[Rhb]

Looking at a Cp and Cpk values that low, I would first look at the specification limits to see if they are realistic for the process involved.  If they are, you have to reduce variation in the process to achieve an acceptable Cp value.  Once that is accomplished, if the process is not centered, you may want to ask for a tolerance shift to center the process.  Most times manufacturers are reluctant to open tolerance ranges, but are somewhat receptive in shifting the tolerance up or down.

0

[Rhb]

Control charts are utilized to analyze an existing process in two ways.  First, to see if the data is normally distributed and doesn’t show abnormal trends or spikes.  Second, to see if the process has been stable over time.  Process Capability can not be accurately calculated without these two attributes.  They are also useful to track performance before, during and after a project to see step-change created due to project activity.

• This reply was modified 2 months, 3 weeks ago by Rhb.
• This reply was modified 2 months, 3 weeks ago by Rhb.

0

[Rhb]

You should not expect a histogram to be normally distributed.  Inherently, it will typically be skewed to the right and represent a Weibel distribution.  You could run a capability analysis  for non-normal data and set your upper boundary at the acceptable level for OTD.  This will give you a graphical picture of the process and an indication of your PPM failures.

0

[Rhb]

Keep in mind that OEE is Overall EQUIPMENT Effectiveness.  So, don’t make it too complicated.  Any delays during a planned/manned production period that degrades throughput counts against availability.

0

[Rhb]

I typically use a Time Series Plot with a reference line for the goal in Minitab.  Since you’re just looking at performance trends you don’t really need control limits  You could also calculate the standard deviation and place a LCL at -3 standard deviations with another reference line.

0

[Rhb]

I generally use Quadratic Regression because there’s almost always some curvature in data distributions. Just my preference, but I typically find it to be the best fit.

0

[Rhb]

Prerequisites for Capability studies are that the data is normal and stable, over time. Additionally, you can’t perform capability studies until you have an acceptable GR&R, so that would be the first step. To Mr. Robert’s point above, there’s obviously “special cause” variation included in the study. Also, you need to think about how the sample data was collected. What variation sources were inherent to the data? Was this a representative sample? Lot’s of unanswered questions.

0

[Rhb]

I prefer to see it as a graphical boxplot or you could use the One-way ANOVA for statistical validation. All types of graphical tools you can overlay histograms on one graph and delete the bars for a quick look.

0

[Rhb]

You have a 1st shift problem. There is a statistically significant difference between all three operators when comparing the two shifts. See attached .pdf’s Reply with email address if you want the MT file w/version you are using.

Attachments:

1. Graphical-Analysis-of-Net-Lbs.pdf
   You must be signed in to download files.
2. ANOVA-Comparison-of-Net-Lbs.pdf
   You must be signed in to download files.
3. Net-Lbs-WS.pdf
   You must be signed in to download files.

0

[Rhb]

Maintenance Mgt > Engineering > Production Mgt > Engineering Mgt > Six Sigma BB > Continuous Improvement Mgr all at World Color / Quebecor World > Quad Graphics Priting industry
CI Coordinator > Six Sigma MBB > Continuous Improvement & Maintenance Mgr > Continuous Improvement Mgr at SRG Global, Inc. automotive supply industry.

1

[Rhb]

Listen to Mr. 99Snowy above! He knows of what he speaks. Very similar to how I was about to respond…:)

0

[Rhb]

Sounds like you are mixing 6S which is adapted 5S vs. Six Sigma. Six Sigma is statistically driven process improvement, while 5S or 6S is Lean attribute improvement. Is this true?

0

[Rhb]

Focus on the “Problem Statement”. I use a check system with the problem statement that requires it answers the questions, “What, when, where and how much?” Also, keep in mind that the Charter is a living document that may require modification as the project progresses. Also, projects may require counter balance metrics. e.g. a counter balance metric may be, quality, scrap, etc. Stated, financial performance will improve by x, while maintaining the same or improved quality level. OEE metrics are good kpi’s for projects.

0

[Rhb]

Answer is, standard deviation within is the short-term standard deviation from a sample, containing “normal” process variation. This standard deviation is used in Minitab control charts to calculate the control limits of a chart. Formula is, Average Moving Range / d2 for a sample size of 2. Standard deviation within is used to calculate Cpk.
Conversely, standard deviation overall is the long term standard deviation of a Population, which contains both “normal” and “special cause” variation. Formula for this is the base standard deviation formula for a population and is used to calculate Ppk.

0

[Author]

Posts