
1) The final development PCBs (the engineering validation articles) are actually identical to the production PCBs. Your final EV revision becomes revision 0 in production; only the manufacturing method changes. So the prototype floor definitely has the ability to make production-quality PCBs.
2) The design change may or may not be complicated. Adding a port typically refers to a port already supported by the chipset; so you add a ribbon connector for it and change the layout only.
3) You generally wouldn't have full pick-and-place automation for making prototypes. There are manual tools (sort of semi-automated) for making a few PCBs. A large company is making prototype PCBs all the time. They know how to do it, and they don't stop manufacturing for it.
4) For the case, the easiest would be to rework an existing case (add a slot). Creating custom injection molds is kind of a big deal. Then again, if Fujitsu does that in house, maybe not that big. Don't know.
5) The expense would not involve a cash outlay, but it would definitely involve diverting engineering, manufacturing and sales resources to even communicate and deliver this whole thing, let alone make it. The engineers may have been pulled off something, or may have been waiting for work; business hasn't been that great for most manufacturers, if you haven't noticed. No, I don't know how Fujitsu accounts for this or why they made the decision to go ahead. I am not a manager at Fujitsu. I can tell you that most companies are very loathe to let prototypes and one-offs to leave their hands at all. So it's definitely impressive on Fujitsu's part that they would go to that length.
> Please fuck off if you don't know what you're talking about, for the sake of anyone that wants to ask a remotely relevant question on this braindead hellhole.
Uh, what? did I miss something?
1. A group of British construction workers in a Discovery Channel documentary loved to start fights in random bars until one day they accidentally started a fight in a bar frequented by members of a local boxing club.
They may have stuck the needed connector on a wire and used an NC mill to make an appropriate cutout from an existing housing part.
It is literally less than 1 hour to design and tape-out a inch-square pc board to accomodate a connector and cable termination. I am a little guy and have to wait a week and pay five bucks to have this manufactured. Fujitsu probably had this done the morning after a solution to the request was devised and approved.
As a customer, I think we bought ~1,000 units in the end. I've rarely seen that type of capability in my career, and it's too bad that the realities of the market have likely nixed it.
Quote:
Principle 3
Use "pull" systems to avoid overproduction.
So yes, you are right — Toyota Way is not viable for mass production, but it perfect for areas where human labor is still plays major role in manufacturing, e.g. programming.
As for being downvoted and needing a new account - I personally never care enough any more to get involved with that, but yeah, that's a good guess by the time after the West Coast wakes up, so have fun with that.
I appreciate your offer of friendship, but since this morning, after a conversation with the neighborhood not-quite-right-in-the-head lady, it would appear that I am getting all the crazy douchebag cultural immersion that I could possibly want. But thanks anyway.