Shift the Mean?
- October 9, 2018 at 1:46 am #56112
is one of the improvement methods in six sigma is to make shifting to the mean of the process as a first step before starting six sigma project (DMAIC)?
and if it is possible how to make this shifting?October 9, 2018 at 5:30 am #203113
What you are asking is this: is there a way to solve my problem before I know what the problem is – and the answer is “no”. The whole point of DMAIC is that of problem identification, investigation, and resolution.October 10, 2018 at 2:24 am #203117
but suppose that the results moving toward the upper spec. limit , so if i can move the specs little bit and examine the process again and gives me better CPk, isn’t this may consider a fast solution ?October 10, 2018 at 6:47 am #203118
No, that’s called cheating. We are assuming your upper spec limit is there for a reason. If it isn’t arbitrary then, given your description of events, you are just trying to pretend that some kind of special cause which is driving your process to the upper spec limit (and presumably at some future date will drive your process past that limit) doesn’t matter. The only solution your approach provides is a way to ignore problems with your process for the time being – this is not process improvement.October 10, 2018 at 7:13 am #203119
could you please explain to me when i can use the shifting as a solution , i google it but i can’t understand
i know that i must find the causes of the variation either they are special or common causes but i also know that i may use the solution of shifting the process mean as one of the solutions
could you please clarify thisOctober 10, 2018 at 8:49 am #203121
@feodor85 First if it is acceptable to move the spec then it should not have been there in the first place. You don’t move the spec just because the process is drifting closer to it.
Let me try an example. I had a process where I was plating gold. The most profitable place to operate is right on the lower spec limit with a mean equal to the spec limit and a std deviation of 0. That won’t happen so I need to set a target as close to the spec as possible. I can do that by using my process mean and std deviation but I am assuming I can shift the mean without affecting the std deviation – you need to verify that.
The mean movement id frequently done with knobs. I try to move my mean closer to the target I calculated and then do a follow on study to verify I am still ok. As I continue to improve my process and decrease the std deviation I can shift my mean closer to my optimal target.
You shift based on data.October 17, 2018 at 8:48 am #203131
What business problem statement are you trying to solve? @feodor85
I’d use your process capability, data gathering (X and Y), and SPC to give you hints as to where you should focus–yes reducing variability will have an effect on mean but typically focus on mean shift OR variability reduction independently. It might sound counterintuitive but if you follow the DMAIC route, you’ll find the X’s having impact and can apply to both mean movement AND s.d. reduction.
You must be logged in to reply to this topic.