Thursday, February 04, 2010

I wish I was more surprised by this

As many of you know, my husband and I lived and worked in Japan in the late 1990s. Our role was to act as liaisons between the Japanese and American groups in the company as they worked toward passing a new quality audit Toyota was requiring. This involved a lot of paperwork and flow charts, and not much actual science, so we were pretty good at it.

While our job there only involved documenting the methods and testing, we heard a lot of stories from our colleagues about what went on behind the scenes with the product development. It turns out that Toyota's policy of continuous improvement is great ... to a point. And after that, it can actually get in the way of producing a useful product.

Let's say, for example, that you make some kind of paint or sealant. Obviously, the stuff doesn't stay good forever, so you need to know how long it can sit before it turns into unusable jello. This takes a different length of time depending on the environment it's subjected to, so in order to develop a useful test, you have to decide what the worst-case scenario is that you want the product to survive.

So, do you want it to be able to be usable after one day at 110F to mimic what might happen if the air conditioning went out in the storage room? Or maybe a few days at 90F to mimic what the stuff might experience in a really hot spell during the summer?

These are reasonable scenarios, but the problem with "continuous improvement" is that the test requirements keep getting more and more stringent, even when the existing requirements are more than sufficient for the job. So instead of passing three days at 90, next year you've got to pass six days at 90. Then 10 days. Then a week at 110. Then even worse conditions, with even smaller tolerances for change in the product. At some point you reach a situation where the test is so far beyond what the product could actually experience, it fails to tell you anything useful at all. Eventually, no one can follow the test method correctly and pass it, no matter how good the product actually is.

And according to our colleagues, this was the case with some of the tests that were required by Toyota - they had gotten so stringent that nobody - not us, not our competitors, not God Himself - could make a product that would pass those tests. Our contacts at Toyota, my colleagues said, agreed that the test was virtually meaningless and that passing it did not do anything to insure that the product was fit for use ... but they were unable to change or modify the requirement because that wouldn't fit with the company's quality philosophy.

This, along with many, many other situations we ran into with the whole quality auditing process (ask me some time about the auditor who refused to acknowledge my presence during meetings because I was a woman), totally enraged me. The product we were making was good and met all of the reasonable specifications, and I was confident it was a good product. But according to our colleagues, we had been virtually forced to change the product (or lie about the test results) in order to pass some tests that had evolved over time into completely irrelevant behemoths that even the Toyota guys said were pretty much useless.

Every time I fumed about this (which was often), my standard line was, "You know, the fact that we have to change the product or lie about our results is bad enough. But it's not like people will die if the product is a little worse for it. But if they do this to us, there's no reason to think that they do anything different for the brake manufacturers or any of the other suppliers. Do you really want the brake supplier lying about test results, or making an inferior product that happens to meet Toyota's asinine quality requirements? I will never, ever buy a Toyota!"

We've been back in the states for a decade now, and the passion behind the statement has dimmed somewhat. The last time we were looking for a new car, Jason cautiously admitted that he would consider a Toyota, if he found one that appealed to him and was in his price range. I was still dead set against buying one, but I now do a lot less foaming around the mouth when I discuss the topic.

Or I did ... until Toyota started having some quality issues ... with things like, I don't know, the accelerator pedals on a bunch of models. You know, the ones that sometimes randomly accelerate and have killed people? Yeah, those. Wonder how their quality audit results looked ...

And as if that wasn't bad enough, now they're reporting problems with - you guessed it - the brakes on some Prius models.

Now, if I weren't such a caring, empathetic person, I'd be chortling with glee while jumping up and down shouting "Hah! I told you so! I told you so! Pthththththtbth!" in the general direction of Toyota-shi. But dude, people have died, so I'll skip the gloating and finger pointing. I do, however, allow myself an irritated snort of scorn and disbelief when I'm watching a television show we recorded a few weeks ago and see one of the Toyota commercials touting the quality of their vehicles.

Guess we won't be seeing those too often anymore.

1 comment:

Anon said...

This is fascinating.

Thank you for helping to crystalize some of my issues with poorly considered corporate metrics. I see some of this with regards to safety in manufacturing facilities in the States. You can never have too much safety, right? Just like you can never have too much quality, right?

The manufacturing firm works hard to improve safety. Anything deemed unsafe is punished. Employment and raises are tied to having a spotless safety record. The result is that folks hide anything that can be hidden. Thus, small issues are hidden until something major occurs.