All Six Sigma LSSGB certification exam dumps, study guide, training courses are Prepared by industry experts. PrepAway's ETE files povide the LSSGB Lean Six Sigma Green Belt practice test questions and answers & exam dumps, study guide and training courses help you study and pass hassle-free!

Six Sigma Measure Phase

1. CLT & Basics Of Measure Phase

I've taken this excerpt from this particular source. You can read it in your own region. But let me explain in my own words. You would never have access to the entire higher population if you wanted to do data analysis. Think about apples, right? Assume Apple wants to conduct a quick survey to find out what features iPhone 6 users dislike and what features they want in the upcoming iPhone 7 model.

What would you do? How would you go about gathering the data? How would you go about collecting the data? Right? You'd obviously not get sufficient time, money, and funding to go and ask each and every iPhone 6 user what features they dislike and what features they want in the upcoming model. That is not possible.

It will probably take ten years or so if you want to gather the data from all iPhone 6 users. And by the end of the analysis, there will be an iPhone 15 or 16 on the market. It would become irrelevant for you. Hence, what do we do? We pick up a small sample, and simple random sampling is the best known sampling technique by far. You can also tend to or opt to use statistical sampling, but it all depends, right? But simple random sampling is the best sampling technique.

Alright, so what do you do? You quickly select a few people at random from the entire population of iPhone 6 users, and then you begin analysing data. On top of that, while you analyze, it is extremely important for you to know what distribution the data follows, or what probability distribution the data follows. And most of the techniques would assume that the data follows a normal distribution, which is welcome, or what is also called a Gaussian distribution.

GEOSIN gaussian distribution All right. Now suppose your population follows a uniform distribution. How does a uniform distribution look like? It looks like this uniform. All of us are uniform. Suppose your population follows a uniform distribution. And suppose I'm picking up a sample from this population. The Central Limit Theorem states that no matter what distribution my population follows, if my sample size increases significantly, data will always follow normal distribution. Look at this. Now, n is equal to one.

Here, it is like uniform distribution. When N is equal to 2, it's a somewhat triangular distribution. When N equals three, it indicates that the distribution is normal. when N is equal to four samples and N is equal to five. Now, it follows a beautiful normal distribution, right? So that is your central limit theorem. It says that no matter what distribution my population follows, if the sample size is large enough, the data tends to follow a normal distribution. That's the beauty of the central limit theorem.

So, assuming your sample size is sufficient, you can simply assume that the data follows a normal distribution and proceed with your analysis. All right, so this is extremely important regarding the distribution of your data follows. Because your entire analysis would depend on what distribution the data follows. Okay, here are the three steps of the measurement phase that we will discuss. Indeed, they contain the critical qualities of high performance. We will look into the measurement system analysis, and we will also look into the current performance measurement. Let us move on and discuss a few more things now. Okay, we are moving on. All right, here's a measure for you. What is a measure? A number or quantity that records the directly observable value of performance Here are a few examples.

length of time, probably speed, or age of the person. You can also measure size in terms of length, height, and weight. You may want to measure cost, sales, revenue, and profits. Accounts of characteristics or attributes are also measured. What is the property size? What is the gender? Male, female, and so on and so forth. Counts of defects, number of errors, number of complaints Right? So you tend to measure all these. It can be a number or a quantity that records a directly observable value of performance.

Why do you need to measure? I need to measure because I want to establish a baseline for current performance. What is my current performance? How am I operating now? This will also help you prioritise the actions and decide whether or not to take some action. Only if I can measure can I do this kind of analysis. Prioritization of actions can only be done if I measure. I will be able to justify the magnitude of the problem and the changes to the process. Based on the magnitude of the problem, you will also be able to predict future performance. People use a variety of dating apps, particularly in the Western world.

There are a lot of dating websites and dating applications, right? These dating apps can predict your match based on your gender, location (geographic location), body characteristics, orientation, and what you've been looking for all along. Systems have become extremely important because people are measuring things, right? And what are the next few steps? What are the first few steps in the data collection plan? What do you need to do? What do you know about your process and where to look for measurement points? You need to first find this out this. Do you even know your process thoroughly?

Where do you find the measurement points? Where do you get these measurements from? What data are you planning to baseline? Right, you have a problem at hand, but what is the data that is required to be the baseline for that particular problem? What are a few of the upstream factors that might affect the process or problem? What are the factors that might impact your problem or the process? You need to look into that aspect. What do we plan to do with the data once it is gathered? Obviously, you will have to do some analysis, but what analysis are you planning to perform that is extremely important here? Let us move on and understand what and where you want to measure.

So this is the most important part. We discussed the Simplifying fields supplier, input, process, output, and customer when creating the side part diagram, right? You prepare the cyborg diagram, and a more detailed process map can actually help the teams select what you want to measure. And choosing good measures requires that you have an extremely clear understanding of the definitions and the relationships between inputs, processes, and outputs. You need to know that. Look at this diagram.

There is an input in the form of an insect. There's a process that happens inside the input, and you have the output, which is an egg, right? That's a process. I have an input, there's a process, and then I have an output. Look at these inputs and processes. X stands for predictor, or input variable. Another name for an input variable is predictive. These are leading indicators because even before my output is generated, I have a chance to tweak my inputs and process them to get the desired output. So whatever work I perform on X occurs before something is delivered or accomplished. Hence, these are called "leading indicators." On the other side of the spectrum, we have Y, which is an output variable.

Another name for this is response. Response variable. These are called lagging indicators. Why do you call these aslagging indicators, by the way? Because I can't do much once I have an output, and I can't take many corrective actions, right? Or if something has gone wrong, I'll have to fix things here. It's a waste of effort and energy, right? So I'm not given sufficient time to ensure that I take corrective action only if something goes wrong. I'll come to know that it has gone wrong, and it is called a lagging indicator, right? So your process measures refer to x variables in data, right? Sorry, but your process measures have failed here.

So let's first discuss input measures. Input measures, referred to as "x variables" in the data, quantify your quality, quantify the speed at which you are working, and they quantify the cost performance of the information or items that are coming into the process. It quantifies the items that are coming into the process. Usually, input measures will focus on effectiveness.

Does the input meet the needs of the process? Yes or no? Right? You get to know about such information. Process measures are also referred to as "x variables" in the data. This also quantifies the quality, the speed, and the cost performance at key points in the process. Some process measures will be subsets of output measures. For example, cycle time per se If you try to identify cycle time for each step of the process and add the total cycle time of each and every process step, you will get the correct and worthwhile total cycle time.

Process and input measures provide an early warning of problems and are key to finding root causes. And it also helps you catch problems before they become serious. That's the most important part. Alright, now let's look at the first step of the measurement phase.

2. Define CTQ Performance Characteristics

Alright, let us look into this example. You must commute to work on a daily basis, right? And, as you commute to work, what is the most important factor you consider in order to reduce time spent at the office? You should probably take a different route, a different vehicle, and arrive at the office earlier. That is, if you are in the early stages of your career, isn't that what you tend to do? If you are measuring the arrival time, then what is the data type?

Is it continuous or discrete? Time is always continuous. What is the significance of the term "continuous"? I can say I've reached the office in 1 hour, or I can say I've reached the office in 60 minutes. Or I can say I arrived at work in one minute or less. Or I can even put that into seconds, right? 3600 seconds. Or I can go to any precision level that I want. I can go to any decimal level that I want. It is still going to make sense for me, right? No matter how much I divide the continuous data, it's still going to make sense for me.

Now that is called a data type, right? Measurement Unit How do you want to measure the time taken to reach the office? If you're a junior employee, you'll probably measure the time it takes to get to work in minutes. Want to quickly reach the office and get the job done? When you reach the top of the corporate ladder, you begin to measure time in hours. You're not concerned with getting to work in a timely manner. You probably want to get to the office as soon as possible. That's fine with you; if you become vice president, you'll probably measure time in days, right? Even if it takes days to reach the office, that's fine with me. I'm vice president of the organization.

When you get a CX job, you're probably going to measure it in months, quarters, or years, right? So the time it takes to get to his office really doesn't matter for a CX, right? He works from anywhere. All right. Jokes aside, the unit of measure is something that you have to identify upfront. And then you have an operational definition. Look for an operational definition using this example. And then we'll see a few more examples. Because this is extremely important. What would you say if I asked you how long it takes you to get to work from your house? You probably say 15 or 20 minutes. That's fine. But how are you measuring the time from your parking lot in your apartment to the parking lot in your office?

Is that the start and end time, or right from your doorstep until the time you log into your system in your cabin? Is that the time you're measuring and claiming as the time it takes me to get to work? Or what is it? If you want to compare two people's time taken, if you ask person A and person B how much time they take to our office, different people may have different definitions, right? One person might say, "I measure it right from the time I put on my tie until the time I connect my laptop and log into my machine." That is the beginning and end time. Based on that, I'm going to calculate my time spent in this office. Some other person might say, "No, from the time I start the car till the time I switch off my car in the office." That would be a sad and time-defining definition for me, right? So different people might measure it differently in order to ensure that everyone measures it in a similar fashion.

You have this operational definition in place. You also define the specification limits. You might say that I want to be in the office not before ten minutes and not later than 20 minutes. It should be anywhere from 10 to 20 minutes. But I also want to get there in 15 minutes. That's my target, which I want to achieve. Look at this rugby ball. Think about this. You ask a person, "What is the circumference of the raghika?" A person might say it's 58 to 62. 58 to 62. I do not see that number here. It's 580 to 60. So, what is the unit of measurement you're employing? That person might say, "I'm using centimeters." Oh, now if I convert that into millimetres, this number would come up. All right. Whatever you have told me is fine. If you ask some other person, "What is the circumference of a rugby ball?" You might say it's 23 to 24 inches, right? If he does not use this unit of measure, you'd be surprised. At 23 to 24, I do not see that number anywhere.

And then you'll ask me, "How are you measuring it?" What is the unit of measure? Then he says "inches." Alright, if I convert inches into millimeters, 23 to 24 inches would be 580 to 620. That's fine. So depending on what you are measuring and how you're measuring, things change. So the unit of measure is also extremely important. Now let us go to the operational definition. This is extremely important. We have looked into that briefing in our commuting to work example. Let us discuss this man. Who is this person? It's Usain Bolt, the reigning running champion. He invents almost all the races, right underneath the dash. 200 meters, 400 meters, I think she's going to beat anyone on the planet. So let me ask you this question:

What is the proper start time for an athlete to start running? 100-meter dash or 100-meter race? You all might have seen the Olympics, right? You all might have followed national games and international events. And running races are considered to be extremely exciting to watch, right? So, let me ask you a question. When should an athlete start running 100 metres rates right? from the gunshot or immediately after the gunshot, right? If an athlete immediately starts running after the gunshot, his race is considered a false start, and he will be disqualified from the events. When should an athlete start running? In that case, an athlete must wait 0.1 second after the gunshot to begin running.

There is a sensor attached to the block on which you stand. You put your feet up to the gun. There's a sensor attached, and they're going to measure the 0.1 second It has been scientifically proven that it takes 0 seconds for sound to reach your ear. Few people tend to start before that, which is wrong. Even one fall start will disqualify you from the event, right? Months and years of hard work will be for naught. So this is the start time, basically in operational definition, and this is the start point. When the task is already considered to be finished, there would be a finish line. I know that I need to touch the finish line with my leg, with my hand, or with any part of my body. When is my wrist considered to be .

When the taskIt's not legs, it's not hands, it's not any part of your body; it's the upper part of your body excluding your hands and head, which is called the astar sole, that has to touch the finish line for an athlete to consider that his race is complete, right? So look at these definitions. They are clearly defined so that each and every judge measures a running race without any bias. According to the International Association of Athletic Federation, these are the rules. We just set extremely important operational definitions. Look at this. If I'm weighing someone and trying to take weight readings, suppose I say, "Hey, this person weighs 150." And suppose I said, "This person also weighs 150," but accidentally measured this person's weight in kg. I forgot this. And I measured this person's weight in pounds. What happens as the end result? Both of these people are going to compete in one weight category.

Because I was such a fool, I forgot to put my units of measurement. So operational definition and units of measure are both extremely important, mind you. Alright, here's a quick exercise to tickle your brain. Here is the customer's voice. Food is too hot. If the temperature is more than 30 degrees, it will be hot. And if the temperature is less than 18 degrees, it would be too cool. Why can't we keep it at 25? The wise customer has come to your Visa centre and has probably discussed the temperature of AC or pizza; you don't know if it is weight (probably AC) or food; it doesn't matter in this case study that we are doing right. What the CPQ criticality customer is speaking about is temperature. All right. Is the temperature continuous on the screen?

It is continuous. Why is it continuous? Because I can measure temperature to whatever decimal level of accuracy I want. It is still going to make sense for me, right? Hence, we refer to that as "continuously." The operational definition is how you're going to measure the ontinuous? Because Are you going to use equipment to take the measurements? If you're using equipment, is the equipment faulty or good? Do you have multiple people who are taking the readings? If yes, then is everyone taking the readings in a similar fashion or not? That is your operational definition. What is the specification limit? 18 degrees. 18 degrees Celsius. Let's clearly define that. What is my outpost specification limit? It is 30 degrees Celsius. And what is my target? I want to maintain the temperature at 25 degrees Celsius. So this is how you need to come up with your operational definition and CTQ measures and things like that. This is the first step in calculating your measurement fees.

3. Measurement System Analysis – AAA

Then comes your measurement system analysis. Yeah. This is extremely important, my friend. Measurement system analysis is extremely important. MSD is used to determine the reliability of a current measurement system that is used to collect output-related data that is used to measure your Why? Essentially, MSA assists in determining whether data can be collected for further analysis, or if your measurement system is faulty, you would say no, let us not proceed with the data collection because the measurement system is faulty. Remember one thing. If there is variation in the data that you are measuring, this variation should not be because of the measurement system. All right, let us look into this. What is a measurement system? A measurement system is a collection of instruments. It's a collection of gauges. It is a collection of standards, operations, methods, software, environment, fixtures, personal, and all that. And remember this. An ideal measurement system will have zero variance. There will be no deviation. Right? There would not be any variance in the measurement outputs. There would be zero bias. And there is zero probability of misclassifying any product it ssifying any prDo you see these kinds of things in reality? I understand that this is an ideal measurement system. But do you see these kinds of things in you see these And if you're wondering whether these kinds of space projects have robust measurement systems, then let me give you this example. NASA lost $125 million on the Mars Orbiter because one engineering team used English units of measurement while another agency's team used metric. Right? And then this unit's mismatch prevented the navigation information from transferring from the Mass Climate Orbital spacecraft to our engineering team and the flight team, because of which the spacecraft has gone for a ht team, bPeople were not able to trace it back. Why a simple measurement mismatch? I measure in a different way. You measure in a different way. And we both did not coordinate. Because of the simple measurement problem, 150 million, or $125 million, was at stake. Hence, we need to ensure that our measurement system is robust. Yeah. First, you analyse your measurement system. Understand that everything is intact, and only then work with the data collection process. Let us first define measurement system analysis. Variations in the measurement process should not be due to the measurement m analysis. VCan we trust our data or not? The evaluation of the measurement system should be completed prior to any capability analysis. We will learn about this capability analysis using Cpcpk later on in the topic. So, measurement system evaluation should be performed before your capability analysis, before control chanting, and before any other analysis to prove that the measurement system is accurate and precise and to prove that the data is trustworthy. But how do I do this? And before that, what is an attribute measurement system? Is it correct that I attributed data? Which has a limited number of categories and a go no go g, correct? Think about this: Your girlfriend makes a statement: "Either you marry me or you go to jail." There are only two options. There are a finite number of options, right? Either you marry me or you go to jail. Either way, you have to undergo the punishment, right? If you get married, you'll be punished. You go to jail; you'll be punished. Which is the best punishment? Attribute measurement system denotes that the measurement values correspond to one or two categories. Or if you're looking for feedback, it might have five to seven categories. satisfied, very satisfied neutral, dissatisfied, very dissatisfied, and five other categories five classifications. Is it possible to divide satisfied? Further. Does it make sense? Can I divide? Very satisfied. Does it make sense? doesn't make sense to me. Your gender? I'm a male or a female, right? Can I divide and subdivide it further? though there are other variants as well. You can't really divide a male and a female, can you? But you can do that on a continuous scale. Weight can be divided with any accuracy possible. Make sense? And that is your attribute measurement system. With an attribute measurement system, what do you use? Which technique do you use to actually gauge the reliability of the measurement system? We use attribute agreement analysis to gauge the performance of the measurement system. Attribute measurement system: what if my measurement system or the data I'm measuring is continuous? If it is continuous, we use something called "gauge RnR," which is discussed in 60 compilers. Here we'll limit ourselves to only discrete data or attribute data, right? And here's an example of that, albeit not consistently. Does your attribute measurement system distinguish between good and bad mobile phone camera lenses? This is attributable to bad luck, right? Hence, you do an attribute agreement analysis in order to measure the reliability of your measurement system. All right, but what is the procedure to perform attribute agreement analysis? Here are the seven steps that you have to perform: First, try to collect 15 to 30 samples of the items you want to measure. Create a master standard, a supervisory reading, or a leader reading that would always, almost always, be right. create that standard. Ask two to three inspectors to review the samples in random order and record their assessment. Next time, randomise the sample order, do not disclose this to inspectors, and have them take the readings again, right? So they'll be taking two readings. Once they take the readings, next time you randomise the sample order and then give it to them once again for reading, and they once again take the measurements.

Calculate the percentage of items where the first trial and the second trial agree with each other. This is called repeatability. Calculate the percentage of items where the first and second measurements of all the two or three inspectors agree with each other. That is called reproducibility. Now calculate the percentage of items from both the first and second trials of these two or three inspectors and measure it against your standard. Which are your supervisor and team leader reading? Because your team leader's supervisor is regarded as an expert, his readings are likely to be accurate. And this is called his accuracy. If these seven points sound alien to you, then now is the time to do an exercise on them.

4. Case study on Attribute Agreement Analysis_Part 1

So we have 20 custom application forms, and this satisfies the first point in our attribute agreement analysis proceedings. Collect 15 to 30 samples of the items you want to measure. We have collected 20 samples. The second step was to develop a master standard that would assign each sample to one of the two categories. I have my team leader. I have given him these 20 application forms, and I've collected the information that he has given. And he comes up to be more or less correct, which is always my standard. I'm going to compare each and every quality representative against the team leader's performance and against the team leader's recordings.

All right, step three says to ask two to three inspectors to review the samples in random order and record their assessment. So, yes, I have three quality representatives or inspectors. I have one QR, one A, two representatives of second quality, and three representatives of third quality. Have three appraisers, three inspectors, or three quality representatives in this case. And I'm asking them to take the readings. So they have taken the readings as part of trial one. As part of trial one, quality representative two has also taken no readings. Quality representative three has also taken readings as part of trial one. Right, now I asked them to take another reading.

Now I need to look into steps 5, 6, and 7, which we have discussed in the attribute agreement analysis procedure. Let us do this. Let us go to Start Quality Tools, and then let us click on attribute agreement analysis. These things are already populated because I have already performed an exercise on them. Okay, this is how it would look initially. All right, since I have data in multiple columns, I select this option. There are several columns, which are listed below. Enter the trials for each appraisal together. That characteristic is representative. Who carried out trial one, trial one? I selected everything and clicked on set. Fill out the form below. How many acquaintances do I have? Three. I have three quality representatives. How many trials did each person perform? Two trials. Do I know the standard or attribute? Oh, yes. and he's my team leader. Select him.

Simply click on "okay." Do not look at the diagram for now. This is known as the "session folder," and it will display what has been done in this session. You have performed attribute-agreement analysis within appraisers. Each appraiser was a standard. There was a standard between appraises and all appraises. Let us discuss each thing in detail. So, first and foremost, within appraisers, let's get to this. Here's one for the appraiser. Yeah. Is quality representative one or quality representative two?

I also have three excellent representatives. There are 20 custom application forms, which each person has inspected. All right, the number of matches is 16 for appraiser one. And the figure is 18%. How did we get to 80, and what is the 16? Let me go back to the worksheet and explain this. In order to go to the worksheet, you need to click on this "Show Worksheets" folder. Look into this policy representative who has taken 20 readings for the first custom application form. He claims that the customer application form is incomplete in the first trial but complete in the second.

Also, he says the customer application form is incomplete in both cases, which is fine for the third form; he says it is incomplete in both cases for the fourth, fifth, sixth, seventh, and eighth forms. Farmers' customer application is inept for all of these reasons. In both trials, he interestingly agrees with himself. Look at this application number nine in his first trial. It says the customer application form is complete. In the second trial, it says it is incomplete. So he is not agreeing with himself. There is one deviation here. For the 10th and 11th, he says that the custom application form is complete in both cases. In both the trials, Look at this application form twelve. For the first time, he's saying it is incomplete. And in the second trial, he changes his decision to be complete. He's not agreeing with himself. The second deviation is here.

The third deviation comes here. Formalized application form 14 In the first trial, he says the custom application form is complete. In the second trial, he says it is incomplete. Key first: look at customer application form 70. He does not agree with himself. He first says the customer application form is complete, and the second time he says it is incomplete. The quality representative disagrees with himself four times out of twenty times. So from 20, if I take out four, the remaining number is 16, right? And that is a 16 that is appearing here. So it matched itself 16 times. So the quality representative agreed with himself on 16 occasions. with respect to 16 custom application forms. If I simply divide 16 by 20 and multiply by 100, I get 80%. At this time, I will not explain the 95% confidence interval or discuss flake up or statistics. Statistics or flake up are discussed in detail in black, not here. All right, what does this 80% mean? Remember this: if any value is greater than 90%, you can say that the measurement system is acceptable.

If the value is between 70 and 90, you have to cautiously accept the measurement system. And if it is less than 70%, the measurement system is unacceptable. So it's 80%. You probably ask your appraiser, who is a quality representative one. But you know what? I cautiously agree with your measurement readings, but you will have to improve your readings, and probably you'll give him some training and ensure that he's up to speed. Quality representative two has a percentage of 90%, which is 18 divided by 20, and quality representative three has a percentage of 90%. Commendable right? greater than 90%. Accent in appraisers is also referred to as "repeatable." Appraisers are repeating themselves if they are matching. If they trial one and trial two, the results match. It is called repeatability. They are repeating themselves right now. Let us look into the difference between appraisals.

Six Sigma LSSGB practice test questions and answers, training course, study guide are uploaded in ETE Files format by real users. Study and Pass LSSGB Lean Six Sigma Green Belt certification exam dumps & practice test questions and answers are to help students.