Jonathon

I should clarify my question. I am not asking about the tamper quote. I am inquiring about “The data belong to the process, not to you.” Is that of your origin, or did you hear it from somebody else? I’d like to use it, and I’d like to give credit where due.

Thanks!

Great quote! Do you know its origin?

In control charts there is a U chart covering defects per unit, which readily accommodate more than one defect. From a standpoint of computing so-called process sigma, though, the opportunity space (area under the normal curve) is defined to be unity. The statistical wonks can debate at length about handling a defect count that exceeds the…[Read more]

More specifically I guess I am wondering what entry level positions I should be applying for to gain the most benifical experience.

Clearly, filling a warehouse with un-needed product is blatant Muda. The part with which I struggle is whether it’s wasteful to spend, say, 10 hours making product, if we can find ways to deliver it in 8. If we can increase demand, all is well. If not, should we operate fewer hours per day? Reduce staff through attrition?Thanks for your thoughts.

Lots of good inputs. Here’s a formula that works well:Plot a “before” and “after” control chart. If your eye and brain cannot detect an improvement, go back to “Analyze.” If the improvement is there but insufficient to meet your needs, go back to “analyze.”If the improvement looks “real,” proceed with hypothesis testing to verify “how sure” we are…[Read more]

IIf you remain at a point where one or more options could be combined into a number of different packages, then Pugh is not the right approach. Pugh applies when the options truly are mutually exclusive. For instance, a given floor can be covered in tile, carpet, wood, laminate, stamped concrete, etc., but it makes no sense to put wood over…[Read more]

I think you and I agree more than not, but I did leave out my assumptions. Shame on me for that omission.I assumed single piece flow of work through a series of reasonably load-leveled stations. If the process has batches, queues, and non-level loads, the equation becomes more complex. Also, the total process becomes slower, and variatioin about…[Read more]

Everything you say is correct. I’d like to build on it.I think of RTY as akin to the “bucket brigade” in old cartoons. Every time the bucket changes hands, a little water sloshes out until there’s hardy any left to throw on the fire. SImilarly, every time a unit of work exits a process step, there’s a likelihood that some “right-first-time”…[Read more]

As others have said, takt time is an idealized “drumbeat” of how often a finished item should come off the line. One can measure how the actual delivery rate varies from that time, although things get a bit messy if the line remains batch-oriented.The actual time that a unit is in the production line can be expressed as: (Takt time) x (# of…[Read more]

Darth:Right on, as usual. I also want to add: whether it’s full-blown QFD, or just the C-E Matrix component, your decision is only as good as your ability to prioritize requirements correctly.

I believe the P-Value is the area in the tail of the test statistic’s distribution. Many such distributions (such as the normal) cannot be integrated directly, so an approximation (Taylor’s series or the like) is used. I hope this is useful.

Lots of energy on this topic. That’s good.
In my experience, the best starting point usually is with what the process delivers, who gets it, and what they need. I have yet to experience an instance where failure to meet needs is “free” of cost. Thus, a decent COPQ analysis should reveal how that applies in your case.
Done properly, improving…[Read more]

Who are the author & publisher of that book?

If you are referring to the Rummler-Brache approach, it’s a variation on a flow chart. Each “box” represents an organizational entity, like a department, or a customer, etc. Each time one entity provides something to another (for instance, a sales department may provide an order to the factory), and arrow links the “supplier” (sales) to the…[Read more]

I guess we should direct the question to the originator of the topic: have you had enough, or do want us to continue this discussion through this forum?

I think we’re in a state of agreement on the practical use of control charts. If I understand your points as the rubber meets the road:

We both continue to use control charts even when data display non-normal distributions.
We seldom mess with the default choice of +/- 3 standard deviations. (Choosing narrower limits increases the likelihood…[Read more]

Gabriel, thanks for pointing out my error regarding the symmetry of control limits on range charts. You’re right, I was in error.
I tried to point out that attribute charts do use +/- 3 standard deviations; I don’t know if I stated that understanding poorly. So we agree on that point.
As for whether normally distributed data is essential for…[Read more]

Ideally, you want to get your process under statistical control before you estimate its capability. As a matter of practicality, sometimes that won’t work, especially to describe the “before” state. After all, the “before” state normally has lots of warts, otherwise we wouldn’t call for a six sigma project to attack it.
I would say it’s vastly…[Read more]