Analysis Part 1

Analysis
Total Quality Methodologies in Engineering

INSE 6210
Zachary Patterson, Professor

Concordia Institute for Information Systems Engineering (CIISE)
4 October 2023
Announcement
• No class next week for Thanksgiving break
2
Informal Course
Evaluation
Last Class
• Wrapped-up Define
• Project Definition
• Process Measurement
• Identifying and selecting metrics
• Data collection
• Statistical sampling
• Descriptive statistics
• Measurement system evaluation
4
Today’s Class
• Wrap-up measurement
• Process Capability Measurement
• Project analysis
• Basic Probability
• Probability Distributions
• Statistical Inference
• Until Confidence Intervals
5
Overview and Principles of Six Sigma
Six Sigma Methodology
Define Measure Analyze Improve Control
How will you

What problem are Why is the What do you
What is the extent ensure that the
you trying to problem propose to do and
of the problem? problem stays
solve? occurring? why?
fixed?
1. Understand
4. Identify what to 7. Identify performance 9. Generate possible 11. Institutionalize &
customer & business
measure gaps solutions monitor solution(s)
requirements
2. Complete high-level, 10. Prioritize & select 12. Replicate & share
5. Plan and collect data 8. Validate root causes
as-is process map solution(s) best practices
3. Complete project 6. Determine baseline 13. Celebrate &
charter performance recognize success
Additional Topics
Drawn from the “The DMAIC Roadmap” Advanced Innovation Group Pro Excellence 2018
Measure What is the extent of the problem?
• What does the detailed process currently

Step 4
look like?
Identify what to • Functional process map
• What are the output (Y) and primary
measure
measures (Xs)?
• Data collection plan
Step 5 • How can we ensure data collection is robust? • Operational definitions of metrics
Plan and collect data • What does the data say? • Sampling
• Measurement system analysis
Step 6 • Is the process stable?
• Run charts
Determine Baseline • Is the process capable of meeting
• Process capability
Performance requirements?
Measure What is the extent of the problem?
• What does the detailed process currently

Step 4
look like?
Identify what to • Functional process map
• What are the output (Y) and primary
measure
measures (Xs)?
• Data collection plan
Step 5 • How can we ensure data collection is robust? • Operational definitions of metrics
Plan and collect data • What does the data say? • Sampling
• Measurement system analysis
Step 6
• Is the process stable?
Determine • Run charts
• Is the process capable of meeting
Baseline • Process capability
requirements?
Performance
Tollgate success: The team understands the current process well and has collected robust data to support detailed analysis.
Today!
Measure
Process Capability Measurement
Process Capability
• Process capability:
• Ability of a process to produce output that
conforms to specifications
• Process capability study:
• A study to evaluate whether a process is
capable
10
Process Capability
• Main idea:
• Goal for processes:
• Target
• Lower & upper specification limits
• A capable process:
• Process mean centered around target
• Process variation falls within the
specification limits
• Note: it’s a bit more complex, but we’ll start with
this
Capability Definitions
• Specifications:
1. Target:
• Mid-point of acceptable values for a
measure of a process
12
Capability Definitions
• Specifications:
2. Tolerance:
• Range of acceptable values for a measure of
a process, for the process to be considered
capable
• Characterized by
• Lower specification limit
• Upper specification limit
• Allowable variation [± z]
13
Origin of Specifications
1. Specifications are determined by designers of

products and their processes
2. They can come from customers or the market
14
Qs for Capability Study
• Where is the process centered?

• How much variability exists in the process?
• Is the performance relative to specifications
acceptable?
• What proportion of output will be expected to
meet specifications?
15
Capability Study Steps
1. Choose a representative machine or process
2. Define the process conditions
3. Select a representative operator
4. Provide the right materials
5. Specify the gauging or measurement method
6. Collect the measurements and interpret the
data
16
Types of Capability Studies
• Process characterization study:
• How a process performs under actual
operating conditions
• Peak performance study:
• How a process performs under ideal conditions
• Component variability study:
• Relative contribution of different sources of
variation (e.g., process factors, measurement
system)
17
Process Characterization
Study
• Information required:
• Mean
• Standard deviation
• Histogram of sample measurements
• If approximately normal, 99.73 % of
observations should fall within 3 std.
deviations
• Natural variation:
μ ± 3σ
18
Study
• Example 4.12
• U-Bolt Mean dimension: x = 10.7171
• Sample standard deviation s = 0.0868
• Histogram
19
Study
• Example 4.12
• Histogram – approx. normal
20
Study
• Example 4.12
• Histogram – approx. normal
• Thus, nearly all U-bolt dimensions should fall
between 10.7171 - 3(0.0868) = 10.4566 and
10.7171 + 3(0.0868) = 10.9766
21
Process Variation E.g.
22
Ideal
Variation < Tolerance
23
Variation = Tolerance
24
Variation > Tolerance
25
Variation < Tolerance

but
Badly adjusted
26
Bimodal
Different Machines?
27
Capability & Control
• A process is said to be “in control” when:
• Variation in the process results only from
common causes, not special causes
• When variation from common causes alone, it
is in (statistical) control
• Process averages & variances are constant
• When special causes present, process is “out
of control”
28
• Why is this important?

• If not in control:
• Conclusions about capability can be
misleading
• We’ll see why when we look at Control
29
• E.g. 4.13
• 30 samples from a process with specs 0.75 ±
0.25 or LSL = 0.50 and USL = 1.00
• Each row n = 5 every 15 minutes
• Mean of each sample last column
• Freq. histogram in Figure 4.17
30
31
• Appears that the process is at least marginally

capable of meeting specifications.
32
Run Charts
• How tell if special causes are present?
• Run chart:
• Plot mean samples over time, e.g. Fig. 4.18.
33
Run Charts
• Run chart Fig. 4.18
• Mean changed around sample 17

• Not “in control”
34
Control vs. Capability
• Control & capability are different concepts
• Goal: capable & in control
• If neither capable nor in control, must:
1. Get it in control by removing special causes
2. Attack the common causes to improve its
capability
• If capable but not in control, we should work
to get it in control
35
Capability Indexes
• Process capability measured with indexes
• Process capability index:
• Relationship between specifications and
natural variation of process
• USL – Upper limit specification

• LSL – Lower limit specification
• σ – Standard deviation
36
Capability Indexes
• Cp – ratio of the specification range to process
variation
• Cp > 1 – capable of meeting specifications
• Cp < 1 – cannot produce 100% conforming
output
37
Capability Indexes
• One-sided process capability indexes are also

used:
38
Capability Indexes
• Example 4.14
• Based on U-Bolt Data
• These indexes show process not capable of

producing to specifications
39
Capability Indexes
• Example 4.14
• Based on U-Bolt Data
• Can we rely on these results?

• Should check that process is in control first
40
Capability & Attributes Data
• Recall: attributes data - e.g. conforming or not

conforming
• Can use the proportion nonconforming in an
analogous fashion as the mean of a continuous
measurement
• the higher the proportion, the worse the
capability
41
Review: Measurement
• Team agrees that current process is well

described
• Data has been collected to describe the
current performance process
• Data available for Analysis phase
42
Analysis
How will you

What problem are Why is the What do you
What is the extent ensure that the
you trying to problem propose to do and
of the problem? problem stays
solve? occurring? why?
fixed?
1. Understand
4. Identify what to 7. Identify performance 9. Generate possible 11. Institutionalize &
customer & business
measure gaps solutions monitor solution(s)
requirements
2. Complete high-level, 10. Prioritize & select 12. Replicate & share
5. Plan and collect data 8. Validate root causes
as-is process map solution(s) best practices
3. Complete project 6. Determine baseline 13. Celebrate &
charter performance recognize success
Additional Topics
Analyze Why is the problem occurring?
Step 7 • Brainstorming
• What are the potential causes of the
Identify performance • 5 Why
problem?
gaps • Fishbone diagrams
• Confidence intervals
• t-tests
Step 8 • How do you know these are the root causes? • Correlation & regression
Validate root causes • What are the validated root causes? • ANOVA
• Multi-vari studies
• Design of experiments
Tollgate success: Team understands the root causes and impact of the problem and ready to Improve.
Analysis
• Examination of processes, facts, and data to gain
understanding of:
• Why problems occur
• Opportunities for improvement
• Focuses on understanding of:
Y = f(X)
• Especially, most important factors (Xs)
influencing results (Y)
46
Analysis
• Typically includes:
• Analyzing uncovering root causes
• Stratifying and charting data
• Statistical analyses & experiments
47
Root Cause Analysis
Cause and Effect Diagrams
Root Cause
• Ultimate goal of Analyze DMAIC phase:
• understand “root cause” of a problem
• Root cause according to NCR corporation:
“that condition (or interrelated set of conditions)

having allowed or caused a defect to occur, which
once corrected properly, permanently prevents
recurrence of the defect in the same, or
subsequent, product or service generated by the
process.”
49
Root Cause Analysis
• Approach using statistical, quantitative, or
qualitative tools to identify and understand a root
cause
• Two techniques:
1. “Five Why”
2. Cause and Effect Diagrams
50
Root Cause Analysis
Five Why Technique
Five Why Technique
• Approach forcing you to redefine a problem
statement as a chain of causes and effects to
identify the source of the symptoms by asking
why, ideally five times
• Classic Toyota Example
• Machine failed because a fuse blew
• Replacing fuse would address symptom but not
problem
• It is not addressing the root cause
52
Five Why Technique
• Use of 5 Why Technique
1. Why did the fuse blow?
Because the bearing did not have adequate
lubrication
2. Why did the bearing not have adequate
lubrication?
Because the lubrication pump was not working
properly
53
Five Why Technique
• Use of 5 Why Technique
3. Why was the lubrication pump not working?
Because the pump axle was worn
4. Why was the axle worn?
Because sludge seeped into the pump axle
• This was the root cause
54
Cause & Effect Diagram
• Developed by Kaoru Ishikawa
• Simple graphical method for hypothesizing a
chain of causes and effects
• Helps sort out potential causes, and organize
relationships between variables
• Ishikawa Diagram
• Because of structure also fishbone diagram
55
• At end of horizontal line, a problem is listed

• Each branch pointing into the main stem
represents a possible cause
• Branches pointing to causes are contributors to
those causes
56
• Diagram identifies most likely causes of a problem

so further data collection and analysis can be
carried out
• Constructed in brainstorming atmosphere
• Everyone gets involved and contributes to the
problem-solving process
57
E.g. Cause & Effect Diagram
• Major hospital concerned about the length of
time required to get a patient from the
emergency department to an inpatient bed
• Quality improvement team identified four main
causes:
• Environmental services
• Emergency department
• Medical/surgery unit
• Admitting
58
E.g. Cause & Effect Diagram
59
Analysis
Basic Probability
Probability & Statistics
• Many useful tools:
• Descriptive statistics (last class)
• Statistical inference (today)
• Drawing conclusions about a population
based on a sample
• Predictive statistics (today & next class)
• Cause-and-effect relationships
• Forecasting/prediction of future
performance
• Central to Six Sigma Body of Knowledge
• Probability at the base all of these
61
Analysis
Basic Probability Concepts
Basic Terminology
Experiment
• A process that results in some outcome
• Sample of 10 light bulbs to determine how
long it takes them to burn out
Outcome
• Result we observe from an experiment
• Number of hours until the light bulb fails
63
Basic Terminology
Sample Space
• Collection of all possible outcomes of an
experiment
• Total number of parts that can fail
• Hours for possible burnout (0 -> Inf)
Probability
• Likelihood that an outcome occurs
64
Probability Properties
• Label the n outcomes in a sample space as O1,
O2, … On, where Oi represents the ith outcome in
sample space
• The probability associated with any outcome
must be between 0 and 1
• 0 ≤ P(Oi) ≤ 1 for each outcome Oi
• The sum of the probabilities over all possible
outcomes must be 1
• P(O1) + P(O2) + … + P(On) = 1
65
Basic Terminology
Event
• Collection of one or more outcomes from a
sample space
• Finding < 2 defects in a sample of 10
• Bulb burns more than 1,000 hours
Complement
• If A is an event, the complement of A, denoted as
A , consists of all outcomes in the sample space
c
not in A
66
Basic Terminology
Mutual Exclusion
• Two events are mutually exclusive if they can’t
both result from a trial of an experiment
• A is event “two or fewer defects in sample”
• B is event “five or more defects”
• A and B are mutually exclusive
67
Probability Rules
• Rule 1:
• The probability of any event is the sum of the
probabilities of the outcomes that compose
that event
• Prob(1 defect) = 0.2 (Outcome)
• Prob(2 defects) = 0.3 (Outcome)
• Prob(< 3 defects) = 0.5 (Event)
• Rule 2:
• The probability of the complement of any
event A is P(Ac) = 1 – P(A)
68
Probability Rules
• Rule 3:
• If events A and B are mutually exclusive, then
P(A or B) = P(A) + P(B)
• Rule 4:
• If two events A and B are not mutually
exclusive:
• Then P(A or B) = P(A) + P(B) – P(A and B)
69
Probability Rules
Example 5.1
• In sample of 100 units:
• three units have defect in the motherboard
• four units had a hard drive failure
• two units experienced both failures
• A = “failure to boot”, B = “hard drive failure”
• P(A or B)?
• P(A) = 3/100 and P(B) = 4/100
• But not mutually exclusive because, P(A and B)
= 2/100
70
Probability Rules
Example 5.1
• Prob. one or the other failure occurs
P(A or B) = P(A) + P(B) – P(A and B)
= 3/100 + 4/100 – 2/100
= 5/100
71
Basic Terminology
• Conditional probability:
• The probability of occurrence of one event A,
given that another event B is known to be true
or have already occurred
• Multiplication rule of probability:
72
Joint Probability
Ex. 5.2
Tree diagram Figure 5.2
73
Conditional Probability
Example 5.2
P(test indicates defective and product is defective)

= P(test indicates defective | product is defective)
* P(product is defective)
= (0.94)*(0.01)
= 0.0094
74
Conditional Probability
Example 5.2
P(test indicates defective and product is not

defective)
= P(test indicates defective | product is not
defective) * P(product is not defective)
= (0.02)*(0.99)
= 0.0198
75
Joint Probability
Ex. 5.2
Tree diagram Figure 5.2
76
Statistical Independence
• Two events A and B are independent if P(A | B) =
P(A)
• If two events are independent, can simplify the
multiplication rule
• P(A and B) = P(B) P(A) = P(A) P(B)
80
Example 5.3
• Prob. of a non-defective part produced in first
step (event A) is 0.95 and the probability of a
non-defective part produced in the second step
(event B) is 0.98
• What is the probability of producing a non-
defective part?
81
Example 5.3
• Two events independent, so probability of
producing a non-defective part in the process is
P(A | B) = P(A)
P(A and B) = P(A | B) P(B)
P(A and B) = P(A) P(B)
= (0.95)(0.98)
= 0.931
• So, if start with 1,000 parts, only 931 will be non-
defective at the end of the process
• In quality control terminology, often called rolled
throughput yield (RTY)
82
Analyze
Probability Distributions
Random Variables
Random Variable
• Numerical description of the outcome of an
experiment
• RVs can be discrete or continuous
84
Random Variables
Random Variable - Example 1
• Experiment: sampling 10 parts and counting
number of defective parts
• Random Variable X: number of defective parts in
sample
Random Variable - Example 2
• Experiment: testing product for pass or fail
• Random Variable Y: 1 if outcome pass, 0 if
outcome fail
85
Probability distribution
• Characterization of possible values a random
variable can take along with the probability of
taking these values
• Can be discrete or continuous
• For RV X, probability distribution of X is f(x)
• xi represents the ith value of random variable X
86
Cumulative Distribution Function
• F(x), specifies the probability that the random
variable X will assume a value less than or
equal to a specified value, x
• P(X ≤ x)
• The probability that the random variable X is
less than or equal to x
87
Important Distributions
• Discrete
• Binomial
• Poisson
• Continuous
• Normal
• Exponential
88
Discrete Distributions
Binomial Distribution
• Describes probability of obtaining exactly x
“successes” in a sequence of n identical
experiments, called trials
• A success:
• Can be any one of two possible outcomes
(binomial) of each experiment
• In some situations, it might represent a
defective item, in others, a good item
• Probability of success in each trial, p
89
• p - probability of a success
• n - number of trials (items) in the sample
• x - number of successes for which the
probability is desired (0, 1, 2, ..., n)
90
91
• What is the probability that Jeanette

makes exactly 3 sales?
Example 5.4
94
Readings
• Chapter 4 Evans & Lindsay

• Chapter 5 Evans & Lindsay
95

Analysis Part 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Analysis Part 1

Uploaded by

Copyright:

Available Formats

Analysis

Total Quality Methodologies in Engineering

Zachary Patterson, Professor

• No class next week for Thanksgiving break

Six Sigma Methodology

Define Measure Analyze Improve Control

How will you

Six Sigma Methodology

Define Measure Analyze Improve Control

Measure What is the extent of the problem?

• What does the detailed process currently

Six Sigma Methodology

Define Measure Analyze Improve Control

Measure What is the extent of the problem?

• What does the detailed process currently

1. Specifications are determined by designers of

• Where is the process centered?

Variation > Tolerance

Variation < Tolerance

• Why is this important?

• Appears that the process is at least marginally

• Mean changed around sample 17

• USL – Upper limit specification

• One-sided process capability indexes are also

• These indexes show process not capable of

• Can we rely on these results?

• Recall: attributes data - e.g. conforming or not

• Team agrees that current process is well

Six Sigma Methodology

Define Measure Analyze Improve Control

How will you

Six Sigma Methodology

Define Measure Analyze Improve Control

Analyze Why is the problem occurring?

“that condition (or interrelated set of conditions)

• At end of horizontal line, a problem is listed

• Diagram identifies most likely causes of a problem

• Multiplication rule of probability:

Tree diagram Figure 5.2

P(test indicates defective and product is defective)

P(test indicates defective and product is not

Tree diagram Figure 5.2

• P(A and B) = P(B) P(A) = P(A) P(B)

• What is the probability that Jeanette

• Chapter 4 Evans & Lindsay

You might also like