Forum Replies Created
Forum Replies Created
June 11, 2009 at 1:17 pm #184524
I have found that using Stepwise Multiple Regression with all the X’s in the Model along with Two Factor Interactions of those X’s is a useful tool to begin the screen of multiple X’s (some of which may be co-linear). Stepwise Regression is a natural way to identify & remove co-linear X’s from the model. Hope this helps. Have a great day, Tony
0June 5, 2009 at 9:04 pm #65292
I believe your question pertains to monitoring Key Process input variables that determine your Process Output. If this is the case, then I would monitor only those Key process Input Variables that significantly deterimine your Key process Output Variable. Typically, Multiple Regression Analysis or formal Design of Experiments are used to determine which process variables are Key.
As to the Key Output variable, I would Control Chart this until I was satisfied that in controlling the identified Key Process Variables, my Key Output variable remains well behaved. Thereafter, you might want to revert to a periodic audit sample of the Output.
Hope this is of some help!! Have a great day, Tony
0March 27, 2009 at 10:39 pm #182860
That is not very helpful.0February 16, 2009 at 12:06 am #62282
no i am actually an instructor monitoring this board to see who is cheating… JUST KIDDING! -No we are not in the same class or school.0February 15, 2009 at 8:51 pm #62280
your calculation is correct.0February 12, 2009 at 8:41 pm #62275
thank you so much that helped. Very much appreciated.
Tony0February 12, 2009 at 7:26 pm #62272
here is my email: [email protected] shoot me an email and we can take it from there. No we have different instructors. most likely a different school as well. thank you so much. I am at my zero hour..0February 12, 2009 at 7:14 pm #62269
would you mind helping me out?0February 12, 2009 at 6:05 pm #62266February 12, 2009 at 5:15 pm #62264
i think it is of the mean. I am coming up with .0879 and 7.5250
if it is from the the mean I am coming up with 0.4896 and 41.89740February 12, 2009 at 4:36 pm #62262
Yes a class project assignment.
The mean is 21.1935 . The standard deviation is 6.9013 . I am having trouble figuring out how to calculate the correct expected variation + or – 3 standard deviations from the mean. So far every answer I have given is wrong. The last hint from the instructor was: my mean is correct, my standard deviation is correct but my lower and higher points are not. Then he hinted: How wide do I beleive this distrubution is? In essence how far to the right and how far to the left do I have..(I think is what he is looking for)0February 11, 2009 at 2:13 pm #62256
It looks like we are doing the same health care project. Did you get an answer?0January 15, 2009 at 2:24 pm #179768
During 6 sigma training, I present the concepts of Taguchi’s loss function to stress that, although inevitable, any variation is bad and has some resulting effect on customer satifaction and therefore cost of poor quality. The message is “never stop reducing variation if it is still possible within obvious commercial constraints”
I agree with trevor – application of all you have mentioned is too complex for all but very sophisticated environments and I would focus on the concepts and principles in an attempt to gain some incremental benefits.0November 19, 2008 at 9:31 pm #177857
Calculating capacity is very easy once you know 2 things: Volume and Productivity.
Here’s a simple equation:
P(roductivity)= V(olume)/TM (Team Member)*H(ours)
The first step is know the maximum hourly productive of the Team Members (TM), (e.g.) if TM 1 has an hrly productivity of 8, TM 2 is 6, and TM 3 is 4 – then (assuming that each TM is equally trained) the hrly productivity rate is 8. Therefore, the Team ought to be able to do 24 tasks an hr. or 192 per day.
To accomodate more volume, you either going to need more TMs, more hours or more productivity.
You’re a self-employed wigget maker and you’re making 8 wiggets a day (1 per hr). Your customer now wants 16 wiggets per day. What are your chioces? You can hire another TM; or you can work 16 hrs; or find a way to increase productivity to 2 an hour – this is where process improvement becomes so important.
0October 15, 2008 at 2:18 am #59462
I am also interested in receiving a copy of the storyboard for the mortage processing cycle … would greatly appreciate if you could send to me by email also.
0July 10, 2008 at 7:44 am #173680
I agree with Daves point. There is a non-linear relationaship between dBm and mW and the distribution of the data in each type of unit should be explored to ensure normality by transformation if necessary. At the end of the day “which unit of measure is most important to the customer – dBm of mV”? That is the one that should be used.0April 11, 2008 at 3:35 pm #171037
I have the customer specs already in my database. I just need a matmatical equation that I can use to determine the sigma value using the specs as well as the deviation value. Any idea?0April 10, 2008 at 5:59 pm #170964
I agree that at the “company” level there are no Six Sigma performers. However, if you look at processes, there are many.
One of the classic examples is aircraft safety as defined by number of passenger flights without casualties vs total flights taken. This process is much higher than Six Sigma, and we wouldn’t tolerate much less. Baggage handling – same industry – about 3 Sigma (more or less depending on which year you look at).
Quality (and safety) are engineered into aircraft, and can also be incorporated into baggage handling given the right priority.
0April 4, 2008 at 5:31 pm #170594
Here is an older example:
Suppliers: Cain, Abel
Specification: Animal offering (blood sacrifice)
Customer received from Cain: Fruit of the ground
Customer received from Abel: Animal offering
Result: Abel’s offering accepted, Cain’s offering rejected
COPQ: Jealousy–> murder–> Cain cursed by God
Lesson: Know your customer’s CTQs.0March 14, 2008 at 1:38 pm #169679
Proven data-driven continuous improvement methodology.0March 4, 2008 at 6:52 pm #169251
I totally agree.
A wise MBB taught us that correlation does NOT imply causation. High correlation between two variables means you need to dig deeper. You need DOE or other tools to verify cause and effect relationships.
Another great example of jumping the gun on correlation: A large city newspaper once reported that fires which caused the most amount of damage had the most number of firefighters on the scene (implying more firefighters cause more damage), when in fact it was the size of the fire that caused both.
BTW, my current company is including Change Management and Project Management in our Green Belt curriculum.0November 29, 2007 at 9:36 pm #165504
The benefit/cost saving of inventory reduction in my company count only interest saving since these materials are still usable. If reduction of obsolete items they can count full cost of materials, plus interest.0June 7, 2007 at 4:09 am #157101
Nice to meeet you in this site.I am now working as a buyer in shanghai ,mainland China.I know six sigma is a good tool for qaulity assurance. I hope that i can comunicate with everyone who knows this knowledge. I am looking forward to your reply via my Email:[email protected] or this email adress MSN for interactive comunication.
Tony.Chau0June 8, 2006 at 7:01 pm #138845May 8, 2006 at 5:37 pm #137402
Could anybody post more examples of using Six Sigma in a Financial or Internal Audit capacity?0February 21, 2006 at 5:42 pm #134071September 27, 2005 at 10:11 pm #56784
i work for a large multinational in Ireland, we have been using SS for about 2 years now but it has really taken off in the last year or so. i would say a lot of other multinationals operating in ireland are doing so at the moment. lets just hope for our sake the all take it up and there is a severe shortage of BB’s $$$$$$$$
tony0September 7, 2005 at 1:40 pm #56767
To add to the other comments, I would suggest also plotting the data on say Individual & Moving Range Type Control Charts if appropriate for your process. Then examine the plots as to the presence of any assignable cause(s) which might explain the unanticipated lack of Normality in your data. Good luck, Tony0July 26, 2005 at 6:49 am #123623
I thought this forum provide insights. A good place to get genuinue feedback.
Yes, I got the answer, value of IQF, etc.,
But guess what the reputation that i had about some like Stans, Darths or the world is gone by now.
Six Sigma is evolving & variation is part of life. It is understandable that some have higher standards than others. A jungle should have tigers, lions, elephants , snakes, deers, etc. to have an balance.
Probably IQF is the deers and consulting firms that Stan & Darth are the lions..
Stan, Darth …. i want to get certified by the lions of the world… Do you certify? what is the criteria ? By how much will it burn my vallet
0June 30, 2005 at 1:08 pm #122392
Assuming Normality and Stability of the process data, I would expect that the overall yield experienced would be lower than that yield projected by Cp or Cpk statistics since we can view the overall yield as being ‘long term’ based upon ‘long term variability’ as compared to classical Cp and Cpk statistics which are based upon short term variability. In a nutshell, Long term variability > = short term variability. You might want to investigate the Pp and Ppk statistics which are based upon long term variability estimates and compare the projected yields from these statistics to that which you have experienced. Hope the above helps. Tony0June 23, 2005 at 1:03 am #121985
Who is the expert on lean six sigma process mapping analysis?
Tony0June 22, 2005 at 8:52 am #121919
You are right! You must determine the logic relation ship between” X”s and “Y”. For an example, when the sun rises, the rooster begin to crow. We can not say “rooster crow” is the reason of “sun rise” . But we can say ” sun rise” leads to ” rooster crow”. And make sure of that “X”s are independent.
0June 1, 2005 at 12:55 am #120488
RIP=raw material in process:
Tony0May 30, 2005 at 2:40 am #120408
Tony0May 27, 2005 at 4:18 am #120310
Thanks very much! It do great help to me!
my email address is [email protected], wish be friend with you!
Tony0May 27, 2005 at 1:23 am #120302
Can someone who has experience on it help me?
Tony0May 26, 2005 at 1:23 am #120207
Thanks for your instruction and your humor.:)
But I do not clearly understand still, do your mean the formular in total variation in MSA and variance in Stat > Basic Statistics > Display Descriptive Statistics is the same? What is difference is only df number? I copied all the data here, can you explain it more detail?
Total variance in MSA is 2.3458 but variance in Stat > Basic Statistics > Display Descriptive Statistics is 2.03062
Two-Way ANOVA Table Without Interaction
Source DF SS MS F P Part 5 66.0719 13.2144 76.7655 0.0000operator 1 0.0078 0.0078 0.0453 0.8329Repeatability 29 4.9920 0.1721 Total 35 71.0718
Source VarComp StdDev 5.15*Sigma Total Gage R&R 0.1721 0.41490 2.13672 Repeatability 0.1721 0.41490 2.13672 Reproducibility 0.0000 0.00000 0.00000 operator 0.0000 0.00000 0.00000 Part-To-Part 2.1737 1.47435 7.59290 Total Variation 2.3458 1.53162 7.88782
0May 25, 2005 at 7:09 am #120104
Thanks very much!
Tony0May 11, 2005 at 6:11 pm #119346
Get your hands on a copy of ISixSigma Magazine, the printed version. The May issue had a fantastic summary of compensation for the six sigma space. It’s really phenominal in it’s breakout.
TonyASQ Certified-MBB0April 21, 2005 at 9:50 am #118186
AndyP Tks! But my question is my %contribution is very low, but %tolerance is very high. and all my data is actually in my specification.
Tony0April 21, 2005 at 2:10 am #118172
Thank you very much! But do you know how can we get mode in Minitab?
Tks!0April 7, 2005 at 5:09 am #117342
Kipper, you definatly need to sit back and take it all in for starters. Take some notes on how things are done and then start asking Why ? when you get the answers be on the lookout for “thats just the way we have always done it”. If there is no other reason, I have found that is a good place to start looking for improvements.0March 24, 2005 at 1:38 am #116746March 23, 2005 at 5:28 am #116703
Urgent, can someone who has experiment on it help me?
Thnaks very much!0March 23, 2005 at 1:01 am #116682
Thanks for your feedback very much!
For this measurement system analysis, I think we can eliminate “man” factor for the its electronic display. My concern is if the equipment suitable for detecting the variance of temperature.
Tony0February 20, 2005 at 7:38 pm #115133
Check out isixsigma magazine, http://www.isixsigma-magazine.com/. I received a copy for being an isssp member and it has a section devoted to ‘final reports’. They show you what the PPT slides look like as well as discuss what topics should be discussed. It’s pretty decent.
Aside from that, the only resource I’ve found are other presentations at my own company. The problem is that I have to weed through the good ones and bad ones and then tend to be more of the latter than the prior, unfortunately.
Tony0January 10, 2005 at 3:45 am #113276
Slava- That’s what the job shop is for (https://www.isixsigma.com). I think this is an inappropriate use of the forums if you are trying to find work by posting like this. If you are trying to find help on how to improve your skills to find a better job, then consider rephrasing your question.
Tony0December 8, 2004 at 1:11 am #111979
Thanks for your all answers! But can you have a detail instruction on these three questions?
1¡¢What is the real role of BB in organization?
Just do the project whihc selected by MBB and function leader or need more functions? 2¡¢What is a real excellent BB? Comapring to the paper BB, what is a real excellent BB? How should be evaluate a real excellent BB? What and waht degree should a BB do to became a real excellent BB? 3¡¢After the discussion above two, let us go to operation issue, how can we be a real excellent BB?
Tks!0December 6, 2004 at 7:08 am #111813
Can you speak more?
0November 10, 2004 at 12:39 am #110524
Thanks for your help very much!
And can you paste your email address or send email to me([email protected]), I have some sample about ANOVA calculation and want to discuss with you.
Tks!0November 9, 2004 at 1:48 am #110475
Thanks for your feedback!
But I need is not for the interpretation but the detail clculation method about the ANOVA in DOE. Can you help me? My emaill address [email protected] .
Tks!0November 8, 2004 at 11:49 pm #110469
Can some experts who has experiments on it help me?
Tks!0November 8, 2004 at 12:18 pm #110401
I request you to pl send me a copy of the case to my emailid tonsunhlj126.com . I am BB in china.
Thanks in advance.
Tony0November 5, 2004 at 4:57 am #110313
But can you have detail explanation on block handling and analysis when the block is significant and it is unontrolled really?
And can we act “the voltage at day and night” as block and deal the difference between the day and night which is uncontrolled clearly?
0November 5, 2004 at 4:35 am #110308
Yes! I agree you. We should do further analysis about block if it is significant, such as act it as a factor then do DOE analysis, or get detail difference in block using graph or other simply statistic tool…
But, unfortunately sometime, we can do nothing on it even it is a significant factor also. What should we do then? I think Taguchi can handle it, act block factor as noise and analysis S/N ratio then find optimize setting. But what should we in normal DOE, is the method I mentioned before right?
0November 5, 2004 at 2:00 am #110303
Thanks very much for your kindly and professional instruction!
But, sometimes, the block is real uncontrolled and we can do nothing on it. And someone told me just has a simple way to deal it:
1/Design the DOE and implement it with the consideration of block;
2/After we finished the DOE with the block included, just ignore block item and do DOE analysis without block items;
3/Then we can use DOE responser optimizer to get out target/max/min.
And he also said block can get the same result and handle the same function of Taguchi.
Do you agree?
0November 4, 2004 at 5:40 am #110236
But what should we do when BLOCK is significant? As we know, block is “uncontrolled”. Can we do DOE with the block involved and analysis it just avoid block?
Tks!0November 3, 2004 at 11:56 pm #110232
Thanks for your feedback!
Yes! The data should be positive, but even the positive data, we will fail at box-cox transformation, my question is waht kind of positive data can be available at box-cox, and what kind of can not?
Tks!0November 3, 2004 at 2:54 am #110162
John , Ray:
Thanks for your feedback!
Yes, column with text or cells with negative or zero values will be hided and can not do box-cox transformation. But also we may not successs even we meet the erquirements of Mnitab, do you know what kind of data(meet Minitab box-cox requierment) we can success?
0November 2, 2004 at 3:50 am #110107
Pls somebody who has experiments on it help me.
Tks!0October 28, 2004 at 5:46 am #109859
Hi all: thanks for your all help firstly!
I know this function in Minitab-14, but can you explain the detail meaning for all those fourteen distribution, or where can I get the explaination(better is in minitab help)?
0October 27, 2004 at 11:53 am #109767
Do we have some fomular to calculate the specimens for each trail? I think 20 specimens for the experment which baseline is 2/10 is too small, it may not distingish te difference.
Tks!0October 26, 2004 at 1:17 am #109699
Thanks for your instruction very much! Many warmhearted experts here and that is why so many people here. And I saw “Forums Say Goodbye to Stan and Darth”, pls forget it and just continuous, or this forum will die offf just as aBBinMN said.
And do you mean we can perform control charts (xbar and R,s) for continuous value even they are not normality distribution? One of my friend asked me how to deal SPC for nonnormality distribution.
0October 21, 2004 at 2:23 am #109459
Thanks for your feedback first!
Pls see belowing sample from Minitab help file, we can find Part and Operator*Part is significant factor for their P-value is less than 0.05. So, my question is what should do on it?
Two-Way ANOVA Table With Interaction
Source DF SS MS F P Part 9 2.05871 0.228745 39.7178 0.00000Operator 2 0.04800 0.024000 4.1672 0.03256Operator*Part 18 0.10367 0.005759 4.4588 0.00016Repeatability 30 0.03875 0.001292 Total 59 2.249120October 20, 2004 at 11:53 pm #109451
Can somebody who has experience on it help me?
Tks!0October 20, 2004 at 11:52 pm #109450
Can somebody who has experience on it help me?
Tks!0October 18, 2004 at 2:32 am #109212
But how should we determine the sample size for each experiment?
tks!0October 18, 2004 at 2:02 am #109211
Can you post a sample on how to transformation of attribute response?
Tks!0October 14, 2004 at 12:45 am #109053
Thanks for your instruction!
And you mean we can use binary logistic reression for attribute DOE anaylysis, but can binary logistic reression provide us both critical Xs and the best setting for excellent output performance, would you pls send a copy of binary logistic reression sample to me? My email address is [email protected]. Thank you firstly!
Can you have a further explanation about “Generalized Linear Models” (unfortunately, I have not Montgomery’s Design and Analysis of Experiments on my hand or in my locak labrary)? Does “data transformation ” you mentioned is Stat > Control Charts > Box-Cox Transformation. And do you mean we can transfer the attribute data to variable then go to the normal DOE analysis?
0October 13, 2004 at 7:56 am #109009
Can somebody who has experience on it to help me? It is urgent? Tks!0October 13, 2004 at 5:51 am #109006
Thanks MJones Tim F Darth mjones
Now I get it. Tks!0October 13, 2004 at 12:06 am #108999October 12, 2004 at 11:51 pm #108997
Can sombody who has experience on it help me?0October 12, 2004 at 12:02 pm #108948
Can somebod who as expeen on it help me? Tks!0October 12, 2004 at 7:43 am #108934
Thank for your instruction very nuch, and I am a BB, but we get training from consultant firms and we have no MBB, so I may ask you some basic concept.
For interaction, do you mean Minitab can not do any further analysis when we doing anove or DOE? What can we do is go to real process and find solution with our process knowledge? But we always go to statistic for lack of process knowledge.
For the confounding, do you mean we need more experiments on it to avoid confounding?
Look forward to your answer! Tks!
0October 12, 2004 at 5:57 am #108930
thanks for your feedback first, but pls go ahead, what is the sigma realy mean?0October 12, 2004 at 2:12 am #108917
Just as mjones said in previous, the test looks at the two distributions and compares them.
For the one sample T, we simulate a distribution and compare to our target mean, if the mean is in the confidence interval, we can determine it is insignificant ,otherwise significant.
Is it right, and can somebody expand it to two sample T test?
Tks!0October 11, 2004 at 11:38 pm #108907
Hi, Can sombody help me?0October 11, 2004 at 8:24 am #108839
Can somebody provide a graph to explain 2-Sample T.0June 8, 2004 at 6:30 pm #60146
Case cart completeness and sterility, Environmental Services responsiveness to clean room, Anesthesia issues, PAT delays………0June 3, 2004 at 2:28 pm #60136
Hi Barb.If you haven’t done so already, I recommend that you complete a detailed process flow diagram of the turnover process. This will identify bottlenecks, delays, rework, etc. Also, the complexity of the process can also be an eye-opener for your team members. From there, I would move on to an FMEA. That seemed to work well in my OR project. Identifying the steps of the process that can fail and potentially lead to significant turnover delays can be a real breakthrough for your project.Hope that helps.0May 12, 2004 at 2:11 pm #60117
I’ll first make the disclaimer that this was my initial BB project, and I would do some things differently if I were to do it again.
With that being said, I used most of the traditional Six Sigma tools. The FMEA was particularly helpful in identifying steps of the process that caused delays in TAT. The team collected additional data surrounding case cart completeness and sterility. We identified both of these factors as significant Xs. The case cart issue has now become its own Six Sigma project.
We also collected data on several time segments within the OR, i.e. room available to cleaning arrival, cleaning time, room ready to pt. in, etc. This shed some light on Anesthesia’s influence on the TAT.
If you have access to the Healthcare Advisory Board, I would recommend looking at some of their materials regarding OR efficiency and TAT.
0May 10, 2004 at 7:34 pm #60114
My training project to become a BB dealt with reducing the turnaround time between cases in the OR. We found that Processing and Anesthesia issues were our biggest factors.0March 22, 2004 at 7:19 pm #97203
The salaries below $50K may be in other countries. Did you check that?0March 3, 2004 at 3:14 pm #96329
Thanks a lot. But I still feel difficult where to start.
So far, I have a big challenge to shorten supplier’s leadtime.
For the critical suppliers, I have provided enough forecast and confirmed weekly PO, how to ensure they shorten the leadtime?
If so, how to define.
A good beginning is an half success.
0March 3, 2004 at 12:32 pm #96318
I encounter the same problem!
To improve soucing process may be a good topic for puchasing.0January 12, 2004 at 4:45 pm #94157
We have done a lot of work in that area. To address your points:
– to improve (reduce variability and injury rates) your production system, I suggest determining the root causes of the lifting problem with an appropriate manual material handling analyisis tool, and removing the causes. You should also be able to tie in your improvement idea to lean through a reduction in wasted motion and then subsequently calculate a ROI.
– awareness and training are helpful, but no matter how you tell someone to lift a 200 lb object off the end of the conveyor, there’s still a problem there. Awareness and training should be focussed on use of the improvement (lift assist).
– implementation of the improvement needs to be completed through a job improvement process so that no other issues are missed (or created). Most ergonomic improvement tools nicely fit in the DMAIC framework, so a proven approach can be utilized.
Hope that helps.0December 3, 2003 at 6:57 pm #59983
Thanks Ty. Your thoughts are helpful.0December 3, 2003 at 6:28 pm #93184
Why won’t your teacher help you? Why do you think that is? Do you think that if your teacher thought you had actually given it the “college try” that s/he would give you a pointer? Perhaps you’re just looking for the easy way out of this situation instead of actually learning to appreciate the finer points. Give it some thought and be honest with yourself.0December 3, 2003 at 4:43 pm #93175
Is this the state of education today? Having to post your question to a forum for help? Why don’t you try working it on your own? You’re never going to learn anything by asking others for the answers. Have you heard of office hours…they’re trained to help teach you topics, not give you the answers. Shame on you for asking the forum to do your homework. That’s not the purpose of this space.0November 24, 2003 at 2:12 pm #92879
It is very typical to gather data in Measure, then move on to Analyze and look at the data — root causes and all — only to realize that you need more data, or didn’t collect all the data you need to determine your root causes. This is especially true on GB projects where the BB or MBB coach is not fully engaged in the project definition and data collection plan creation. Don’t fret — just be pleased that you realized that you need more information. It’s a learning curve!0November 21, 2003 at 4:52 pm #92807
SSS,That is great advice! Thanks for sharing.Tony0November 20, 2003 at 4:18 pm #92752
Here is a thread with that article link and another:
0November 11, 2003 at 4:03 am #92357
This topic is regularly beaten to death in the software area of iSixSigma. If you haven’t seen it, I do recommend browsing it’s discussion area: http://software.isixsigma.com/forum/0November 11, 2003 at 4:02 am #92356
Are you looking for dashboard software, or project management software? Or both? It sounds like just process management, so check out ProjX and Instatis. I think they may be able to meet your needs. Otherwise, I think someone with database skills in your IT department might be able to point you in the right direction.0November 3, 2003 at 1:53 pm #91977
I really like the analogy! Just for the sake of further discussion and “deeper” understanding, if a DOE were to be conducted what could the high and low settings be for the two factors water and temperature. I foresee temperature settings being hot (+) and cold (-), but how would we deal with water being wet (+) or dry (-) ??0October 13, 2003 at 2:01 am #90931
I have got great information from your feedback. But, I still do not know how to get typical samples in r%r study clearly, how do your get your samples in your r%r study to make it typical and present process variation enough? Thank you very much!
0October 11, 2003 at 6:57 am #90911
Thanks for your feedback, we all know the sample should represent the variation for your process, but know can we do it?0October 11, 2003 at 12:54 am #90908
Who can help me? How can I get typical samples fot r%r study.0September 24, 2003 at 5:38 am #90223