Chinese factory

Loading thread data ...
I love this statistic:
``...Since the robots came to the plant the defect rate of products has dropped from over 25 per cent to less than 5 per cent...''
Comments welcome
i
Reply to
Ignoramus29244
Ignoramus29244 fired this volley in news: snipped-for-privacy@giganews.com:
Obviously, it's because the robots "do what they're told" instead of "doing what they want to do"!
What's missing in all that is that NO plant is TOTALLY automated. There have to be people there (at the very least) to monitor the operation, and handle repairs. We aren't quite yet at the point where the robots will also do their own maintenance.
What's fishy about the claim of defect rate, is that any plant turning out 25% defective product could never have remained in business long enough to automate!
Lloyd
Reply to
Lloyd E. Sponenburgh
First, I can't see India ever allowing an unmanned factory anywhere in their country. They have some of the cheapest labor available anywhere and hundreds of millons of poor to feed.
Second, is a company admitting to a 1:4 failure rate. Ghastly. Haven't they heard of "training" and/or "Quality Control"?
Third is "down to 5%"?
Reply to
Larry Jaques
Larry Jaques fired this volley in news: snipped-for-privacy@4ax.com:
Fourth, "INDIA"???? Really, Larry?
Lloyd
Reply to
Lloyd E. Sponenburgh
It isn't "subsidies." There are some processes that produce scrap rates on that order. At one time, single-crystal turbine blades had about a 30% acceptance rate. It's near 100% now, but it was still worthwhile at 30%.
Some high-quality investment castings turn out 25% or more scrap. And there are others. Some semiconductors have very high scrap rates.
There still are processes that are too expensive, or impossible, to control to the scrap rates generally acceptable in industry today. They tend to be very specialized.
Reply to
Ed Huntress
Robotics in some processes reduce scrap rates on that order of magnitude, but I'm suspicious of the whole story. I'd have to see what they're doing to "polish modules" with those robot arms to believe it.
Reply to
Ed Huntress
I worked on some cabinet doors machined from a large aluminum casting. We rejected about half for porosity which we couldn't find until we did a lot of cutting.
Another job involved SS bearing housings that had one end cut in one turning center, the second in another. Concentricity from one end to the other had to be +/-.0002". We were rejecting 35% when I discovered a fundamental flaw in the inspection process: before checking concentricity, we would wipe the part with a pink cloth shop rag. One night, I used white paper towels. 5% reject rate. Same the next night, while the day guy still got 35%. I mentioned it to the inspection dept. They pointed out that I was not following procedure. I said we had been out of shop rags, which was true the first night. Shortly thereafter, procedure was changed to require use of white paper towels. Inspection dept took credit, no attaboy.
David
Reply to
David R. Birch
Imagine if it had been toilet paper, and it became part of your ISO 9000 documents. d8-)
Reply to
Ed Huntress
This was a bit before the ISO fad, it was near the start of the SPC fad.
David
Reply to
David R. Birch
On Tuesday, August 11, 2015 at 11:33:01 AM UTC-4, Lloyd E. Sponenburgh wrot e:
Nucor has a couple of fastener plants that come close to being TOTALLY auto mated. They run three shifts and the graveyard shift runs totally automate d. They have automated the monitoring. If something goes wrong , the mach ine shuts down and it gets fixed on the next shift.
Dan
Reply to
dcaster
Nucor has a couple of fastener plants that come close to being TOTALLY automated. They run three shifts and the graveyard shift runs totally automated. They have automated the monitoring. If something goes wrong , the machine shuts down and it gets fixed on the next shift.
Dan =========================
formatting link
"FANUC, the Japanese robotics company, has been operating a "lights out" factory for robots since 2001. Robots are building other robots at a rate of about 50 per 24-hour shift and can run unsupervised for as long as 30 days at a time. "Not only is it lights-out," says Fanuc vice president Gary Zywiol, "we turn off the air conditioning and heat too."
American inventor Oliver Evans built the world's first fully automated factory:
formatting link
"In the 1780s, Evans built a completely automatic grist mill in New Castle County, Del. Powered by a water wheel, the mill was the first continuous flow, production line mill in the world. An English book of the day described the mill: "Mr. Oliver Evans, an ingenious American, has invented ... a flour mill upon a curious construction which, without the assistance of manual labor, first conveys the grain ... to the upper floor, where it is cleaned. Thence it descends to the hopper, and after being ground in the usual way, the flour is conveyed to the upper floor, where, by a simple and ingenious contrivance, it is spread, cooled, and gradually made to pass to the boulting hopper." The product wasn't touched by human hands from the time the grain was dumped into the receiving hopper until the finished flour flowed into a bin ready for packing into barrels or bags."
The sole operator handled shipping and receiving.
-jsw
Reply to
Jim Wilkins
Fanuc was running the first modern lights-out factory, building wire EDMs, from around 1991. It worked very well but I don't know how cost-effective it was. One guy monitored the whole thing, walking about with a lighted clipboard. They built it partly as a demonstration of what they could do with their robots and networked CNCs at the time.
There is enough lights-out activity going on that my publishing group made a prototype of a new magazine, called _Lights-Out Manufacturing_, four years ago. We decided that there wasn't enough editorial material to sustain a full-scale magazine at that time, but there is activity.
Reply to
Ed Huntress
But John, it isn't clear what kind of process they're talking about -- whether it's some group of processes, or just polishing that cell-phone "module."
I'd want to know more before getting critical about it. For reference, when P&W started making their single-crystal jet-turbine blades, the reject rate was 70%. But their performance was so much better than the blades they made previously that it was worth it.
There are some needs for parts in production that can barely be met by the best known or best practical processes -- they have low Cp or Cpk, if you're into that stuff. There are other processes where the material is so cheap (plastic phone cases) where a high reject rate isn't much of a hardship, compared to the cost of improving the process.
We're so locked into sigma values over 4 these days that we sometimes forget that there are some things that, in practice, we can barely make to spec no matter what we use. But they can be things that we really need.
Reply to
Ed Huntress
There are other processes where the
Or like using aluminum for pickups............... Rejects and scrap just get recycled.
Dan
Reply to
dcaster
I used to design burn-in and production test equipment. What kind of failures do you see?
-jsw
Reply to
Jim Wilkins
Finally someone saw that statement for what it REALLY means!
25% defect rate? FUCK!!!
i
Reply to
Ignoramus28978
I tracked this story back to it's source: People's Daily. It's virtually the same story.
I have no confidence that People's Daily got the story right.
Reply to
Ed Huntress
But, your labour cost is $15/hour, theirs is $15/day :-)
If a worker can make 10 pieces a day it cost you $120 to make 10 units. It costs them $15 to make 10 units. they throw 25% away so individual unit cost is $2.00 while your's is $12.00. they sell theirs with 100% mark up for 4 dollars. You go belly up.
Reply to
John B.
I remember a post in alt.machines.cnc, someone relating a story about making boatloads of small brass parts for a customer. By the barrel full. Customer goes to China to get them for some significant amount less. First delivery shows up and roughly half the parts are scrap, out of tolerance. Company complains, China factory says they'll make twice as many, same total price. And yea, the US company set up an inspection line to sort good from bad... Forgive if I got it a bit wrong, this was years ago.
Jon
Reply to
Jon Anderson

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.