If you allow me to make 2 assumptions..
Assumption 1 – You are tracking project cycle time because you would like to reduce it
Assumption 2 – The fewer days a project takes the better
I would count total days (weekends included). Keep it simple. If there is a bank holiday somewhere or some holidays or whatever, you will start messing with the…[Read more]
We have completed a SMED kaizen event yesterday. Our goal was 60% changeover reduction. We have achieved 32% on our first attempt. Our approach was to prepare all data necessary for the 3 day Kaizen event. Please drop me a line on firstname.lastname@example.org if you would like me to show you what we mean about “prepara the data”.
If you have, for example, a data set like 100.1; 100.3; 100.6; 102.4; 103.7; 100.4
You can get chunky data by:
1- rounding off result (100; 100; 101; 102; 104; 100)
2- using a measuring system not precise enough (same as above if instrument does not read to 1/10 of reporting value)
hope it makes sense..if you still have questions try to do…[Read more]
I am no expert but It might have something to do with chunky data. I have not heard the rule but have seen the effect of not having precision. Chunky data in a Control Chart means that you will probably have points outside the limits when the underlying process is stable/predictable.
Hope it helps
Thanks to all for the replies,
Instrument is able to provide 2 d.p.(example: 7,82)
QC reports result as 7,8 (for the reason you just mentioned and I agree). The result is to be expressed with 1 dp because the second d.p. is not reliable.
The issue is when it comes to the usage of Control Charts. QC is quite resistent to the suggestion…[Read more]
Thanks for you offer. I have run the experiment this morning and I have the data.
When you find it is a suitible time for me to email you the data, please say so. As it is Friday and we have a bank holiday on Mon and Tues please contact me on rodrigo_s_pereira(at)hotmail.com
thanks a lot
Annon, Deep, Sandor, Mike 86 n DOE, thanks for your replies
Sandor – Ill give you a shout after the experiement.
Mike86 – there is pre-compression before compression. I was aiming at keeping pre-comp. force constant throughtout the experiement. (at “normal” level). I agree with you with regards to the granulation and raw materials.However,…[Read more]
As for Ppk specifications I use the following:
<1.33 = Bad (<4.5 sigma)
1.33 – 1.67 = OK (4.5 – 5.5 sigma)
>1.67 = Good (>5.5 sigma)
A simple way of interpreting Ppk is “the higher the Ppk the better”. I am not aware of a max. value for Ppk.
If I understood you correcly, you dont have a specific USL or LSL but you work with a target…[Read more]
i have contacted minitab to know exactly what you are trying to find out. In a heavy regulated industry you must have your software validated. If you want to know the details of mintabs reply email me to email@example.com
speak to you soon
Where I work we measure the uniformity of the batch in terms of Mass Uniformity. The procedure we use is ok/accepted byt the Regulatory authorities:
We sample the batch (say mil-std 105) take 20 units and verify if the individual weights are within x% of mean.
hope it helps.
I havent done any work on that front but I am reading a book at the momment that might be of interest. Its called:
“Lean Six Sigma for Service: How to Use Lean Speed and Six Sigma Quality to Improve Services and Transactions ” by Michael L. George and others.
Have a look on amazon and see if it helps.
- Load More