Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

6-Sigma Green Belt > Measure > Flashcards

Measure Flashcards

1
Q

What question is the MEASURE phase answering?

A

How big is the problem?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is Tool 2?

A

Identifies outputs to measure. Converts CTCs from Tool 1 to measurements.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How is information in Tool 2 analyzed?

A

Using a rating scale: strong, medium, weak, null

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the Operational Definition?

A

What is being measured?
How is it being measured?
What is the start/stop for measurements?
What tools?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the 2 types of data that can be collected? Which type is preferred?

A

Continuous & discrete. Continuous is preferred. You can derive discrete from continuous data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is continuous data?

A

Physical measurements - temp, velocity, pressure, location, speed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is discrete data?

A

Counting data - categories, yes/no, pass/fail

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Which statement is true:
Discrete data can be derived from continuous data
Continuous data can be derived from discrete data

A

Discrete data can be derived from continuous data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are ‘Location’ based continuous data types?

A

Mean/Average
Median
Mode
Quartiles

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What are ‘Spread’ based continuous data types?

A

Variation
Standard deviation
Range

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the mean? What is its symbol? How is it calculated?

A

X bar
Sum measurements/Count of measurements

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the median? What is its symbol? How is it calculated?

A

Value at the midpoint of the frequency distribution
X~
Sort the data set in ascending order, divide the set in half, pick the middle value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the mode? What is it’s symbol? How is it calculated?

A

Value with the highest frequency in data set.
X dot

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are Quartiles? What is it’s symbol? How is it calculated? What is its visual representation?

A

Quarters; related to the median (which is 50%)
Symbol: Q
Q1: 25% less than
Q3: 75% less than
Think box plot

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is variation? What is its symbol? How is it calculated?

A

Spread across the data set.
S^2
= Sum (each # from data set - mean) / (count of measurements - 1)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is standard deviation? What is it’s symbol? How is it calculated?

A

68% of the data falls within the standard deviation
S
Square root of variation (s^2)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the range? What is its symbol? How it it calculated?

A

Max/min
R = range
R= Xmax - Xmin

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is proportion? What is it’s symbol? How is it calculated?

A

P^ = # of parts of certain criteria/# total parts = %

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are influencing factors for sampling?

A

Time & Money
Granularity (delta)
Confidence level (z)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What does the Confidence Interval tell you? What is its symbol? How do you calculate it?

A

Tells you how representative your sample is to the population.
CI = x bar +/- z (s/square root n)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is the symbol for confidence levels? What are the 2 options for confidence levels? Which one is most often chosen?

A

Z = confidence level
95% = 1.96 (most common)
99% = 2.57

95% probability that the real population mean is within this interval.

You can also say this as… “with a 95% confidence the mean of the population lies between x-bar - delta and x-bar + delta

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What type of Measurement System Analysis (MSA) is used for discrete data?

A

Gage R n’ R

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is Gage R &R?

A

A type of Measurement System Analysis (MSA) to tell use if the data can be trusted.
Gage R n’ R testes repeatability & reproducibility.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the purpose of a Measurement System Analysis (MSA)?

A

To determine if the data can be trusted. Is there process variation or measurement variation? The goal is to make sure that measurement variation is reduced in order to focus solely on process variation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What are the 2 causes of variation?

A

Common/expected
Special/unexpected

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What key figure does a box plot show? What do the boxes represent? What do the whiskers represent? What does the * mean on the chart?

A

Median & Quartiles
Q1 - 25% less than
Q2 - median 50% less than
Q3 - 75% less than
Min/max
Outlier - distanced too far from the rest of the data%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What does an outlier in the data tell you?

A

It could be a typo on the data, or it could be ‘special cause’ variation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What is a visual representation of continuous data?

A

Box plot
Histogram

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What is a visual representation of discrete data?

A

Pie chart, bar chart, Pareto chart

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

What is the purpose of a control chart?

A

To tell you about the stability of a process by showing development over time. Can be used to see improvement before & after implementation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What are the key parts of a control chart? What do these parts tell you about the process variation? What is the main type of control chart?

A

Shows the mean.
Shows upper control limits (UCL) & lower control limits (LCL).
Always calculated using 3 standard deviation

Data above or below +/- 3s would be considered special cause variation.
Data within +/- 3s would be considered normal cause variation.

IIMR chart - there are always 2 charts that go together
-individual
-moving range

32
Q

What is the difference between an individual & moving range control chart?

A

Individual control chart - each data point is plotted over time & the UCL/LCL are calculated 3s
Moving range - measures the spread in the data by taking the difference between each single data point

33
Q

What is the difference between stability and capability of a process?

A

Stability is the voice of the process - calculated & shown through a control chart
Capability is the voice of the customer - can the process meet their demand?

34
Q

What are the main Quality/Process Performance key figures for discrete data?

A

Yield
Rolled throughput yield
Parts per million (PPM)
Defects per million opportunities (DPMO)

35
Q

What are the main Quality/Process Performance key figures for continuous data?

A

Process cabability (Cp)
Process location (Cpk)

36
Q

How do you calculate Yield? Rolled throughput yield? What are the symbols for each?

A

Yield (Y) is the # of good parts/# total parts x 100 to give you a percentage.
Rolled throughput yield (YTD) is the yield from each individual process step (Y1 x Y2 x Y3). Make sure to convert % into decimals before multiplying together.

37
Q

What does DPMO stand for?

A

Defects per million opportunities.

38
Q

How is DPMO calculated? What is the target DPMO value?

A

Remember D/NO!
DPMU = D / (n x o) x 10^6
D = defects
N = # parts/samples
O = failure opportunities

Target is 3.4 DPMO (6-sigma)

39
Q

What does Cp stand for? How is it calculated? What is the goal?

A

Process capability.
Cp = allowed spread (from customer)/actual spread
Goal > 1.33

40
Q

What does Cpk stand for? How is it calculated? What is the goal?

A

Process location.
Cpk = [(USL - x-bar) / 3s]; [(x-bar - LSL) / 3s]
Gives 2 different results; choose lower value to describe the location of the spread
Goal > 1.33

41
Q

What is considered a good result for process capability & process location?

A

> 1.33

42
Q

What items are required for gate closure in the MEASURE phase?

A

Measurements (KPIs)
Operational definition
Sample size & influencing factors
Visualize the data
Do you trust the data (MSA)?
Process performance

43
Q

Which one should be smaller, process variation or measurement variation?

A

Measurement variation. You want to reduce measurement variation as much as possible so that you can see your process.

44
Q

What is the symbol for granularity? What is the granularity rule of thumb?

A

Delta symbol. Measure 0.1 (1-tenth) of what needs to be seen.

45
Q

Granularity is part of which formula? Explain.

A

Part of the CI formula.
CI = x-bar +/- z * (s/square root n)
Delta = z * (s/square root n)
Want CI to be as small as possible to see Mu. Delta is how close you’ll get to the true value.

46
Q

If I increase the # of samples does the confidence level increase or decrease?

A

Increase

47
Q

How many samples are needed for continuous data?

A

30 data points

48
Q

How many samples are needed for discrete data?

A

100 data points and at least 5 points from each category.

49
Q

Once you know how many samples you need, where do you get data from?

A

Preference is an existing data source & existing measurement. Second would be a new measurement from an existing data source. And so forth.

50
Q

What 2 factors make up total variation?

A

Process variation and measurement variation.

51
Q

What are the requirements for a measurement system?

A

Measurement system includes both the instrument & the operators.

  1. Accuracy
  2. Repeatability
  3. Reproducibility
  4. Linearity
  5. Stability
  6. Granularity
52
Q

What is the instrument bias?

A

The difference between the calculated value and the known true value.

53
Q

What are common causes for measurement variation?

A

Man, machine, Mother Nature, method, material

54
Q

What are the 4 ways to tell if you can trust the data?

A
  1. Expert to verify
  2. Plausibility test
  3. MSA Type I - instrument accuracy, accuracy & precision
  4. MSA Type II - Gage R&R, yes for repeatability, reproducibility, and granularity
55
Q

True or False: It is common to use Tool 2.

A

False. It is rare to use Tool 2. Most often, use Tool 0 (common sense). Don’t bother the team with Tool 2 unless you really need it.

56
Q

Let’s talk about the measure phase. Describe the outline of the topics to be covered.

A
  1. What to measure? Tool 2.
  2. How to measure? Operational definition
  3. Types of data? Continuous vs. discrete
  4. Statistical key figures
  5. How do I find the truth? Is my data representative of the population? Confidence interval.
  6. How many samples do I need?
  7. Where do I get the data? Hopefully an existing data source
  8. Can I trust the data? MSA Type 1 & 2
  9. Graphical Representation of the data
  10. How good (or bad) is my process?
  11. Quick wins
  12. Gate review
57
Q

What is the difference between location & spread?

A

Location - did you hit the target?
Spread - how much variation (precision)

58
Q

There is a pain & then a project… How do I find out the truth about the population?

A

You cannot measure the population – all data from everyone & all of time – but instead take a sample of the population which won’t show the entire reality.

Population > Sample > Data > Statistics > Draw Conclusions about mu where mu is not x-bar.

The confidence interval will tell you how well your sample reflects the population in terms of the mean.

59
Q

True or false? More samples increase the confidence level and decreases the confidence interval.

A

True. The more samples, the more confident, and the range of the interval becomes smaller.

60
Q

Where to get your data - what are the 4 possible data sources/measurements? Which option is preferred?

A
  1. Existing data source/Existing measurement - preferred
  2. Existing data source/New measurement
  3. New data source/Existing measurement (e.g. new machine)
  4. New data source/New measurement
61
Q

What is repeatability?

A

Same person can repeat the same result

62
Q

What is reproducibilty?

A

Person B can replicate results for Person A

63
Q

What is Linearity when it comes to your Measurement System?

A

It’s the linearity of the operating range. For example, a heavier person would not have more bias than a lighter person.

64
Q

What is Stability when it comes to your Measurement System?

A

Means that the bias doesn’t change over time.

65
Q

What is Granularity when it comes to your Measurement System?

A

It’s the resolution that you can see.

66
Q

What is MSA Type I?

A

Tells you the accuracy & capability of your instrument.

67
Q

What are the prerequisites for a MSA Type 1?

A

1 reference part (aka reference value or true value)
1 operator (>/ 30 data points)

68
Q

How do you calculate the capability of the gage? What is the goal result?

A

Cg + Ggk
Rule of Thumb: If Cg & Cgk >/ 1.33 = good
For Cgk, if less than 1.33, there is a bias - tells you how far away the mean value is

69
Q

What are the requirements for Gage R&R for discrete data?

A
  1. Operational definition
  2. At least 30 data points
  3. 1 expert
  4. 2 operators
  5. 2 measures per part per operator
70
Q

What are the requirements for Gage R&R for continuous data?

A
  1. Operational definition
  2. 10 parts & samples across the whole range (not random!)
  3. 2 operators
  4. 2 runs
71
Q

Finish this sentence… “If Po is low…”

A

“…Ho must go.”

72
Q

Describe the results for % Contribution.

A

<1% good
<9% check
>9% bad

73
Q

Describe the results for % Study Variation

A

<10% good
<30% check
>30% bad

74
Q

Describe the results for % Tolerance

A

<10% good
<30% check
>30% bad

75
Q

What is ‘ndc’? What is the basic requirement?

A

Distinct Categories (aka resolution/granularity)
The smallest unit within the data represent the resolution of the measurement system.
Basic requirement: ndc >/ 5

76
Q

Describe Overall Equipment Effectiveness

A

OEE = A (available time) * P (performance) * Q (quality).
85% is world class

6-Sigma Green Belt (7 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions