Lecture 5 - MSA & Process Capability

Question 1

What does performance measurement of core business processes involve?

Answer 1

1. Determine which data to capture.
2. Develop a data collection plan for the process & collect data.
3. Establish baseline performance.
4. Compare to customer results to determine the shortfall.

Question 2

Terminologies of DPMO:

What is a Unit (U)?

Answer 2

• Parts
• Components
• Assemblies
• Sub assemblies
• Systems tested or inspected

Question 3

Terminologies of DPMO:

What is CTQ?

Answer 3

The characteristic of a product/process that is Critical To Quality in the eyes of the customer.

Question 4

Terminologies of DPMO:

What is a defect (D)?

Answer 4

Non-conformity in a Unit.

Question 5

Terminologies of DPMO:

What is an Opportunity (O)?

Answer 5

The total number of chances per unit to have a defect. Anything that can be measured, or tested for the presence of a defect is opportunity. There could be multiple opportunities in a unit that could lead to a defect.

Question 6

Calculating DPMO: How do you calculate

Total opportunities (TOP)?

Answer 6

TOP = U x O

Question 7

Calculating DPMO: How do you calculate

Defect per unit (DPU)?

Answer 7

DPU = D / U

Question 8

Calculating DPMO: How do you calculate

Defects per unit opportunity (DPO)?

Answer 8

DPO = DPU / O

Question 9

Calculating DPMO: How do you calculate

Defects per million opportunity (DPMO)?

Answer 9

DPMO = DPO x 10⁶

Question 10

What is the goal of a Six Sigma project?

Answer 10

To improve Sigma quality level and not necessarily to reach Six Sigma performance immediately.

Decide Sigma level that is good enough, then compare your process to competitors - most important is that you are better than them!

Question 11

What does Measurement System Analysis (MSA) involve?

Answer 11

1. Selecting the correct measurement and approach
2. Assessing the measuring device
3. Assessing procedures and people conducting measurement
4. Assessing any measurement interactions (people, measurement device and object measured)
5. Calculating the measurement uncertainty of individual measurement devices and/or measurement systems

Question 12

MSA elements

What is accuracy in terms of a measurement system?

Answer 12

The difference between the measurement and the part’s actual value.

Question 13

MSA elements

What is precision in terms of a measurement system?

Answer 13

The variation you see when you measure the same part repeatedly with the same gauge.

Question 14

Accuracy & Precision related terms

What is true value?

Answer 14

Theoretically correct value

Question 15

Accuracy & Precision related terms

What is bias?

Answer 15

Difference between the average value of all measurements of a sample and the true value for that sample.

Question 16

Accuracy & Precision related terms

What is stability?

Answer 16

Stable processes are those that are free from special cause variation.

Question 17

Accuracy & Precision related terms

What is linearity?

Answer 17

A measure of any change in accuracy or precision over the range of instrument capability.

Question 18

Accuracy & Precision related terms

What is repeatability?

Answer 18

Variability inherent in the measurement system under constant conditions.

or

Variation when one person repeatedly measures the same unit with the same measuring equipment.

One piece - one measuring equipment - one operator

Question 19

Accuracy & Precision related terms

What is reproducibility?

Answer 19

Variability among measurement made under different conditions (e.g. different operators, measuring devices, etc.)

or

Variation between means when two or more people measure the same unit with the same measuring equipment.

One piece - one measuring equipment - various operators

Question 20

Describe accuracy and precision in terms of a target.

Answer 20

Accurate but not precise - On average, the shots are in the centre of the target but there is a lot of variability.

Precise but not accurate - The average is not on the centre, but the variability is small.

Question 21

Name some sources of poor measurement.

Answer 21

• Poor or non-existent operational definitions
• Lack of understanding of the definitions
• Difficult measures
• Poor sampling
• Inaccurate, insufficient or non-calibrated measurement devices

Question 22

What is an operational definition of measurement?

Answer 22

A precise description of how to obtain the measurements you require.

It defines exactly what and how to measure (including how to use the measurement equipment).

Question 23

What are specification limits?

Answer 23

The allowable range of product deviation from the customer’s requirements that he/she is willing to accept.

It is derived from customer requirements.

Question 24

Specification Limits in pub

Who sets the lower specification limit (LSL)?

Answer 24

• Government: “UK law requires certain steps be taken to ensure that a pint of beer is indeed a pint… Until recently glasses had a crown stamp and number etched upon them indicating that the certification had been done by an agency of the Crown.”
• Customer

Question 25

Specification Limits in pub Who sets the upper specification limit (USL)?

Answer 25

- Glass capacity | - Profitability of pub

Question 26

What is process capability?

Answer 26

- Ability of a process to meet its specifications - Measures the performance of a process - Process must be stable in order to calculate process capability (only common cause variation) The two types of process capability measures are: - Cp Index: measures the process potential - Cpk Index: measures the process capability but takes into account the 'off-centredness' of a process

Question 27

Equation for Cp index is:

Answer 27

Cp = (USL-LSL)/6σ = Voice of the customer / Voice of the process or = specification width / process width

Question 28

What does Cp < 1 imply?

Answer 28

Process incapable Inherent or natural variability in the process, as measured by the process spread 6σ, is greater than the difference between the specification limits. ``` Cp < 0.67: Terrible capability rating 0.67 < Cp < 1.00: Poor 1.00 < Cp < 1.33: Fair 1.33 < Cp < 1.67: Good 1.67 < Cp < 2.00: Excellent Cp = 2.00: Terrific ```

Question 29

Process capability yogurt example Factory producing and packaging yogurt in 500g pots. The process mean is 500g and the standard deviation is 2g. The LSL is 495g and the USL is 510g. What is the process capability, Cp?

Answer 29

Cp = (510 - 495) / (6 x 2) = 15/12 = 1.25: Fair capability rating

Question 30

What are the corrective approaches to increase process potential?

Answer 30

Increase the specification limits - make sure that customers are happy with the changes of specification limits! Reduce the process spread - investing in better raw material, new equipment / processes or experienced operators. Side note: Many companies prefer Cp values >= 1.33 as this ensures an extremely low (0.006%) non-conformance rate.

Question 31

What are the limitations of Cp?

Answer 31

- Does not incorporate the process mean, μ, or centres process. - Process can be capable and still not met the customer requirements.

Question 32

Equation for Cpk index is:

Answer 32

Cpk = min { (USL-μ) / 3σ, (μ-LSL) / 3σ } or Cpk = proportion of natural tolerances between process centre and nearest spec. Desirable to have Cpk > 1

Question 33

Process capability yoghurt example Factory producing and packaging yoghurt in 500g pots. The process mean is 500g and the standard deviation is 2g. The LSL is 495g and the USL is 510g. a) What is the process capability, Cpk? b) An equipment provider offers a new high-tech packaging machine with specifications of standard deviation of 0.5 g. Does it solve our problem? c) What would happen if we used the new packaging machine and set it up for a 496.5 gr mean (instead of the current 500 g)?

Answer 33

a) Cpk = min { (510-500) / 3 x 2, (500-495) / 3 x 2 } = min { 1.67, 0.83} = 0.83 b) Cpk = min { (510-500) / 3 x 0.5, (500-495) / 3 x 0.5 } = min { 6.67, 3.33} = 3.33 c) Cpk = min { (510-496.5) / 3 x 0.5, (496.5-495) / 3 x 0.5 } = min { 9, 1} = 1 = (500-496.5) / 500 = 0.7% less yoghurt

Question 34

Introduction to Sampling What is statistical inference?

Answer 34

Using sample data to draw conclusions about a population.

Question 35

Introduction to Sampling What is parameter?

Answer 35

Numeric value that describes population (μ, σ)

Question 36

Introduction to Sampling What is statistic?

Answer 36

Numeric value that describes sample (𝑥, s)

Question 37

What are the reasons for sampling?

Answer 37

1. To save time, money and effort. 2. Produces relatively accurate conclusions about the population from sample (if sample is representative) - Appropriate sampling method - Appropriate sample size 3. Impossible to examine all population - Too large population (e.g. voters) - Sampling is destructive (e.g. testing wine bottles) - ££££

Question 38

What are the four sampling approaches?

Answer 38

- Random sampling - Systematic sampling - Sub group sampling - Stratified sampling