Process Validation Statistics

Bass Masri
Principal Consultant
Statistical Techniques
www.statisticaltechniques.com.au
Introduction

• Minitab Statistical Software was first developed in 1972 by three

professors to teach their students statistics. Minitab is still a leading
provider of software and services for quality improvement.

• Statistical Techniques is a Sydney consulting business specializing in

quality analysis and strategic research. ‘Statistical Techniques’ is an
important section of Quality Management Systems Regulations.
Tutorial Overview

• This tutorial was developed based on the Global Harmonization Task

Force Quality Management Systems Guidance on Process Validation.

• A ‘heat sealing’ process uses equipment to seal a pouch to act as a

sterility barrier for a medical device. This process requires validation.

• Where the results of a process cannot be fully verified by inspection

and test, the process shall be validated with high degree of assurance.
Process Validation

Defined as establishing by objective evidence that a process consistently

produces a result or product meeting its predetermined requirements.

1. evidence is the documented collection of facts, data and analysis

that demonstrate your process is operating at a validated state.
2. consistently produces links the statistical techniques of process
stability (statistical process control) and process capability (Cpk) .
3. predetermined specifications are the specifications that, if met
result in the product meeting customers needs and intended use.
Benefits and Why

• Process validation results in a process that consistently produces a

high quality product for your customers .
• Think about this in terms of your customer’s satisfaction, customer
experience and your brands reputation.
• Process validation and control results in benefits for your organization
(cost of quality) as well as your customers.
• Scrap reduction, reduced rework, increased productivity, reduced
customer complaints and reduced product liability.
Statistical Methods

• Three process variables: time, temperature and pressure are used to

illustrate statistical methods in manufacturing process validation.
1. Histogram Distributions
2. Statistical Process Control
3. Process Capability
4. Design of Experiments
5. Measurement Systems Analysis
6. Sampling Plans
Method 1 – Statistical Distribution

Characterise the ‘distribution’ of process parameters

closure time, temperature and pressure using a
Histogram - Normal overlay.
Closure TimeHistogram of Closure Time Normal

6 Mean 1.533
StDev 0.09497
N 20
5

4
Frequency

0
1.4 1.5 1.6 1.7
Closure Time
www.statisticaltechniques.com.au
Temperature
Histogram of Temperature Degrees
Normal
Mean 157.9
9
StDev 6.711
N 20
8

6
Frequency

0
140 145 150 155 160 165 170
Temperature Degrees
www.statisticaltechniques.com.au
Pressure Histogram of Pressure kPa
Normal
Mean 324.9
6
StDev 3.307
N 20
5

4
Frequency

0
318 320 322 324 326 328 330 332
Pressure kPa
www.statisticaltechniques.com.au
Method 2 – Statistical Process Control

Determine whether the parameters closure time,

temperature, and pressure are in ‘statistical control’
using control charts (with tests for special cause 1, 2, 3)
Closure Time I Chart of Closure Time
1.8 UCL=1.8055

Individual Value 1.7

1.6
_
X=1.5325
1.5

1.4

1.3
LCL=1.2595

1.2
1 3 5 7 9 11 13 15 17 19
Observation
www.statisticaltechniques.com.au
Temperature
I Chart of Temperature Degrees
170
3
UCL=168.40

165

160
Individual Value

_
X=157.9

155

150

LCL=147.40
145
1

140 1
1 3 5 7 9 11 13 15 17 19
Observation
www.statisticaltechniques.com.au
Pressure I Chart of Pressure kPa
335
UCL=333.72

330
Individual Value

_
325 X=324.9

320

LCL=316.08
315
1 3 5 7 9 11 13 15 17 19
Observation
www.statisticaltechniques.com.au
Method 3 – Process Capability

Determine whether the process parameters are in

‘statistical control’, whether the ‘normal distribution’
models the data well and whether these process
parameters are ‘capable of consistently’ meeting
specifications.
 Closure Time Process Capability Sixpack Report for Closure Time
I Chart
UCL=1.8055 LSL
Capability Histogram
Target USL
1.8
Overall

Individual Value
Within
1.6 _ Specifications
X=1.5325 LSL 1.0
Target 1.5
1.4 USL 2.0

LCL=1.2595
1 3 5 7 9 11 13 15 17 19 1.05 1.20 1.35 1.50 1.65 1.80 1.95

Moving Range Chart Normal Prob Plot

UCL=0.3353 AD: 0.439, P: 0.264
0.30
Moving Range

0.15
__
MR=0.1026

0.00 LCL=0
1 3 5 7 9 11 13 15 17 19 1.2 1.4 1.6 1.8

Last 20 Observations Capability Plot

1.7 Within Overall Overall
StDev 0.09099 StDev 0.09497
Cp 1.83 Pp 1.75
Values

Cpk 1.71 Ppk 1.64

1.5 PPM 0.14
Within Cpm 1.66
PPM 0.44

1.3 Specs
5 10 15 20
Observation

www.statisticaltechniques.com.au
 Process Capability Sixpack Report for Temperature Degrees
Temperature 168
I Chart
UCL=168.40
Capability Histogram
LSL Target USL
Overall

Individual Value
Within
_
X=157.9 Specifications
156
LSL 150
Target 160
LCL=147.40 USL 170
144 1
1
1 3 5 7 9 11 13 15 17 19 140 145 150 155 160 165 170

Moving Range Chart Normal Prob Plot

UCL=12.90
AD: 0.720, P: 0.051

10
Moving Range

5 __
MR=3.95

0 LCL=0
1 3 5 7 9 11 13 15 17 19 140 160 180

Last 20 Observations Capability Plot

Within Overall Overall
StDev 3.499 StDev 6.711
160 Cp 0.95 Pp 0.50
Values

Cpk 0.75 Ppk 0.39

Within
PPM 12260.71 Cpm 0.47
150 PPM 155275.95

140 Specs
5 10 15 20
Observation

www.statisticaltechniques.com.au
 Pressure 336
Process Capability Sixpack Report for Pressure kPa
I Chart
LSL
Capability Histogram
Target USL
UCL=333.72 Overall

Individual Value
Within
328 Specifications
_
X=324.9 LSL 300
Target 325
USL 350
320

LCL=316.08
1 3 5 7 9 11 13 15 17 19 304 312 320 328 336 344

Moving Range Chart Normal Prob Plot

UCL=10.83 AD: 0.356, P: 0.422
10
Moving Range

5
__
MR=3.32

0 LCL=0
1 3 5 7 9 11 13 15 17 19 320 328 336

Last 20 Observations Capability Plot

330 Within Overall Overall
StDev 2.940 StDev 3.307
Cp 2.83 Pp 2.52
325
Values

Cpk 2.82 Ppk 2.51

PPM 0.00
Within Cpm 2.52
PPM 0.00
320
Specs
5 10 15 20
Observation

www.statisticaltechniques.com.au
Process Capability Study in Details
Determine whether process parameters are ‘capable of
consistently’ meeting specifications
𝑈𝑈𝑈𝑈𝑈𝑈 − 𝑋𝑋�
𝑃𝑃𝑃𝑃𝑃𝑃 =
3(𝜎𝜎)
𝑋𝑋� − 𝐿𝐿𝐿𝐿𝐿𝐿
𝑃𝑃𝑃𝑃𝑃𝑃 =
𝑈𝑈𝑈𝑈𝑈𝑈 − 𝑋𝑋� 3(𝜎𝜎)
𝐶𝐶𝐶𝐶𝐶𝐶 =
3(𝜎𝜎�) 𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 𝑜𝑜𝑜𝑜 𝑃𝑃𝑃𝑃𝑃𝑃 𝑎𝑎𝑎𝑎𝑎𝑎 𝑃𝑃𝑃𝑃𝑃𝑃
𝑋𝑋� − 𝐿𝐿𝐿𝐿𝐿𝐿
𝐶𝐶𝐶𝐶𝐶𝐶 =
3(𝜎𝜎�) ∑𝑛𝑛𝑖𝑖 (𝑋𝑋 𝑖𝑖 −𝑋𝑋� )2
𝜎𝜎 = �
𝐶𝐶𝐶𝐶𝐶𝐶 = 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 𝑜𝑜𝑜𝑜 𝐶𝐶𝐶𝐶𝐶𝐶 𝑎𝑎𝑎𝑎𝑎𝑎 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛−1
Closure Time Process Capability Report for Closure Time

LSL Target USL

Process Data Overall
LSL 1 Within
Target 1.5
USL 2 Overall Capability
Sample Mean 1.5325 Pp 1.75
Sample N 20 PPL 1.87
StDev(Overall) 0.0949723 PPU 1.64
StDev(Within) 0.0909854 Ppk 1.64
Cpm 1.66
Potential (Within) Capability
Cp 1.83
CPL 1.95
CPU 1.71
Cpk 1.71

1.05 1.20 1.35 1.50 1.65 1.80 1.95

Performance
Observed Expected Overall Expected Within
PPM < LSL 0.00 0.01 0.00
PPM > USL 0.00 0.43 0.14
PPM Total 0.00 0.44 0.14

www.statisticaltechniques.com.au
Temperature Process Capability Report for Temperature Degrees

LSL Target USL

Process Data Overall
LSL 150 Within
Target 160
USL 170 Overall Capability
Sample Mean 157.9 Pp 0.50
Sample N 20 PPL 0.39
StDev(Overall) 6.71134 PPU 0.60
StDev(Within) 3.49944 Ppk 0.39
Cpm 0.47
Potential (Within) Capability
Cp 0.95
CPL 0.75
CPU 1.15
Cpk 0.75

140 145 150 155 160 165 170

Performance
Observed Expected Overall Expected Within
PPM < LSL 150000.00 119575.42 11988.30
PPM > USL 0.00 35700.53 272.41
PPM Total 150000.00 155275.95 12260.71

www.statisticaltechniques.com.au
Pressure Process Capability Report for Pressure kPa

LSL Target USL

Process Data Overall
LSL 300 Within
Target 325
USL 350 Overall Capability
Sample Mean 324.9 Pp 2.52
Sample N 20 PPL 2.51
StDev(Overall) 3.30709 PPU 2.53
StDev(Within) 2.93953 Ppk 2.51
Cpm 2.52
Potential (Within) Capability
Cp 2.83
CPL 2.82
CPU 2.85
Cpk 2.82

304 312 320 328 336 344

Performance
Observed Expected Overall Expected Within
PPM < LSL 0.00 0.00 0.00
PPM > USL 0.00 0.00 0.00
PPM Total 0.00 0.00 0.00

www.statisticaltechniques.com.au
Method 4 – Design of Experiments

Determine the parameters which affect seal strength

and their optimal values using a two level full factorial
design of experiments.
Select and Create the Design

• Create an experimental design in a worksheet to record the response

variable. Create a 2 level factorial design, with three replicates
View the Design
Analyse the Design
Pareto Chart of the Standardized Effects
(response is Strength, α = 0.05)
Term 2.306
Factor Name
A Size
B Time
A

AB

0 1 2 3 4 5 6 7 8 9
Standardized Effect
www.statisticaltechniques.com.au
Check the Model
Residual Plots for Strength
Normal Probability Plot Versus Fits
99
1

90

Residual
Percent
50 0

10
-1
1
-1.0 -0.5 0.0 0.5 1.0 10 12 14
Residual Fitted Value

Histogram Versus Order

3 1
Frequency

Residual
0

-1
0
-1.0 -0.5 0.0 0.5 1.0 1 2 3 4 5 6 7 8 9 10 11 12
Residual Observation Order

www.statisticaltechniques.com.au
Check for Interactions
Interaction Plot for Strength
Fitted Means
Size * Time Time
15 1.0
2.0
14

Mean of Strength
13

12

11

10

Small Large
Size

www.statisticaltechniques.com.au
Interpret Results
Main Effects Plot for Strength
Fitted Means
Size Time
14.5

14.0

Mean of Strength 13.5

13.0

12.5

12.0

11.5

11.0
Small Large 1 2

www.statisticaltechniques.com.au
Draw Conclusions
Cube Plot (fitted means) for Strength

9.0000 13.8267
2

Time

13.0500 14.4167
1

Small Large
Size
www.statisticaltechniques.com.au
Method 5 – Measurement System Analysis

Determine whether measurements are accurate and

precise and if measurements can be trusted – are
repeatable and reproducible.
Measurement System Analysis

• Five main measurement studies for Continuous Variables

1. Type one Gage Study
2. Gage Run Chart
3. Gage Linearity & Bias Study
4. Gage R&R Study (Crossed)
5. Gage R&R Study (Nested)

• Another study for Attribute Variable

1. Attribute Agreement Analysis
Measurement System Analysis

• Before proceeding, quality engineers need to determine whether the

measurements from the digital scale can be trusted.

• To determine acceptability, the engineers use a type one and a gauge

repeatability and reproducibility study, and an …

• Attribute study to show inspectors consistently agree with their own

appraisals, amongst each other and against a standard.
Type One Study

• Through repeated measurements of a known standard, the type one

study can determine bias and how capable the gauge is of measuring.

• One operator weighs a 100 gram which is a known standard weight

thirty times and records the results into a Minitab worksheet.
Type 1 Gage Study for Weight
Reported by: Bass Masri
Gage name: Type One Study - GMP and Engineering Forum Tolerance: 2
Date of study: 16 July 2018 Misc: Statistical Techniques

Run Chart of Weight

100.2 Ref + 0.10 × Tol

100.1
Weight

100.0 Ref

99.9

99.8 Ref - 0.10 × Tol

1 4 7 10 13 16 19 22 25 28
Observation

Basic Statistics Bias Capability

Reference 100 Bias 0.00 Cg 1.21
Mean 100.00 T 0.033 Cgk 1.20
StDev 0.055 PValue 0.974
6 × StDev (SV) 0.331 (Test Bias = 0) %Var(Repeatability) 16.57%
Tolerance (Tol) 2 %Var(Repeatability and Bias) 16.60%

www.StatisticalTechniques.com.au
Interpretation and Conclusion

• The output shows that for a tolerance of +/- 1 gram the digital scale is
acceptable with Cgk that is greater than 1. All measured values are
within +/- 10% of the tolerance, at an average of 100 grams.

• The null hypothesis is that there is no bias in measurements from the

reference value. The p value (0.974) is greater than 0.05 so we can
conclude there is no bias and proceed to the gauge study.
Repeatability & Reproducibility Study

• Gage R&R studies determine how much of the observed process

variation is due to the measurement system: equipment, operators
environment and procedures and how much is due to part variation.

• Repeatability is variation observed when the same operator

measures the same part repeatedly with the same device.

• Reproducibility is the variation observed when different operators

measure the same parts using the same device.
Gage R&R (ANOVA) Report for Measurements
Reported by: Bass Masri
Gage name: Gage R&R Study - GMP and Engineering Tolerance: +/- 5 Grams
Date of study: 16 July 2018 Misc: Statistical Techniques

Components of Variation Measurements by Parts

% Contribution
% Study Var
100 % Tolerance 25.0
Percent
22.5
50

20.0
0 1 2 3 4 5
Gage R&R Repeat Reprod Part-to-Part
Parts
R Chart by Operators
John Jack Measurements by Operators
UCL=0.6178
Sample Range

0.50
25.0
_
0.25 R=0.24
22.5
0.00 LCL=0
1 2 3 4 5 1 2 3 4 5
20.0
Parts John Jack
Operators
Xbar Chart by Operators
John Jack Parts * Operators Interaction
Sample Mean

Operators
25.0 __ John
UCL=23.816
X=23.57 25.0 Jack

Average
LCL=23.324
22.5
22.5
20.0
1 2 3 4 5 1 2 3 4 5
20.0
Parts 1 2 3 4 5
Parts

www.statisticaltechniques.com.au
Interpretation and Conclusion

• Engineers run a gauge study using two inspectors and five parts
selected from across the range of the part values 20-30 grams.

• Gauge repeatability and reproducibility or measurement system

variation is around 8% of process variation (study variation).

• Most of the study variation comes from the part to part variation.
Therefore they can trust their measurement system and proceed.
Attributes Agreement Analysis

• In the last step of the process, operators visually assess the seal as
closed or open (conforming or non-conforming, binary variable).

• Two appraisers’ rated 50 bags [to reflect the range of parts] as either
good-to-go (G) or not good-to-go (NG).

• Each appraiser [Fred and Lee] rate each bag twice, in a random order
to mimic the typical production environment.
Attributes Agreement Analysis

• Attribute Agreement Analysis shows operators consistently agree with

their own appraisals, amongst each other and against a standard.

1. Within Appraiser: Does each appraiser rate the same parts,

the same way, each time?
2. Between Appraisers: Do the appraisers ratings consistently
agree with each other?
3. Appraiser versus Standard: Does each appraiser evaluate parts
accurately against a standard?
 Assessment Agreement Date of study: 16 July 2018
Reported by: Bass Masri
Name of product: Attribute Agreement Study
Misc: Statistical Techniques

Within Appraisers Appraiser vs Standard

100 95.0% CI 100 95.0% CI
Percent Percent

95 95
Percent

Percent
90 90

85 85

80 80
Fred Lee Fred Lee
Appraiser Appraiser

www.statisticaltechniques.com.au
Interpretation and Conclusion

• The Assessment Agreement graph shows the consistency of each

appraiser's answers and the agreement against a standard.

• Fred rated all 50 parts consistently and correctly across both trials.
From the session output Lee incorrectly rated 3 of 50 parts.

• Lee rated parts incorrectly across the trials and may need training
against approved procedures to assess product more effectively.
Method 6 – Sampling Plans

The Quality team want to inspect for the correct weight

of bags and to ensure the seal on the bag is closed.
Sampling plans

• Measurement studies were successfully completed so we can trust

the measurements of weight and inspector results, we now need to
determine the sample size required to randomly select from the lot.

• It is not always possible, efficient or cost effective to inspect the

entire lot. To determine an appropriate sample size the team use
international standard sampling procedures…
1. Variables Method 2. Attribute Method
ISO 3951 (AS2490) ISO 2859-1 (AS1199.1)
Zero Acceptance Plans
Variables Method

• Variables Method: Weight is a variable that is measured so the

continuous variables method is used. This method has the advantage
of requiring a much smaller sample size than the attributes method.
But the variables method does require some very basic computation.

• Sample Size Question: Quality Assurance engineers want to evaluate

the weight of bags without having to inspect the entire population.
What is an appropriate sample size for a lot size of 1000 bags?
 Variables Method
From the supply agreement the maximum bag weight should not exceed 1 kg, in lots of 1000
items and inspected using a Normal Inspection with an AQL of 1%. Using this information we
need a random sample size of n=35 to demonstrate an AQL of 1% for a lot size of 1000.
Variables Method
• Randomly select a sample n=35 from the population and compute
the sample mean, sample standard deviation and quality statistic.
Sample size n= 35
Sample mean x= sum of x’s /n = 0.66
Sample Standard deviation s= sqrt [sum(xi-x)2/(n-1)] = 0.10
Specification (upper limit) U= 1
Quality Statistic Qu= (U-x)/s = (1-0.66)/0.10 = 3.34
Acceptability constant K= 1.89
Acceptability Criterion 3.34>1.89
• Since the quality statistic is greater than the acceptability constant K
the entire lot can be accepted. The team can be (95%) confident that
the weight of at least 99% of the bags will be conforming.
Acceptance Sampling by Variables

Accept or Reject Decision Using Weight (Kg)

Sample Size 35
Mean 0.663429
Standard Deviation 0.100585
Upper Specification Limit (USL) 1
Z.USL 3.34614
Critical Distance (k Value) 1.89

Decision: Accept lot.

Attributes Method
• The attributes inspection method consists of examining items and
classifying them as either conforming or nonconforming.

• To compute the sample size for an attribute variable the team use the
sampling procedure for inspection by Attributes. The advantage of
attribute over variable sampling is that there is less time taken to
inspect and compute statistics but at a cost of an increased sample.

• A sampling plan requires the acceptance and rejection criteria to be

specified in advance. The acceptance and rejection criteria may be
detailed in a supplier contract, quality document or test protocol.
Attributes Method

• Seal strength is assessed by visual inspection of the seal. If the seal

appears closed, then the seal is considered sufficient to ensure
integrity. If a seal is open then the bag is considered nonconforming.

• Sample Size Question: The Quality Assurance team want to evaluate

the seals in a lot of bags without having to inspect the entire
population. What is an appropriate sample size for a lot of 1000 bags?
 Attributes Method
Using the same quality specifications, from table 2A we see a sample size of n=80 items is
required to demonstrate an acceptable quality level of 1% for a lot size of 1000

80 bags are drawn at random and inspected. No bags were found open, so the lot is accepted.
Attribute Method – Zero Acceptance

• Where no failures are permissible, Zero Acceptance Sampling Plans

(Nicholas Squeglia) may be used. Note that Zero Acceptance Sampling
is specific to the ‘Normal Inspection’ (not tightened or reduced).

• This method has an advantage over AS1199, as its the simplest plan
to interpret as the acceptance number is zero in all cases.

• This plan is suited to low cost items. The disadvantage, is that if one
unit fails the entire lot is rejected.
 Attribute Method – Zero Acceptance
Using the C=0 sampling plan to demonstrate an acceptable quality limit of 1% for lot size of 1000.

The advantage using C=0 plan over AS1199 is the number sampled has reduced from 80 to 34.
Summary

• Three process variables: time, temperature and pressure are used to

illustrate statistical methods in manufacturing process validation.

1. Statistical Distributions,
2. Statistical Process Control,
3. Process Capability,
4. Design of Experiments,
5. Measurement Systems Analysis and
6. Sampling Plans…
Minitab Project Files

• Keen to give it a go?

A copy of the Minitab file is available on request!

• Don’t have Minitab? Download a free 30-day trial from

www.minitab.com
Further Resources
• Introduction to ‘Statistical Quality Control’ by Douglas C.
Montgomery, Fifth Edition.

• Need assistance with quality analysis and strategic research?

Contact Bass Masri [email protected]

• Global Harmonization Task Force SG3 Edition 2 - January 2004 –

Process Validation Guidance.
 PART 820 -- QUALITY SYSTEM REGULATION
Subpart G--Production and Process Controls Sec. 820.75 Process Validation

(a) Where the results of a process cannot be fully verified by subsequent inspection and test, the process shall be validated
with a high degree of assurance and approved according to established procedures. The validation activities and results,
including the date and signature of the individual(s) approving the validation and where appropriate the major equipment
validated, shall be documented.

(b) Each manufacturer shall establish and maintain procedures for monitoring and control of process parameters for validated
processes to ensure that the specified requirements continue to be met.

(1) Each manufacturer shall ensure that validated processes are performed by qualified individual(s).

(2) For validated processes, the monitoring and control methods and data, the date performed, and, where appropriate, the
individual(s) performing the process or the major equipment used shall be documented.

(c) When changes or process deviations occur, the manufacturer shall review and evaluate the process and perform
revalidation where appropriate. These activities shall be documented.
 PART 820 -- QUALITY SYSTEM REGULATION

Subpart O--Statistical Techniques Sec. 820.250 Statistical techniques.

(a) Where appropriate, each manufacturer shall establish and maintain procedures for identifying valid statistical techniques
required for establishing, controlling, and verifying the acceptability of process capability and product characteristics.

(b) Sampling plans, when used, shall be written and based on a valid statistical rationale. Each manufacturer shall establish and
maintain procedures to ensure that sampling methods are adequate for their intended use and to ensure that when changes
occur the sampling plans are reviewed. These activities shall be documented.
 GUIDE TO INSPECTIONS OF QUALITY SYSTEMS

Production and Process Controls

4. If the results of the process reviewed cannot be fully verified, confirm process was validated by reviewing the
validation study.

If the chosen process requires process validation, review the established Process Validation Procedure(s).

Verify via a review of the Process Validation Study Summary and Approval that objective evidence has demonstrated
that the process will consistently generate a product or result meeting its predetermined specifications.
 GUIDE TO INSPECTIONS OF QUALITY SYSTEMS
NOTE: If there are indications of unresolved, potential problems with a validated process, in addition to a review of process
monitoring and control activities, a comprehensive validation study review should be conducted.

1. The instruments used to generate the objective evidence were properly calibrated and maintained prior to the study;

2. Predetermined product specifications were established;

3. Test sample sampling plans were based upon a statistically valid rationale;

4. Objective evidence demonstrates predetermined product specifications were met consistently;

5. Process tolerance limits were challenged;

6. Process equipment was properly installed, adjusted and maintained;

7. Process monitoring instruments are properly calibrated and maintained;

8. Changes to the validated process were appropriately challenged; and,

9. Process operators are appropriately qualified.
Q&A