All entries for Thursday 02 December 2010

Taguchi DOE, and how I wish it wasn't used.

The day spent learning about Taguchi methods for experimenting (DOE) was very interesting. I’ve often wondered how practical it is to try and isolate problems by changing only one factor at a time, especially in complex situations, where there are so many possible variables, that there are a hundred or more possible permutations. This approach allows you to identify what is important in an incredibly efficient way I feel.

The BAE session the next day was useful in putting it into true manufacturing context. But, I found it hard to get through the session. I intensely dislike so-called ‘defence’ companies. Really, is there a more abhorrent use of intelligence than for the creation of materials designed solely for the purpose of ‘keeping order’? Or killing people, as I see it. I spent a lot of time wincing, particularly when listening to Roger’s description of developing warheads the size of dustbins! I do not ever want to work in the arms/defence industry, or towards anything else that is deliberately (or even not) designed to harm people – it is just contrary to my world view.

I did like that he was quite aware of all this – some people who work on these kinds of projects have no appreciation for the effects of what they are doing. Roger seemed to have a remorseful tone, and said that it was something you had to think about. His view was that what BAE Systems and others do is a necessary evil, as if we are asking people to risk their lives on our behalf, we have to provide them with the tools that give them the best possible chance of coming back alive. While I’m not sure if I can agree with the necessity of going into unjustifiable wars, I can understand his outlook. I don’t really mean to preach or offend anyone affected by the issue, I just couldn’t live with myself if it were me.

Back to Taguchi. I asked if DOE was used more as a problem solver in industry (the main context I had seen so far), or if it was ever used during the design and development stage to make better products. I was informed that some companies (Ford, Xerox) do use it the latter way, but at BAE, it had never really become embedded. A lot of the time, what happened, is that when it was applied, it brought success to those who had used it, and they moved up the chain of command or out of the company.

In my opinion, good! I don’t really want them to figure out how to murder people more effectively...

: 02 Dec 2010 23:55 |  Tags: Defence Industry Mbe Piuss Taguchi |  Comments (0) |  Close comments |  Report a problem