Thursday, October 12, 2017

How did the Kobe Steel faking happen?

Doubtless you've heard about this, but in case you haven't, via the New York Times (article by Jonathan Soble and Neal E. Boudette) 
TOKYO — Big manufacturers of cars, aircraft and bullet trains have long relied on Kobe Steel to provide raw materials for their products, making the steel maker a crucial, if largely invisible, pillar of the Japanese economy. 
Now, Kobe Steel has acknowledged falsifying data about the quality of aluminum and copper it sold, setting off a scandal that is reverberating through the global supply chain and casting a new shadow over the country’s reputation for precision manufacturing... 
...Kobe Steel said on Sunday that employees at four of its factories had altered inspection certificates on aluminum and copper products from September 2016 to August this year. The changes, it said, made it look as if the products met manufacturing specifications required by customers — including for vital qualities like tensile strength, a measure of material’s ability to withstand a load without breaking when being stretched — when they did not. 
On Wednesday, the company said it was investigating possible data falsification involving another product, powdered steel, which is used mostly to make gears. The company said the powdered steel it was examining had been sold to one customer it did not name...
So here's what I want to know - how the heck was this not caught by the customers? Is metallurgy different than chemistry? Did customers only rely on Kobe Steel's testing? (Do they not have their own QC labs?) Man, that's remarkable if so.

(Of course, how often did you QC stuff from Aldrich in grad school? Rarely, if ever, for me.) 

10 comments:

  1. In defense of supply chain and production managers, I'd point out that one of the reasons that OEMs buy higher-priced raw materials is the assurance that the supplier has a robust internal QA/QC program. OEMs pay a little more up front so that they don't have to QC raw materials as they come in. In terms of price value, strict adherence to production specs ranks right up there with excellent technical support, local rail spurs, dedicated transit fleets, rebate programs, etc. Unfortunately for Kobe Steel, the market tends to punish those who charge a premium for value-added product features, then fail to deliver the value.

    ReplyDelete
    Replies
    1. "OEMs pay a little more up front so that they don't have to QC raw materials as they come in."

      This is something I did not know - thank you!

      Delete
    2. there is a close analogy - a manufacturer of a dietary supplement blend containing vitamins and aminoacids, buying the raw ingredients in the certified food grade quality from a reputable western supplier (rather than from a cheaper supplier in China)

      Delete
    3. I thought my company was sloppy and careless for not QC'ing incoming raw materials, but I've since found out that this is a standard thing in industry today. I suspect the real reason is because it's cheaper to deal with the occasional quality issue than to employ more QC techs, and we have enough leverage with suppliers that we can often get them to pay for batches ruined by mistakes on their end.

      Delete
  2. Anonymous @ 9:35 AM,

    Well-stated. You are absolutely correct. It will be interesting to see if accidents have resulted from materials not meeting specs. Law suits also punish liars.

    ReplyDelete
    Replies
    1. it may take time to determine that - as you know bearings take time to wear out and fail.

      Delete
    2. A while ago there was a similar story about bad aluminium from a company in Oregon. It is suspected that sub-standard aluminium caused two NASA satellite launch failures and a loss of more than $500 million.

      http://www.oregonlive.com/business/index.ssf/2017/08/after_failed_space_flights_nas.html

      Delete
  3. The question is if management oversaw it (either directly or with a wink and a nod) or if the plants had some issues that made them unable to make certified steel and unwilling to remake non-specd products. With people at four plants lying, it's harder to think that someone in management didn't know about or facilitate the falsification.

    ReplyDelete
  4. How often did you QC stuff from Aldrich? Only when things weren't working...

    ReplyDelete