0:14

So, if you think about Measurement Systems, what you're concerned about is,

Â you're concerned about variation that might be coming from different places.

Â So when you think about a measurement system,

Â you should be thinking about, are people trained to measure it the same way?

Â Are the devices going to be calibrated such that they stay consistent over time?

Â Are the measurement instruments going to be such that they don't get affected

Â by who is using them, or by dropping them, or things like that?

Â And again if you're thinking about perceptual measurements, you're thinking

Â about surveys, are they going to mean the same thing to the same people?

Â Is the wording going to be

Â current when you're talking about using the same battery of questions over time.

Â So you should be thinking about that as well.

Â And then the procedures to actually do the measurement.

Â So how are these going to be actually done?

Â Are they going to be done taking five samples and taking the average and

Â then how exactly are you going to get that measurement to be done.

Â Because you're trying to get away from getting variation from the measurement

Â system itself, or errors from the measurement system itself.

Â What kind of standards to you have for the measurement system?

Â And then what kind of standards do you establish for

Â saying something is beyond a particular threshold or not.

Â And then training people for using those measurement systems.

Â Training people to do this correctly, getting a demonstration.

Â So when you think about a criteria and a test,

Â you should be able to demonstrate this and

Â somebody should be able to understand it and then be able to replicate it for you.

Â They should have the same meeting of that critical to quality characteristic and

Â the criteria and the test that you do so that when anybody does this,

Â they should be able to do it the same way and get the same result.

Â If it's the same packet, they should be able to get the same result.

Â There shouldn't be variation in results from measuring the same thing over time.

Â So what are we concerned about when we are talking about measurements?

Â And some of this can be looked at based on data.

Â When we collect data, we can take a look at the variation in the data and

Â we can parse that variation out into does it look like.

Â Does it look like there's variation that's coming in because of measurement?

Â So what are some of the things that we're concerned about with measurements?

Â With measurements we're concerned about sampling bias.

Â So, did we choose particular samples at a particular time all the time?

Â And one way to get around it is have some sense of random sampling and

Â even including some kind of random sampling with some rules, like saying,

Â we take a random sample from the 8 o'clock batch or

Â we take a random sample for every hour's batch and so on and so forth.

Â So depending on what is it that you're trying to look at.

Â If it's going to be a hypothesis test, then random samples are better.

Â And if it's going to be looking at a process' performance,

Â then you want it to be timed and make sure that you're getting a sample to represent

Â each of those times, or each of those days of the week and so on and so forth.

Â But you have to be aware of the idea that your

Â sampling can bring in some bias into your measurement.

Â And again you could be thinking of perceptual measurements and

Â how did you sample the people that you talked to them,

Â the customers that you talked to or employees that you talked to in

Â order to get some sense of whatever you are measuring.

Â 3:41

What we are concerned about is repeatability by the same appraiser.

Â So, if you give the same person the same object to weigh over time,

Â the weight should be the same, right?

Â If you give it to them today and you give it to them tomorrow and day after,

Â the weight should be the same.

Â That's what we mean by repeatability.

Â Reproducibility is, if you give me that object and if you give my kids that

Â object, if they're trained the same way, they should get the same measurement.

Â So it's multiple appraisers should get the same measurement.

Â That's what we call reproducibility.

Â We're concerned about linearity over range.

Â If you think about looking at any weighing instrument,

Â it usually says this is accurate up to a certain weight.

Â If you go beyond that weight, don't expect it to be accurate, right,

Â that's what they're trying to say.

Â So when you have a measuring scale that is meant for measuring your spices,

Â it's going to be a small measuring scale and if you're trying to take a pound of

Â flour and you're trying to weigh it on that.

Â It may not be calibrated to go beyond half a pound of weight.

Â So if you're trying to measure something like a pound of flour,

Â don't expect to get accuracy.

Â And that's what we mean by linearity over range,

Â that it's not going to be linear over range.

Â It should be strictly linear over range if you're talking about a continuous

Â measurement and if it's not then there's a problem with it.

Â Stability over time is that if you do the measurements over time, they should

Â give you the same kind of result if it's you're talking about the same thing.

Â So, in order to work with measurements, in order to get a sense of

Â the measurements and how good they are, you can do something called a Gage R&R,

Â a reproducibility and repeatability analysis.

Â And this can be based on getting data from multiple respondents for multiple objects.

Â And getting them to do the repeated kind of measurements from that same.

Â We collect that data and you start to look at if

Â there was any variation in the data, and then you look at whether that could be

Â coming from training of people, or whether that could be coming from the instrument,

Â or whether that could be coming from instructions that you're giving people.

Â So gauge R&R is something that you can use based on data to look at these kinds of

Â questions.

Â Now when we turn to perceptual scales, also we can do similar kind of analysis,

Â and here I have an example of an employee satisfaction item.

Â So this is a single item that says my supervisor

Â encourages innovation by tolerating failure.

Â Respond to this on a scale of 1 to 5 from Strongly Disagree to Agree,

Â and a 9 of not applicable.

Â So, here are the options that you have, how would you test this?

Â You would test this based on a test, retest reliability.

Â You are giving multiple administrations of the same scale to respondents.

Â And you give again to them at different times,

Â or you could be testing people who are working at exactly the same conditions,

Â but you are trying to get inter-rater reliability.

Â You're measuring whether two people would be giving the same response about

Â the same thing when you expect it to be the same.

Â So that's how you could be testing this.

Â And again you could do data analysis for this.

Â You can collect the data and

Â do some assessment of the variation that you're seeing in the data

Â to see where that variation is coming from or might be coming from.

Â So in some about the quality of process data, what can we say?

Â We can can be looking at the validity of data, its measuring,

Â what it's supposed to measure, its measuring of something that we expect

Â to be measuring and it's also giving us a comprehensive measurement of something.

Â So it's something that's useful for

Â us to make an assessment about the product or the process.

Â So that's what we mean by validity.

Â It's reliable, it's consistent and accurate, it's sensitive to changes.

Â There should be data,

Â there should be measurement that should be sensitive to changes.

Â It should be calibrated to a level of granularity that it's sensitive to

Â changes.

Â Otherwise, you're not going to be able

Â to make out the difference between two different levels when you are supposed to.

Â And it should be accessible.

Â It should be accessible to the people who are going to do the measurements, and

Â who are going to use it in a timely fashion.

Â So it should be accessible and be able to be used in a timely fashion.

Â So the simpler it is, the more people can use it.

Â And the simpler it is, the more quickly you can get that measurement and

Â the results from that measurement, for it to be used in a timely fashion.

Â Finally in terms of using data for six sigma and how it can be used for six

Â sigma, so we've talked about this earlier with the idea of process control charts.

Â So what are process control charts,

Â they look at the inherent potential of the process.

Â So we can use juridical process control charts to figure out,

Â what is the common cause variation in the process and

Â what is beyond that, and what is the special cause variation?

Â So we can look at the performance of the process based on process control charts.

Â And then once we've established that a process is in statistical control

Â using SPC, using statistical process control charts,

Â we can do process capability analysis or Cp and Cpk ratio calculations.

Â And these are meant to take the voice of the process and

Â compare it with the voice of the customer.

Â How does all this relate back to the idea of 6-sigma?

Â So, once you learn about where, once you learn about calculating these

Â process capability ratios, the Cp and the Cpk values.

Â How you can relate this back to the idea of 6-sigma is that a Cp and

Â Cpk value of 1, exactly 1 indicates that it's a 3-sigma process.

Â What that's saying is that, if you were to look at the distribution of the process,

Â and you go plus or minus three standard distribution by going plus or

Â minus three standard deviations from the mean.

Â You're going three standard deviations from the mean on either side.

Â You reach the upper and

Â the lower limit of the specifications that are given to you by the customer.

Â So plus or minus three standard deviations from the process matches the upper and

Â the lower limits that are given to you by the customer.

Â The spec limits that are given to you,

Â the specification limits that are given to you by the customer.

Â So a 3-sigma level process means a Cp or Cpk value of 1 and

Â a 2 Cpk value indicates a process that's working at a 6-sigma level.

Â So when we say a 6-sigma level performance,

Â we're in fact saying that it should be at a 2 Cp, Cpk level.

Â And remember we're adjusting for that 1.5 sigma, the addition of that 1.5 sigma that

Â we've talked about earlier in different sessions in this particular course.

Â