Desirability, an Important Skill for Six Sigma Practitioners
Posted by: meikah | 13 August 2008 | 9:13 pm
This is new to me, too. So, what do we mean by desirability here?
It is a single number that describes how well a person’s performance comes to meeting the targets for all requirements simultaneously.
An article on iSixSigma has a good discussion about it.
The math in the desirability score card requires practitioners to know what the customer wants and what they will accept, known as voice of the customer (VOC); and what can be delivered and with how much variation, known as voice of the process (VOP).
Filed under: Tools/Toolkits, Statistics, Six Sigma
No Comments » |
SixSig Discovers Statistical & Scientific Thinking blog
Posted by: meikah | 8 July 2008 | 8:42 pm
I went over to Curious Cat Management blog and discovered an interesting blog: Statistical & Scientific Thinking blog.
In this blog, John Dowd discusses a wide variety of topics that are useful if you’re going into Six Sigma or any other process improvement. It talks about statistics, quality, productivity, research methods and the like.
Here’s the link to his take on Six Sigma.
Check out the blog!
Filed under: Statistics, Six Sigma
No Comments » |
Six Sigma and Statistics
Posted by: meikah | 23 June 2008 | 9:29 pm
Deploying Six Sigma means you will be handling data, analyzing that data, and so you will surely be handling statistics.
Here are two posts that will give you good insights on statistical concepts:
- 10 Key Statistical Concepts Every Engineer Should Know from Objective DOE
- 10 Statistical Concepts Everyone Should Know from Today’s Six Sigma
These two posts show us more than just the figures; rather, these underscore the role of these data, their interpretation in a Six Sigma or quality initiative.
Filed under: Statistics, Six Sigma
No Comments » |
Using a Scatter Plot in Six Sigma
Posted by: meikah | 10 December 2007 | 9:10 pm
A scatter plot, also called a scattergraph or scatter diagram, is a kind of chart that shows relationships of two variables in a graph. The scatter plot or diagram shows how one variable is affected or influenced by the other variable.
It is especially useful when you are dealing with a large number of data points. The scatter plot will show the relationships in terms of strength, shape, direction, or presence of outliers.
If you are into Six Sigma, all the more reason you will find use for the scatter plot, which is in fact one of Six Sigma’s basic tools. Below is an example:
If the overall production team was able to produce 500 cars before stating the Six Sigma plan. Then with the changes in waste reduction created a change of an additional 200 cars in the same amount of time. Then the numbers would show that there was a 200-point increase. This increase could have been due to the relationship of changes that occurred with the implementation of Six Sigma.
*Photo from NetMBA
Filed under: Tools/Toolkits, Manufacturing, Statistics, Six Sigma References, Six Sigma
No Comments » |
Six Sigma Improving Processes at BayCare Health System
Posted by: meikah | 28 May 2007 | 8:40 pm
I have always been concerned about how hospitals run their business, especially how they respond to emergencies or life-threatening situations. So when I read about Six Sigma or Lean being used to improve processes in hospitals or healthcare, I am happy.
BayCare Health System is a case in point. When Baycare started Six Sigma, it was Black Belt Angi Jennings’s first project. The goal was to reduce patients’ ventilator days and length of stay in ICUs. Using DMAIC, SIPOC, ANOVA, and other tools, Baycare was able to achieve its goals.
Here’s a summary:
Business Problem: High number of ventilator days for ICU patients
Methodology: DMAIC
Solution: Interdisciplinary rounding and standard weaning protocols
Benefits/Results:
- Savings of $650,000 annually
- Ventilator days reduced by 38%
- ICU LOS reduced by 23% for vented patients
- Reduction in ventilator related complications
- Improved teamwork and communication among care providers
Source:
BMG: Breakthrough Management Group
Filed under: Tools/Toolkits, Benefits and Savings, Services, Lean Six Sigma, Deployment, Statistics, Healthcare, Processes, DMAIC, Baycare Health System
No Comments » |
A Quality Quiz from PQ Systems, Inc.
Posted by: meikah | 15 May 2007 | 6:41 pm
I subscribe to PQ Systems, Quality e-Line, and so I get interesting insights on quality every time. Today, I got this interesting Quality Quiz, which I was able to answer correctly. 
I’m posting here the quiz now. Take the quiz and tell me how you fared.
Polly Yurathane considers herself the SPC queen, after successfully explaining the beta coefficient of regression analysis (standardized regression coefficient), to her new boss, Hy O. Pinyon. Now she is preparing a presentation for division manager Emily Dickinson, a poetic manager who leads with an iron will nonetheless. Emily has requested information about the relationship of the cost of coal mined by Blackern Dirt Mine Company to the amount of coal mined and the cost to mine that coal.
Polly, knowing that Emily is a poet, assumes that she knows little about statistics, so she feels quite relaxed about her presentation, seeing it as an opportunity to once again mount her throne as SPC queen.
Opening the Power Point presentation she has prepared, she explains the concept of simple regression, using this data:
DATE
TONS
(in 1000s)
TOTAL COST ($1,000s)
JAN
2
2
FEB
5
8
MAR
7
10
APRIL
1
4
MAY
7
7
JUN
8
9
JUL
2
5
AUG
3
5
SEPT
5
7
OCT
6
7
She then shares a printout, generated by software that does simple regression, and explains patiently to Emily that this is a regression model that shows the relationship between the cost (Y) and the amount of coal mined (X).
(Click on the printout for a larger version.)
Where
dependent variable
independent variable
“You can use the model to predict your cost if you know the amount of coal mined,” she says slowly to the division manager, who has remained quiet through the first part of Polly’s presentation, reinforcing Polly’s inaccurate conclusion that Emily is not absorbing the message. Now, however, Emily asks a question:
“What assumptions are you making to assure that this model works?”
Not expecting such a sophisticated question, Polly stumbles for a minute, then asserts firmly, “It is very important that the dependent variable is a parameter and the independent variable is a random variable.”
Is Polly’s response to Emily’s question appropriate and accurate?
a) Yes; now she has really assured her permanent position as SPC queen.
b) No: Polly Yurathane can consider herself only a princess at best.
You can also click here for a more complete video explanation.
Filed under: Data, Statistics, Data Analysis, Regression Analysis
No Comments » |
Control Chart and Six Sigma
Posted by: meikah | 12 March 2007 | 4:24 am
A Control Chart is one of the seven basic statistical tools often used in Six Sigma, along with the Pareto Chart, histogram, check sheet, cause-and-effect diagram, flowchart, and scatter diagram.
Also known as the Shewhard chart or process-behavior chart, it can monitor processes and assure that they remain stable.
Six Sigma on its part is data-driven and aims for processes to be stable and continue to improve for as long as the processes are working.
When you use a Control Chart therefore in your Six Sigma project, you can better evaluate your data and monitor your processes.
Business Knowledge Source, Manufacturing Info reports that:
Control Chart Control charts help distinguish process variation due to assignable causes from those due to unassignable causes. Both these types of process variation are charted on a control chart. Assignable causes or special causes are meaningful factors of a process and are not always present or normal. These types of causes can be avoided and should be investigated. Unassignable causes are also known as common causes or chance causes. These are factors that occur by chance. They are not always present, but are normal and expected within a process. They are unavoidable and inherent in a process.
Source:
Business Knowledge Source, What is a control chart and how is it used in Six Sigma? with a link from Six Sigma Zone
Filed under: Tools/Toolkits, Manufacturing, Data, Statistics, Processes, Control Chart
No Comments » |
Pareto Chart and Six Sigma
Posted by: meikah | 5 March 2007 | 3:41 am
If your organization is going though some quality improvement, I’m sure you have asked the following questions:
- What is the largest issue facing your company or organization?
- What 20% of sources are causing 80% of the problems?
- Where should we focus our efforts to achieve the greatest improvements?
These questions can be answered by using a Pareto chart, also called the 80/20 Rule. The assumption is that most of the results in any situation are determined by a small number of causes.
For example, 80% of process defects arise from 20% of the process issues or 80% of delays in schedule arise from 20% of the possible causes of the delays.
BusinessKnowledgeSource.com shares that a Pareto chart:
…is a type of bar chart where the values being plotted are arranged in descending order.
…graphically summarizes and displays the importance of the differences between different groups of data.
…is constructed by splitting your data into groups. These groups are also known as segments, bins or categories. The left-side vertical axis of the Pareto chart is then labeled Frequency, Cost or some other unit of measure depending on your data.
Source:
BusinessKnowledgeSource.com Manufacturing Info, “What is a Pareto chart and how is it used in Six Sigma” with link provided by Six Sigma Zone
Filed under: Tools/Toolkits, Data, Statistics, Pareto chart
No Comments » |
Using Six Sigma to Determine Business Value
Posted by: meikah | 1 March 2007 | 9:21 pm
For most companies, they determine their business value through return of investments, or ROI. To me that is not a very accurate way to get the real value of your business.
For one, a company’s bottomline has to be checked against the defects in the processes and evaluate. The ROI may register positively but how far is your business processes from the standard deviation, and how are you able to sustain your ROI through the years?
Tom Bowers, writing for SearchSecurity.com, shares his experience with working in the security department at a large pharmaceutical company. In his own words, they had success using Six Sigma to identify relevant data that can show the value of initiating a new project or technology. Instead of attempting to prove ROI, we used Six Sigma tools to define what can be measured, conduct the “measurement” and provide an analysis of that data to show business value to the CFO. The bottom line is always showing “business value” — hopefully in real dollars saved.
Basing on DMAIC, Tom’s department only used the DMA and added their own methodologies:
Define. The goal in this case is simply to identify events that can be measured. Let’s consider, for example, the theft of laptops storing valuable data. Another example might be a paper-based information risk audit thrown into a regular (versus shred) trash basket.
Measure. We decided what “units” will be used to define the measurement. What is measured and what units to use are completely based upon the process being measured. For example, a forensic examination can be measured as “each” or in dollar terms based on the information recovered or lawsuit won. The measurement for the example of the stolen laptops might be each or a dollar value (of the information and/or the device itself) – or both.
Analyze. We evaluated the business value of our measured events versus a planned security project, a headcount increase – anything that requires approaching the CFO for funding.
Next we looked regulatory compliance breaches, and then used external events such as information theft reports and regulatory noncompliance reports at other firms.
To create a larger statistical sample, we used freeware and commercial risk assessment tools.
For the IT security data points, we looked at our SEM/SIM, which provides log and event correlation that supports events seen in one location by providing corroborating evidence in other locations. These tools provide data points that are of higher quality and can speak more clearly of the ROI/business value provided by security in protecting infrastructure.
It’s good that Tom’s company uses the tools of Six Sigma, and enjoys the benefits brought by it. While the complete DMAIC process, may not be the answer to all process improvement initiatives, I think the Improve (continue to do better) and Control (institutionalize the improved system) phases of DMAIC will still prove useful in determining or presenting the ROI to senior management.
The popular question from management is always, “What’s in it for us?” or “What do all those figures mean to our business?” I believe the Improve and the Control phases will determine the impact or implication of ROI or data leading to the determination of ROI. This is where the value lies.
Source:
SearchSecurity.com, “Forget ROI; Use Six Sigma to prove business value” with link provided by iSixSigma.com
*Photo credit: MorgueFile.com
Filed under: Tools/Toolkits, Benefits and Savings, Finance, Data, Statistics, Processes
No Comments » |
Lean and Six Sigma Blogs Round-up
Posted by: meikah | 28 February 2007 | 2:31 am
Today, we’ll make the rounds of Lean, Six Sigma, or any quality blogs.
Mike Wroblewski of Got Boondoggle? has two interesting posts inspired by the book, My Life and Work by Henry Ford. Mike shared some interesting points about Ford and the Ford Hospital as the first “Lean” Hospital. I also like this quote: “It is not good management to take profits out of the workers or the buyers; make management produce the profits. Don’t cheapen the product; don’t cheapen the wage; don’t overcharge the public. Put brains into the method, and more brains, and still more brains-do things better than ever before; and by this means all parties to business are served and benefited.” Many management delay rewards and recognition till bottomline is stable.
Ron Pereira of Lean Six Sigma Academy talks about hypothesis testing, which you can use to test if two samples of variable data represent different populations. Six Sigma because it is data driven involves statistics, and it’s good to arm yourself with such tools. Read more…
Over at Quality Hero, Rob Thompson shares a non-believer of the successful marriage of Lean and Six Sigma. Like Rob, I also believe that the two can really be merged and can make a difference in an organization. Those who think otherwise may not be aware of it, but they’re actually working on the concept of Lean and Six Sigma. Only that they have given it another name. Possible, right? 
