Ron

I’ve just returned from a business trip to China and Korea… Business is definitely booming in those countries and appears to be improving in my business.
 

By definition Inspections will never be considered a value added activity. Now that I made a lot of friends let me go on…
Value is in the eye of the customer..what they are willingto pay for. Since qualit of products and services is expected by the customer they may not be willing to pay extra for it.
However, as we all know allowing defective…[Read more]

You have touched on a very critical and important function…Team members. I am surprised that you would be perfomring a “Performance Assessment ” on team memberas by definition these people are those who are performing at least two job functions by participating on your team.
Team members needs to be a part of the solution and as such you would…[Read more]

Shewhart SPCcharts are robust to the data type and do not require normal distribution data.
I am not aware of any charts other than and ImR that you can use either type of data.
There are application in using ratings as continuous data that can be interpreted using continuous methodology.

anything you can obtain data on can be plotted on a control chart. It is up to you what you do with the data and the output.
 

This of course is NOT CORRECT.
 
when you samples sisze are small as they would be in statistical sampling plans using the formal equation for calculating process standard deviation are inaccurate therefore your software is inaccurate.
you must use an approximation of process standard deviation which is what minitab does.
Understanding the…[Read more]

The short term answer is of cource Yes. But I question why you would even report COQ. COPQ (Cost of Poor Quality) is a significantly better metric to track.
COPQ would contain : Scrap, Rework, Engineering evaluation time, warranty costs, etc.
COQ contains many cost items that you really do not want to report because the metric drives the action…[Read more]

This is a classic two sample T test.
 

Please let us know your product line I want to be sure I don’t buy any..
A product recall is a function of the product failure mode criticality and also the likelyhood of occurrence.
The decision to recall can never be a statistical “coin toss” for reputable companies.
We base product recalls on the nature of the defect, the potential failure…[Read more]

Yes,
You refer to Cpk upper and Cpk lower in your case there is only a Cpk upper.
The tool is being used to visually display where your data falls relative to the desired specification
 

Don’t know if you are a six sigma person but here is my thoughts:
Use a simple Pareto methodology to group similar failure modes from your returns. Start with the largest failure mode and drill down in that group to determine root cause (s) and fix that issue then continue in this manner.
Statistical sampling does not make any sense until you…[Read more]

There are many types of control charts and shewhart charts are only a few. These Shewhart charts do not require any specific distribution so sure go ahead and use them.
EWMA may be your best choice since you can control the weighting on more recent data.
Let us know how they work for your client.
 

Detection implies your ability to detect the defect you are inquiring about in your PFMEA. 1 – always detect it 10 never detect it.
Do you have a method to detect the failure mode and mechanism you are describing before you deliver the product to the customer.

Damaris,
First Pass Yield and RTY are totally different and should never be confused with each other.
Your analogy is correct: RTY is the product of the yields at each step in a process. Each process steps should have a FPY. Never confuse FPY and RTY they are totally different.
The reason this is so highly taught in Six Sigma training is that…[Read more]

Sure..remember that minitab loves the vertical columns  so enter your four heading above each column you will have to make 24 sets of these data one for each filler head.
When you select the chart you are going to use to monitor the process you follow the simple minitab instructions and you have it.
If you stack all your data vertically you can…[Read more]

In six sigma the important metric is DPMO or defects per million opportunities.
The posting of the printed circuit boards jogged my memory.. In the automotive world PPM was sufficient but in most industries including the automotive DPMO is more accurate as not all defects make an item defective.
 

This is sort of preaching to the choir but….. The first thing I noticed is that the author works at a university (this is not a recommendation for a person to comment about six sigma methodology). The second thing I notice is that his statement regarding people to people processes restates my first point..people working in education need to get…[Read more]

Linear regression investigates and models the linear relationship between a response (Y) and predictor(s) (X). Both the response and predictors are continuous variables.
A Pearson correlation coefficient measures the extent to which two continuous variables are linearly related.
One models the relationship the other quantifies the relationship.
 
 

Pete,
You have an interesting problem statement; Tool usage as many have replied is subjective. However, You would not use either ofthese tools in the Define Phase or in the Control phasein most cases
The problem with tools such as you have mentioned is that they do not carry and statistical validity. As such many people like to use them because…[Read more]

PDCA is a concept that should never have developed as far as it has. Plan Do Check Act totally misses the fact that decision need to be data driven.
I believe this is whuy the creators of the six sigma DMAIC methodology so rapidly over took this outdated terminology.
Define – Define the problem you are working on.
Measure – gather data on the…[Read more]