Vishwadeep Khatri

Hi Prabhu, You may try this link - http://stattrek.com/lesson5/hypothesistesting.aspx

Here they are - http://academic.pgcc.edu/~ssinex/excelets/

Response Optimizer - Provides optimum solution (Unique) but can provide output for given set of inputs (common) Contour Plot - Provides a single view that indicates expected outcomes at various combinations (unique), Does allow flagging and display of output for a given set of inputs (common) Response Optimizer - Can be used for optimizing more than one responses and shall provide the best solution at one click (unique) Contour plot - Can help highlight pockets of great responses at a glance (unique) Response Optimizer displays how close you are to the ideal solution with desirability measure. (unique) While Response optimizer allows to play around with more variables and types, Contour plot can be drawn only when you have at least two continuous X's and are willing to keep remaining variables constant. With Response Optimzer, you need to know what you want or expect, while with Contour plot, you can explore with an open mind.

If the causes are known and the solution is easy to find, the project does not classify as a Six Sigma project. If the project is all about listing average time taken in each task and reducing or eliminating non value adding time, you may take it as a Lean assignment.

In absence of trust, such projects seldom succeed. The expectations of each side should be made clear and agreed to at the beginning of the assignment.

Good share. It also has a saving calculator that allows intelligent visualization of savings. http://www.cisco.com/web/applicat/TPcalcul/home.html

Interesting and very valid question. Subject matter experts may like to decide what is a good sub-grouping logic in the process. If hourly sub groups provide big between group variation (as compared to within group variation), then it should be taken as hourly. More comments welcome.

Good article. And agreed - Unfavorable randomization may need tweaking. Thanks for posting this.

Though this question was not directed to me, let me add a quick response. You use Chi Square test primarily to test dependence or independence. In other words, (using an example) you may wish to know whether the defectives % is same for all shifts. If the test disproves the null, you conclude that defective % are different. In other words - defective % DOES depend on shift or defective % is not INDEPENDENT of shift.

Standard VSM does not have scope for decision boxes as it is done at a macro level to primarily show FLOW of material, information, communication. Use wait time of one invoice from the moment it enters the queue till it is processed. Considering an average queue length of 100, each invoice becomes 100th first before becoming 99th or 3rd in the queue. Depends - Ideally you may do detailing on an area that requires action. Hope this helps. More comments welcome.

Just because Motorola decided to call it Six Sigma and not "5" or "7", as they thought Six is just the right goal. Also, the focus initially was primarily on variation reduction, which later shifted to a full fledged management approach.

Commenting here just to make this discussion "live" again.

Good to see the suggestions and comments. There is a need to finalize a metric. QFD is relevant and could be used here. To build the discussion back to Dina's focus( my feeling is that Dina needs some direction to identify a measure that relates closely to pain), may I ask everyone for ideas on CTQ options that we may suggest Dina to consider ?

Dear Deena, Each project is unique in itself. Most projects shall have confidential data and a situation different from yours. Please pose your specific questions related to the project. Best Regards, VK

Hi Krishna, What you have highlighted is an important topic and I am glad you raised it here. One way of working with this situation is Try to categorize defects as critical, major and minor. Define the requirement for each type of defect. For example zero defect is the expectation for critical defects, while up to 3% minor defects may be considered acceptable. Sigma level can be separately calculated for each category of defect, if you wish. Another option is Assign weights to each category of defect. Calculate weighted scores (similar to a quality score) Fix a weighted improvement goal for the process. Hope this answers your question. Do let me know if you have further queries. Opinions from others welcome . Note - Probably the CTQ you would like to consider would be accuracy (if you use first option) and score (if you use second).

No, as points beyond 3 standard deviation from the mean is considered a special cause indicator for (many of the) other distributions as well. Original usage of SPC was demonstrated for distributions that deviate a lot from Normality.

Stability can be checked using control charts. Although there is some debate on applicability of I-MR chart's usage for non normal data, the fact is that it works well in most cases. Xbar chart shall be ok as normality can be assumed for averages.

Interesting situation, Mayank. Output/time is the right metric. However, it seems the question is that of reliability of data. One method, though imperfect, is where the computer throws random number at the agent every few minutes. If the agent types it quickly, he gets a pass. If he fails to enter in a few seconds, it is a fail. You may allow some fails if total time is long. The method only checks that the agent is right there! May still miss whether he was really working on the process.

A real project shall always need validated real data. See if you can connect with some small business (even if it is a shop or small trading company) and carry out a project.

Hi Manik, Very important question for service sector professionals. Here is my answer - Anything that needs improvement should be measured first. If something is critical, you can surely find methods to measure it. In service sector - Dissatisfaction is measured by poor referral, poor repeat sales, or loss of customers. Employee motivation is measured by absenteeism, productivity, attrition. Success of social activity is measured by number of posts, new friends and "likes" Taste of liquor is measured by scores given by experts (with their tongue as measuring device) Quality of service is measured through carefully designed feedback forms. Success of schemes is measured by increase in sales. Success of email campaign is measured by clicks. Success of outbound sales is measured by various conversion rates - cold contact to leads, leads to hot leads, hot leads to accounts, accounts to up-sell I presume you got the idea. Let us play a game - I suggest you and other viewers to challenge the creativity/ingenuity of this community by providing situations where measures are tough to devise.

Good question, Prashant. Capability analysis for non normal data can be done after data transformation. You may like to try box-cox transformation for capability analysis. Make sure that the specification limits are also transformed.

Important observation there Mayank and your example fits perfectly here. Let us say a specific small difference (X) in proportions (or averages/medians) when comparing two populations (based on hypothesis test done with large sample data) is found to be significant. The same difference (X) may be absolutely insignificant when you take smaller sample sizes from the same two populations. This means if you work with insufficient sample sizes, small real differences in populations cannot be detected. Large sample sizes provide you with higher power (ability to detect small differences if they exist in populations) Views, anyone?

If you have gone through a structured learning, you can follow the steps starting with creation of a project charter. To start with, do you have a good business case (if successful, what is the business impact of this project?) if yes, you need to sell the business case internally.

Hi Sumit, It seems from your post that the reasons are known "below average aptitude or untrained Medical transcribers (both QA & editors) are responsible" If the cause is known, one can focus on finding solutions (using I phase techniques of DMAIC). A Six Sigma project (DMAIC) may not be needed, however Six Sigma techniques (as presented in I phase and C phase) can be used.

I have compiled quotes for Six Sigma beginners and my comments are in green. Feel free to change the green text and use wherever you like. However, the black text is protected by the wise men who developed them. Here you go.. "It's better to be prepared for an opportunity and not have one than to have an opportunity and not be prepared" - Whitney Young Jr *There is no better way to prepare than to learn the world best techniques in decision making "It does not matter where you are coming from. All that matters is where you are going" - Brian Tracy *Chances are - you may not be going anywhere with the old ways. "The right place and the right time is here and now" - Michelle Ustaszeski *Absolutely! "The more you seek security, the less of it you have, but the more you seek opportunity, the more likely it is that you will achieve the security you desire" - Brian Tracy *Taking this further - when you succeed with one opportunity, you get loads of them. "It is not necessary to change. Survival is not mandatory" - W. Edwards Deming *Bow. "The greater the difficulty the more glory in surmounting it. Skilful pilots gain their reputation from storms and tempests." - Epictetus *The BHAG is what you need. The big hairy audacious goals. Do not set unless you are prepared well to tackle them. "Big jobs go to the men who prove their ability to outgrow small ones." - Ralph Waldo Emerson *Undoubtedly "Your life changes the moment you make a new, congruent, and committed decision." - Robbins, Anthony *Make it now "Without a sense of urgency, desire loses its value." - Jim Rohn *Do it now "If you are not willing to risk the unusual, you will have to settle for the ordinary"- Jim Rohn "You must take personal responsibility. You cannot change the circumstances, the seasons, or the wind, but you can change yourself. That is something you have charge of." - Jim Rohn *Rohn says the same thing in so many ways "We can't solve problems by using the same kind of thinking we used when we created them." - Albert Einstein *Change the thought process. Welcome to the Six Sigma world. "The larger the island of knowledge, the longer the shoreline of wonder." - Ralph W. Sockman *After so many years with these techniques the wonders have not ended for me yet "Intellectuals solve problems; geniuses prevent them." - Albert Einstein *I can only take a bow on this. "First they ignore you, then they laugh at you, then they fight you, then you win." - Mahatma Gandhi *You need to ensure that the first three stages do not get extended too much "One that desires to excel should endeavor in those things that are in themselves most excellent." - Epictetus *Get all the ammunition that is excellent in itself. Even if you miss a few goals, it will not matter. As the jackpots you hit shall reap great rewards for you.