Rupinder N

This was a tricky question and the answer is that exact reversal of the same statement is not possible. Hence, one situation's null hypothesis, in the same form, cannot be alternate hypothesis in another situation. Keeping this in consideration, Ronaaq's answer has been chosen as the best answer - owing to the fact that this is the only answer that says so. Close second is Venugopal's answer in which he has indicated so with the help of an example. A big applause to everyone who attempted this question as it undoubtedly was a tough one.

This is a question that fetched us a lot of similar answers, especially about the definitions of maturity levels. The similarity in answers made it specifically difficult to choose the three best ones. Raghavendra Rao answered all parts of the question asked and also articulated the answer in his own words, though it could have been structured better to further improve the quality of the response. Venugopal R has mentioned how assessing process maturity and not rolling back is the way to go. Not verbose at all, very well structured. Rajesh Chakrabarty has clearly outlined how a process evolves and matures and also shared a tabular summary of maturity levels. This is the chosen best answer to the process maturity question.

This question again challenged us to think beyond what we always stress upon - collecting continuous data wherever possible.Kudos to everyone who made an attempt. Great examples provided by Venugopal, Mohan PB, Kavitha and R Rajesh. The chosen best answer is that of Kavitha's as she provides a direct comparison of examples where we have continuous data but still choose to collect discrete data.

It is heartening to see that the complexity of the topic did not deter so many of us from taking a stab at the answer. That shows the true spirit of excellence! There were two parts to this question, and that was the tricky part. Not all attempts answered both parts - the meaning of statistically significant and practical application. Needless to say, articulation in your own words is a key, too. The answer which was complete in all respects was by Mohan PB. Rajesh Gadgil and Venugopal R made a very good attempt too. Cheers to everyone who contributed.

We have the winning answer for this question. Thank you all for taking out the time to contribute to the World's Biggest Dictionary, during the festive season. The best answer is given by Venugopal R of Club 46. This answer clearly explains, with the help of examples, where over emphasis on VOC could cause more harm to the business than benefits. There were other answers that were close, but examples and articulation in your own words is always going to get some brownie points. Cheers to excellence!

Hypothesis Testing is among the most powerful tools used in Business Excellence. It takes away the decisions based on gut feeling or experience or common sense e.g. Site A has better performance than Site B, we should hire more experienced employees as their accuracy is higher, it takes lesser time if we use System A vs System B, older customers are less likely to use self-help as compared to other age groups, are we meeting the cut-off defective %age or not, based on the proportion defectives we see. Hypothesis testing allows to collect valid sample sizes and make decisions for population - it keeps the gut feeling and statements such as "in our experience" out of the picture. You have statistical proof of whatever you "feel" or "think" is right. What must be kept in mind is that it is an OFAT testing technique - only one factor under consideration can be varied while all other Xs must be maintained constant. Hypothesis Testing can be used in any and every phase of the DMAIC cycle. Define - Usually all "1" tests or tests where we compare a population to an external standard are used in this phase e.g. 1 proportion test (if I have x out of y defects, am I meeting the client quality target of 95%?), 1 Sample Z, 1 Sample T, 1 Sample Sign etc. (Has the cost of living gone up as compared to the mean or median cost 10 years ago?). It helps us decide "do we even have a problem". Measure - One can look at data and the eye can catch a "trend". But can we really say that the performance has dipped, is the difference in performance statistically significant. Hypothesis testing can give you the answer. Analyze - this hardly needs any explanation as everyone has using hypothesis testing extensively in this phase to compare two populations or multiple populations e.g. do the five swimming schools create the same proportion of champions out of all enrolled in them, is the lead time for a process on machine A better than machine B, does Raw Material X give better quality than Raw Material Y, does Training Methodology 1 give better results as compared to Methodology 2, 3 and 4, does Vendor A have fewer billing discrepancies than Vendor B etc. Improve - tests involving two populations are generally used. E.g. comparing Y pre and post solution implementation (we implemented a solution to improve the yield of a machine). Is the post-solution yield higher than pre-solution yield, is the TAT post solution better than the TAT before implementing the solution, are more customers buying our product than before etc. Control - We get different CTQ numbers every month post we made an initial improvement. Can we really say that we have improved as compared to before? For 5 months after improve, if we saw a lower number for the metric, was that really different than other months. Can we say that we are consistent? We can use Hypothesis testing again. Business Excellence is nothing but an iterative process to drive excellence throughout the business. As Hypothesis Testing helps us validate or invalidate what we suspect every step of the way in the DMAIC cycle, it is a "must use" tool for the armor.

Necessary - X is a must for Y to occur. Y cannot occur unless X is present. Sufficient - X is enough to cause Y. However, Z may also cause Y. Scenario 1 - Cause is necessary but not sufficient. X occured at some time for Y to have occured but alongside other factors. In this case, other causes that could have caused Y when combined X have to be found. E.g. there was a case of cars catching fire if hit from behind when the right indicator was on. Having the right indicator on was necessary but not sufficient for a car to catch fire. It had to be combined with the other factor of being hit from behind in order for it to catch fire. Hence, we are looking for critical combinations of other causes with this X. 2. Cause is sufficient but not necessary - means that X on it's own can cause Y. But this is not the only cause leading to Y. It is required in this case to make sure that other causes are also found out, else the problem may remain unresolved even when X is fixed. E.g. not having enough water in a day can cause headaches. But so can not eating on time. Even if you keep having water, but not having food could.still trigger the headaches. 3. Neither sufficient nor necessary - Even if X happens, Y will not occur. In this case this cannot.be deemed as a root cause. Solving for this X will be futile. Other causes ought to be explored in order for the problem to be solved. E.g. an executive assistant not having an app for calling a cab for her boss is neither a reason sufficient to not get a cab, nor is it necessary. A cab can still be called via a phone call, by asking someone else to order, or booked through a website, by hailing from the street. 4. Both sufficient and necessary - must be solved for as whenever X occurs Y will occur. If this is not solved, you have not resolved the problem

Dear Abhishek, As a Change Analyst, the tools mentioned above by VK may be helpful in your role, too. As a System Administrator, you may be responsible for the installation, support and upkeep of servers or systems. You may work on projects that involve fixing reasons for frequent breakdowns in a certain area. You may work on projects which involve reducing security policy breaches, time taken to resolve any issues/problems, reduction in escalation for resolution etc.

Dear Krishna, I see that you completed your BB post you asked this question. The BB training would have answered this question for you. However, for everyone's benefit, in order to determine sample size, the following are the pre-requisites: 1) What is the hypothesis being tested (what do you want to ascertain with this sampled data) 2) Level of Significance and Power of the Test In the example that you have provided, what are you trying to test using this data is not clear. If you could provide more clarity, I would be able to guide further.

Dear Forum Member, At the risk of over simplification, companies prefer candidates with Six Sigma knowledge because it tells the companies that these candidates will be able to put a structure to solving any business problem they are faced with. Anyone who has learnt the methodology will look at opportunities to address areas which could lead to business benefits for any organization they work for. Good luck!

Dear Rajiv, I am sure by this time you would have completed this project and many more. Gurshit's direction was good. In this kind of problem, you could start your analysis with time taken to make a call, time taken to make a good call (successful interview) by observation (or time in motion as it is called) and then working out the number of FTE required. If you find that you have too many surveyors, you will need to do an analysis around where the time in the day is being spent. Other data, I believe, such as these lines being used for other purposes could be easily available. This data can be analyzed and reasons eliminated or addressed.

Dear Pankaj, What you need to do in this case is "logical validation" which means that you need to ensure that the data you are capturing from the system has the same start and end points as defined on your Operational definition. MSA needs to be done only when there is manual assessment involved. You are using the start and end time from when the ticket is logged to when it is closed in the system. You will still need to ensure that the person who works on it closes the status real time and not as per their convenience. In TAT projects, it is typically seen that there can be a few outlier cases. If you remove the outlier cases, you could do a normality check again and use methods and tools that you would use for normal continuous data. But you should go this way only if your metric is TAT in number of days or hours. If your metric definition is in pass/fail, you should calculate the sigma level as per the defectives method.

Dear Rahul, Some things to keep in mind while writing a problem statement are - 1) Mention specific issue 2) Where is the issue occurring (company, deptt etc) 3) What is the trend in performance against expected target 4) For how long has the problem existed 5) How do you know the problem exists If you use these as guidance, you will arrive at your problem statement. E.g. what makes you think that telephony cost is an issue? Is It higher as compared to other similar businesses or departments? Is it the top spending area in your company etc. So you may have something like, The ISD call cost for the last 6 months were___ (which were higher than the previous year/which are higher than comparable businesses). The targeted cost is___ hence a gap of ___ and so on.

Dear Bharani, Going by the review effectiveness definition, you would use the P chart (where the defects not found are defectives, with variable subgroup size). Please note that the chart will calculate the UCL and LCL based on the data that you use unless you specify the limits, which can be done by clicking on P-chart options -> Limits. If you choose to test for all Special Causes (P chart options - > tests), it will show if the variable is out of control even for reasons other than >1 std dev from mean.

Dear Bharani, Here is an example. Let us say in Back Office data entry work, you are trying to test time taken to upload a transaction using Software A and Software B. If you collect data from different operators for multiple transactions, you will say that the external conditions are different, as the skills may vary by the operator, the comfort levels with two softwares etc may be different. However, if you collected data for the transaction for the same operator using both Software A and Software B, you would say that the external conditions are the same, since in the latter case you can truly test if the Software is causing any difference in time taken to upload and you have removed the "noise" that could have been introduced by different skills from different operators. In your case, you would have to elaborate on what is it that you are trying to test. If your intent is to test if the time varies by the tester, and you are keeping the code and the machine same (and nothing else is changing either e.g.length of the code etc) you could say that the external conditions are the same. This is what I can say with the level of information available.

Dear Vikram, You have already mentioned two approaches that you use. You are correct in saying that an "estimate" of the goal is made at the initial stages. Other approaches that are used are: 1) reduce defects by 50% 2) estimate what can be reached by eliminating top reasons, set stretch goals 3) use DOE to optimize the solution and settle for the goal that is viable for your budgets

Dear Ajith, Project Charter is a living document.It can be updated even after the Define phase. In fact, DMAIC cycle is such that you may have to go to any of the previous stages during the course of the project and it is ok to do so.

Dear Shalini, You may have moved on to the latest version of Minitab since you asked this question. Here is the answer for the benefit of those who are new to this world. You may create a C&E diagram to the 3rd level, however, beyond that Minitab directs one to create a new fishbone. 1) The main bone labels can be listed as column headers and the second level as items in the column 2) If you want to add a 3rd level, right click on the second level labels and click on "Graph Options" 3) Click on "Sub" for the branch where you want to add the 3rd level. Write the 3rd level bones separated by commas CE - Minitab.pdf

Dear Kumkum, This may have lost the time relevance for you, but here are the answers: 1) If you know that standardization is going to resolve most of the issues, there is no need to do a Six Sigma project. You just need to have an implementation plan for the replication of processes. However, if you are seeking to attain goals beyond what standardization can get you at, you can think of doing an improvement project. 2) In our experience, if your company does not have the Six Sigma culture and does not support it either, you could look for other avenues such as a friend or a relative's business i.e. hotel, shop, etc 3) I am afraid we are unable to offer freelancing opportunities as most of our clients request data confidentiality and only Benchmark employees to be associated with their projects. Wish you luck.

Dear Richard, By now, you may have worked on many projects since you completed your LSS GB training. However, for the benefit for other new professionals, I want to mention that the methodology can be used in any and every domain. There isn't one core field that Six Sigma applies to, though it may be a common misconception that it applies just to manufacturing. Click here to hear from the experts in manufacturing - https://www.benchmarksixsigma.com/manufacturing/ If you have any specific questions on an area of improvement in your current work area, please reach out.

Dear Premanand, Let me start by saying that the Lean and Six Sigma are methodologies that are applicable across industries and domains. The methodologies help you solve business problems irrespective of the organization and geography that you work in. If you come with those skills, it is intuitive that your skills would be widely sought after by any organization. The more you exercise the skills, the better you get and hence the demand for your skill also increases. To review a case study in Sales and read feedback from some of our participants from sales and marketing background, visit us at https://www.benchmarksixsigma.com/lean-six-sigma-in-marketing-sales-advertising-market-research-PR-business-development/ Please visit http://www.benchmarksixsigma.com/globalnews/ if you want to stay current on how it is changing lives across domains.

Dear Vishwanath, Anything that is outside of the specification limits is a defect. If the customer gets a lab test report sooner than 2 hrs, it will not be a defect. As is mentioned by Mr Nair, not all projects will have an LSL and a USL. What you have mentioned is what will add to the variation in the TAT. Let us take another example, if we order a pizza, we expect it to be delivered within 30 min. If it is delayed, that is a defect (as it doesn't meet the customer specified max limit or USL of 30min). However, this will have no LSL as the customer will not complain if the pizza arrives in 5, 10 or 15 min. The sooner it happens the better it is.

Dear Shalini, Some examples of DOE in the banking sector could be - 1) Lead Conversion Loans/Credit Cards - Factors could be Region, Interest Rate, Features and benefits provided in writing or not, Other Existing Relationships with the bank, Warm up Calls done or not 2) %age of Correct or Complete CC/Loan Applications - Rep assisted or not, Example Responses provided or not, Multiple language options available or not, Number of Questions on the form, Checklist of documents available or not 3) Promotional Campaigns Signed Up - Factors to be tested could be advertised on Website or Social media, Processing Duration, Self Sign up Option available or not, Length of Existing Relationship with the bank 4) %age Goods Loans Honored by Retailer - Type of Good, Tie up with Retailer or not, Cost of Good

Dear Syed, Shalini has given excellent ideas in her responses. As far as your question Some of these seem to be solutions to some problem areas and not problems in themselves as far as Six Sigma is concerned. e.g. why do you want to develop the laboratories? why do you want to add extra courses?- it seems those underlying problems are the actual problems that you are seeking to address. Alternatively, these could be Six Sigma "design" problems e.g. Design a process for ongoing laboratory development or Design a process for course change/addition in the college. What all should happen, or should be taken into account in order to decide whether laboratory needs to be developed further or whether a new course should be added - this becomes an ongoing process. This will follow DMADV approach. Whereas this could be an improvement problem "Current average academic and industrial collaboration count is X" which could be a cause of concern because it is lower than other similarly rated institutes and hence your goal could be to improve and sustain these collaborations at Y per year by a certain timeline. This will follow the DMAIC approach.

Hello Vamshi, If the intent was to just look at the pattern of "number of email" per 30 min interval, you could use a bar chart, with the intervals on the X axis and count or frequency on Y axis. Count of email is discrete or categorical data. However, if it can assume more than 10 different values, you may "treat" it as continuous data.