Your two graphs are very good. 0:45 Traditional Loss Approach. This is the "inspect and reject" mentality that all parts within specification are equally good, with essentially no loss cost. 2:32 Taguchi Loss Function. This leads to the six-sigma approach when looking at large quantities of parts. A picture of the famous bell-shaped curve might be valuable here. You want your average value to be the nominal value (otherwise the loss is greater on one side than the other). You want your six- sigma (variance) value (2 or 3 failures per million) to be within limits. You have very few parts at the "almost to the limits" area where the loss cost is significant, even though technically acceptable. The narrower the variance, the less the loss. Which is distinctly different from the "anything in the limits is zero cost" analysis. Nicely explained!
Hello Ruth, in case of Taguchi there are several: - always aim for the target value, not only within specification. This is what drives operations excellence - every problem is a design problem, not the fault of the operator, it should be solved by "robust design" (such as Poka Yoke) to make failures impossible - Taguchi design of experiments (maybe I make a video about it later) ...
Your two graphs are very good. 0:45 Traditional Loss Approach. This is the "inspect and reject" mentality that all parts within specification are equally good, with essentially no loss cost. 2:32 Taguchi Loss Function. This leads to the six-sigma approach when looking at large quantities of parts. A picture of the famous bell-shaped curve might be valuable here. You want your average value to be the nominal value (otherwise the loss is greater on one side than the other). You want your six- sigma (variance) value (2 or 3 failures per million) to be within limits. You have very few parts at the "almost to the limits" area where the loss cost is significant, even though technically acceptable. The narrower the variance, the less the loss. Which is distinctly different from the "anything in the limits is zero cost" analysis. Nicely explained!
Hello Camgere, thanks for your comment,
Dear sir, as a foundation on taguchi is clear enough with your explanation and request to do the function regarding this also. Thank you very much 👍
Another Question: Did boeing, vaccine and toyota applied such method to their products?
What about the formula? :-) Loss=k.(y-m)2
Question...How is this different from engineering margin??
That was clear, thank you!
Thank you!😀
USEFUL
Well explained! btw the "loss" concept actually feels like sth in economics
Great explanation. Thanks a lot!
glad I could help you!
Thank you!
HI I would like to ask what does quality practices impacted by the contribution of the guru?
Hello Ruth, in case of Taguchi there are several:
- always aim for the target value, not only within specification. This is what drives operations excellence
- every problem is a design problem, not the fault of the operator, it should be solved by "robust design" (such as Poka Yoke) to make failures impossible
- Taguchi design of experiments (maybe I make a video about it later)
...