Transcript fraction defective - Qualfon University

Continuous Improvement
and Six Sigma Green
Belt Certification
Course Wrap-up
What is Continuous Improvement?
is a defined and disciplined business set of
methodologies to increase customer satisfaction and
profitability by streamlining operations, improving
quality and eliminating defects in every organizationwide process in a daily basis.
One of these methodologies is Six Sigma.
The Hidden Factory
The “Hidden Factory” refers to the
activities included in a procedure.
These include Band-Aid treatments,
workarounds and eventual re-work.
The Hidden Factory is the set of activity (or activities)
in the process that results in a reduction of quality or
efficiency of a business process or service
department, and is not known to managers or others
seeking to improve the process. We get used to live
with these activities that add no value to the client and
customer.
History of Six Sigma
Key Concepts:
William A. Shewhart (1920s) – demonstrated that statistics can be
used to determine production Quality and the subsequent impact on
productivity.
W. Edwards Deming (1950s) – applied Shewhart’s theories on
Statistical Process Control to Management techniques (TQM).
Post WWII – Japan adopted Deming’s teachings and experienced
improvement in Quality.
Motorola (1980s) – Quality engineers Dr. Mikel J. Harry and Bill Smith
developed a series of filters to help improve Quality. This will later be
known as Six Sigma. Motorola became the first recipient of the Malcolm
Baldridge National Quality Award.
History of Six Sigma
A History with Periods
1920s
1940s
1950s
1979
1986
1999
W. A. Shewhart
W. E. Deming
Made in Japan
Motorola
Statistics can be used
to determine production
quality and productivity.
Shewhart’s theories can
also be applied to
management
techniques.
Deming gives
management lectures in
Japan on management
and improving quality.
In 1979, after experiencing
aftermarket quality control
issues, 2 engineers create
the basis for Six Sigma.
Six Sigma Improvement Process
Excellence Disciplines
Waste
Reduction
(Lean)
Problem
Solving Steps:
DMAIC
PROCESS
Process Excellence and
Process Documentation
What is Six Sigma Improvement Process?
•
•
Six Sigma was originally developed by Motorola, led by its Chairman Bob Galvin, as a
company-wide activity in the late 70s.
It uses statistical tools to identify and eliminate variation.
•
•
Six Sigma is developed as a “Top Down” methodology in 1987.
Six Sigma is a set of activities to achieve a financial impact (profit and cost reduction)
throughout the entire organization using statistical methods, evidence-based
administration and policy alignment of senior management.
•
•
The fundamental structure and the methodology, matures by mid 90's.
Many success stories have organizations worldwide.
•
Actually, it is considered a one of the methods to solve problems for Continuous
Improvement activities.
Problem Solving Approach vs Design Approach
• Reduction of defects / faults / errors.
– Eliminate errors.
– “Doing things correctly” (level 1).
Problem solving
approach
• Certainty (consistency, predictability).
– Certainty in the processs (process stability)
– “Doing things correctly” (level 2).
•
Problem solving
approach
Task achievement & Innovation.
– Go to the forefront.
– Achieve defined tasks and challenges.
– “Do things correctly vs. Do the correct things”.
Design
approach
(innovation)
Six Sigma vs BPR
Business Process Reengineering
BPR
6 Sigma
General
tendency
Radical redesign
Align & Maintain
Approach
Design & innovate
Problem solving
Identify
the
process
Standardize
Tools
Process maps &
Project management
Statistics
Medium and long term
Short, medium term
Deployment
Top Down
Top Middle Down
Implementation
time
1 a 2 years
6 months
Process change
Process variation
reduction
Impact
Benefits
Implement
actions
towards To
Be state
Analize
current As
Is process
Design
desired To
Be
process
Problem
•Understand the problem that impacts
Business performance
Measure the
problem
•Evidence Based understanding of the
problem
Causal
relationship
•Find the origin of the problem to define a
reliable data driven solution
Practical solution
•Implement the best solution (quality,
cost, delivery, safety)
Control plan
•Create a system to stabilize and sustain a
long term solution
Results
•Measure the benefits of the improvement
(financial, service, sales, growth)
Results
Problem
Define
Control
Improve
VOC
&
Voice of
the
Process
Measure
Analyze
Continuous Improvement (CI) Culture promotes customer satisfaction and value added by
streamlining operations, improving safety, quality and service, reducing costs, eliminating
defects and errors in every organization wide process in a daily basis.
What is Quality?
Different individuals and organizations have given different
definitions for Quality.
“Quality is defined from the customer’s point of
view as anything that enhances their satisfaction.”
Deming
“Fitness for use. Those product features which
meet the needs of customers and thereby provide
product satisfaction. Freedom from deficiencies.”
Juran
“The totality of features and characteristics of a
product or service that bear on its ability to satisfy
stated or implied needs.”
ASQ
“Quality is defined as knowledge of agents that would
enable them to provide accurate and consistent
solution to the customer at the very first attempt.”
“Degree to which a set of inherent characteristics,
of a product or service, fulfil requirements.”
COPC
ISO
What is Quality?
From the Customer’s Point-of-View
UNDERSTANDING customer
requirements
DESIGNING products and
services that satisfy those
requirements.
DEVELOPING processes that
are capable of producing those
products and services.
CONTROLLING AND
MANAGING those processes so
they consistently deliver to their
capabilities.
CONTINUOUSLY IMPROVING
processes to develop a culture
towards constant change
Global Quality Characteristics
•
•
•
•
•
•
Q: Quality: physical & affective, design, attitudes and performance attributes
C: Cost: price, cost, waste, investments and value added attributes
D: Delivery: distribution, service or operation time & quantity attributes
S: Safety & Security: hygiene, information security attributes
M: Morale: work environment, proactiveness, well being attributes
S: Service: customer focus and satisfaction, after service attributes
4 Types of Quality: NORIAKI KANO Analysis
Customer Satisfaction
I am delighted
Quality of
Attractiveness
One dimensional
quality
Not a problem
Delighter
requirements
Service Fulfillment
I am angry
It´s obvious
Mandatory
quality
Performance
requirements
Mandatory
requirements
Reverse
quality
KAIZEN
• “KAIZEN” japanese word meaning improvement or
change towards the correct destiny.
改 = KAI = Change, Alter, Correction.
善 = ZEN = Good, Right, Better.
改善 = KAIZEN = Improvement, Change for the better
Poka Yoke: Mistake Proofing
Mistake proofing o POKA YOKE (from Japanese poka [inadvertent error] y yokeru [prevent]) helps to
reduce human error, specially when it impacts process performance (“Y”).
POKA YOKE reduces and eliminates the potential problem to happen.
Basic functions:
• Shutdown: “Stop”
• Control
• Warning: Detects flaws
3 detection methods:
• Contact (e.g. connectors)
• Fixed value (e.g. “pepto” cup)
• Fixed motion step (e.g. checklist and
actions prior to operation)
KAIZEN 改善 IMPROVEMENT
• To Change for the better
A
P
C
D
P= Plan
D= Do
C= Check
A= Action
5W – 2H
•
•
•
•
•
•
•
What?
Who?
Where?
When?
Why?
How?
How much / many?
BASELINE RULE
• First.
– Process stabilization.
•
Stable = Controlled = Predictable
– Compliance with standards and procedures.
• Meet client, customer needs.
• Meet regulations and laws.
• Meet our corporate standards and procedures.
• Verification of everybody´s skills.
• Second:
– Improvement efforts.
Performance metrics
•
•
•
•
•
•
Percentage defective
PPM: Defective parts per million units
DPU: Defects per unit
DPO: Defects per opportunities
DPMO: Defects per million opportunities
Rolled Throughput Yield
Fraction Defective
•
Fraction Defective: “Good vs Bad”, “OK vs No Good (NG)”
– Fraction Defective is the number of defective units divided by the total
number of units
– Note: a defective unit is any unit containing one or more defects.
– Ex. 50 services (online chat) are randomly selected and inspected. Two
services have typos, one service has typos and misspelling of customer
name (2 defects in same one service), three services give wrong
directions. 6 services were defective (low Customer Satisfaction Index)
6 defective services out
of 50 services
6/50 = 0.12
Fraction Defective
•
Fraction Defective: “Good vs Bad”, “OK vs No Good (NG)”
– Ex. 50 services (online chat) are randomly selected and inspected. Two
services have typos, one service has typos and misspelling of customer
name (2 defects in same one service), three services give wrong
directions.
•
•
•
•
Total of 7 defects in 6 out of 50 services sampled.
Number of defective services = 6
Fraction Defective = 6/50 = 0.12
Interpretation: At this quality level, we can expect this proportion of
defectives in average in our process (e.g. 12 defectives each 100
services on average).
Defectives
Defects
7 defects in 6 defective
services out of 50 services
Percentage Defective, PPM
•
Percentage Defective.
– Percentage Defective is the Fraction Defective multiplied by 100 or the
number of times a defective unit or service will occur in 100 units or services
produced.
•
Defective PPM: Parts per million
– Defective PPM or simply PPM is the Fraction Defective multiplied by
1,000,000 or the number of times a defective unit or service will occur in 1
million units or services produced.
– Ex. 50 services (online chat) are randomly selected and inspected. 6
defective services are detected.
• Fraction Defective = 6/50 = 0.12
• Percentage Defective: At this quality level, 12 defectives per 100 services
• PPM: At this quality level, 120,000 defectives per 1 million services
DPU, DPO, DPMO
•
DPU: Defects per unit
– Number of defects in a sample divided by the
number of units sampled
– Ex. 50 services (online chat) are randomly selected
and inspected. Two services have typos, one
service has typos and misspelling of customer
name (2 defects in same one service), three
services give wrong directions.
•
•
•
•
Number of non conforming or defective services = 6
Total of 7 defects in 6 out of 50 services sampled.
DPU = 7/50 = 0.14
Interpretation: At this quality level, each service
can contain on average this number of defects
(e.g. 14 defects each 100 services on average).
One unit,
service or
operation with 3
defects
DPU, DPO, DPMO
• DPO: Defects per opportunities
– Number of defects in a sample divided by the total number of defect
opportunities or defect types.
– Ex. Each service (online chat) could have 4 defects: Typo, incorrect
name, wrong direction and incomplete information. Therefore each
service has 4 defect opportunities. 50 services are randomly
selected and inspected. Two services have typos, one service has
typos and misspelling of customer name (2 defects in same one
service), three services give wrong directions.
• Total of 7 defects out of 200 opportunities (50 services x 4 defect opportunities /
sample)
• DPO = 7/200 = 0.035
• Interpretation: At this quality level, each service can contain on average this
number of DPO (also, we can say 35 defects each 1000 opportunities on
average).
DPU, DPO, DPMO
• DPMO: Defects per million opportunities
– Number of defects in a sample divided by the total number of defect
opportunities multiplied by 1 million. DPMO standardizes the number
of defects at the opportunity level (not per unit) and is useful when
comparing processes with different characteristics.
– Ex. Using the DPO example, 50 services randomly selected and
inspected. 2 services have typos, 1 service has typos and misspelling
of customer name (2 defects in same one service), 3 services give
wrong directions. Each service could have 4 defect opportunities.
•
•
•
•
Total of 7 defects out of 200 opportunities (50 samples x 4 defect opps / sample)
DPO = 7/200 = 0.035
DPMO = 0.035 x 1,000,000 = 35,000
Interpretation: At this quality level, if our process remains the same, each
1,000,000 opportunities will generate 35,000 defects on average
Note
• DPMO (Defects per Million Opportunities) differs from
Defective Parts Per Million (PPM) as it comprehends
the possibility that one defective unit or part under
inspection may have multiple defects (same kind or
different types of defects).
• DPMO is equal to PPM only when the number of
opportunities for a defect per unit or service is 1.
Process Yield
Different types of fulfilment can impact the quality level measured in processes.
Yield can be understood as Classical Yield (YC), First Pass Yield (Yft or FPY),
and Rolled Throughput Yield (Ytp or RTY).
FPYx = Units Passed / Units input for First Time (per process)
FPY1 = 91/100 = 0.90
100
Process
1
Yield
1
FPY2 = 82/91 = 0.90
Process
2
91
Rejects
(9)
Yield
2
FPY3 = 70/82 = 0.85
Process
3
Inspect /
Test
4
Yield
4
Pass
65
70
82
Rejects
(9)
Yield
3
FPY4 = 65/70 = 0.93
Rejects
(12)
Rejects
(5)
YC = Units Passed / Final Units Tested
= 65/70 = 0.93 (last process)
RTY = 𝒏𝒊 𝑭𝑷𝒀i = FPY1*FPY2*FPY3*FPY4 = (91/100)*(82/91)*(70/82)*(65/70) = 0.65
FPY all = Total units passed / Total Units input for First Time = 65/100 = 0.65 (no rework allowed)
Process Yield
•
•
•
YC (Classic Yield) metric only provides the view of the “last process”, commonly a test or final
inspection operation.
FPY (First Pass Yield) metric only provides the view of the sub process yields. Those FPY
running high do not drive for Improvement efforts.
RTY (Rolled Throughput Yield) metric provides a more complete view of the process. It is only
when the total process yield becomes visible, real actions occur.
FPY1 = 91/100 = 0.90
100
Process
1
Yield
1
FPY2 = 82/91 = 0.90
Process
2
91
Process
3
Yield
3
Inspect /
Test
4
FPY4 = 65/70 = 0.93
Yield
4
Pass
65
70
82
Loss 2
(9)
Loss 1
(9)
Yield
2
FPY3 = 70/82 = 0.85
Loss 3
(12)
Rejects /
Loss 5
(5)
YC = Units Passed / Final Units Tested = 65/70 = 0.93 (last process)
RTY =
𝑛
𝑖 𝐹𝑃𝑌 i
= FPY1*FPY2*FPY3*FPY4 = (91/100)*(82/91)*(70/82)*(65/70) = 0.65
Brain Flex:
Calculating Yields
Customers need to go through a 3 step human interaction
process to solve his situation. We have Accuracy Rates for
each of the 3 steps.
Calculate the FPY for each step and RTY of the whole process
in the following diagram.
100
Process
1
Yield
1
Process
2
90
Incorrect
(10)
Yield
2
Process
3
77
81
Incorrect
(9)
Yield
3
Incorrect
(4)
Video: Process distribution & Standard Deviation
And why only 6 sigmas?
One Standard Deviation
99.7%
Six Standard Deviations
Mean
Central tendency
Brain Flex
Standard Deviation and its meaning
•
If a customer response process has a
mean value (average) of 30 seconds
and a standard deviation (sigma) of 5
seconds, what can you say about this
process?
Brain Flex
Standard Deviation
and its meaning
• What are your
conclusions of the
following graph
where the mean
value of the
process is 28
minutes and the
standard deviation
is 0.3 minutes
Process distribution, Standard Deviation & Sigma Level
SIGMA level
measures the
distance to a
specification
or standard
in terms of
sigmas
5 Standard Deviations AWAY
from the specification
SIGMA LEVEL = 5 Sigma
One Standard Deviation
Six
Standard
Deviations
99.7%
Mean
Central tendency
Specification or
Standard to comply
SIGMA LEVELS at GLANCE
One Standard Deviation
Six
Standard
Deviations
99.7%
Specification
or Standard to
comply
Specification
or Standard to
comply
One Standard Deviation
Mean
Central tendency
3 Standard Deviations AWAY
from the specification
SIGMA LEVEL = 3 Sigma
Six
Standard
Deviations
99.7%
Mean
Central tendency
6 Standard Deviations AWAY
from the specification
SIGMA LEVEL = 6 Sigma
Costs of bad quality
Quality Costs (Bad Quality)
> 50% of sales (Out of the market)
30-40% of sales (Non competitive)
20-30% of sales
15-20% of sales (Average organization)
10-15% of sales
3-10% of sales (World class leader)
DPMO*
**
697.672
1
308.537
2
66.807
3
6.210
4
233
5
3.4
6
* DPMO: Defects per Million of Opportunities
** Average Sigma Level in Most Relevant processes of the organization
35
How CI & Six Sigma Work
Business Problem
Cause Analysis
Creating a problem
statement and
identifying problem
magnitude & scope.
Testing and Root
Cause Analysis
(RCA).
Business
Problem
Statistical Problem &
Causes
Analysis of the
Solutions
Identifying key
solutions to root
causes through
statistical analysis.
Causes & Statistical
Solutions
Business Solution
Implementing
solutions and
standardization for
sustainability
Business Solution
& Control
The Focus of CI and Six Sigma
Six Sigma mathematics
Understanding the factors
o Y is the outcome or result desired
o X is/are the factors that affect the result Y
o Ɛ is the error of the function
o f is the change (transformation) fuction to be applied
FOCUS ON THE CAUSE (X),
not the Result (Y).
Descriptive Statistics
Simply Describing the Data
Objective
Measures of
Central Tendency
Mean
Median
Mode
Measures of
Dispersion / Variation
Quartiles
Range
Variance
Standard
Deviation
Additivity of mean and variance
P1
Average
time 1
µ1=20
Whole
Process
(minutes)
P2
Average
time 2
µ2=30
Total average
time
µ=20+30+10=60
P3
Average
time 3
µ3=10
Probability Distribution
Means and Standard Deviations
Probability distribution is simply a
theoretical frequency distribution
characterized by Mean and Standard
Deviation.
σ
Normal Distribution is
represented by µ and σ.
Calculation of z-value (z score)
𝑥−𝜇
𝑧=
𝜎
Where:
x = Value of the data point of interest
µ = Mean of the population or data points
σ = Standard Deviation of the data points
Z = Number of standard deviations between x & the
mean (µ). z-value is negative when the data
point of interest is below the mean.
What is the probability?
Assume the mean of service process A is
28.0 minutes with a standard deviation of 0.3
minutes. Estimate the % of process A
services finishing:
a)
Before 27.4 minutes
b)
Between 27.4 and 28.3 minutes
c)
Over 28.9 minutes
µ=28.0 minutes and σ=0.3 minutes
σ
0.05%
0.05%
R:
a) 0.05+0.1+2.1 = 2.25% (in graph due to
rounding). In table: 2.28%
b) 13.6+34.1+34.1 = 81.8%
c) 1-(0.15+99.7) = 0.15%
0.15%
0.15%
27.1
27.4
27.7
28.0
28.3
28.6
28.9
PARETO GRAPH
100.0%
120
110
93.3%
89.1%
n=120,
Dec. 2015
120
80.8%
100
107
90
80
70
60.0%
70
50 %
36.7%
50
30
40
44
30
44
20
28
25
10
20
10
5
8
0
10
0
A
B
Pareto Graph Objectives:
1. Stratify the whole into its parts for
better understanding of the
situation.
2. Be able to Prioritize important in
order to Focus on key issues.
60
72
60
40
90
97
80
Frequency
112
100
C
D
E
Others
Interpretation:
•
Claim type A represents 36.7% of
all claims.
•
Claims A, B &C represent 80.8%
of all claims.
•
From the frequency standpoint,
our priority is to reduce claims
type A,B&C.
Brain Flex
Analyze the following Pareto graph.
Which one is your most important
issue to improve?
Why Why Analysis
A Simple and powerful Root Cause Analysis technique
Why the user complains about the service received?
Because of the errors in the bill account.
Why are they complaining about errors in the bill account?
Promised discount in an ad is not being applied
Why are promised discounts not being applied?
Online web based discount is applied, however “face to face” operations in Region 3 are
charging without applying the discount.
Why are Region 3 sales executives did not apply the discount?
Not aware of the existence of such a promotion and as their do a manual process without web
services, our “automated sales system” was not able to prevent these cases.
Why did Region 3 sales executives did not know about this particular promotion?
50% of Region 3 agents are commission agents (contractors) and were not considered in the
mailing list and two way communication protocols.
“Errors in the bill account occur because commission agents (contractors) only have
currently serving distant populations are not in the mailing list”
FOCUS ON THE CAUSES, NOT REASONS
Cause & Effect Diagram
The Cause & Effect diagram was first used by
Dr. Kaoru Ishikawa of the University of Tokyo to
organize a discussion with Kawasaki Steel Co.
managers in 1953. This diagram is used to
identify all of the contributing possible causes
(factors) likely to be causing an effect (problem
or deviation).
Outside Japan, is also known as Ishikawa
Diagram and/or Fishbone Diagram.
This tool offers several benefits:
– Visual tool
– Straightforward and easy to learn
– Promotes participation of the whole team
– Organizes discussion & maintains focus
– Promotes systems thinking
– Supports further analysis and corrective
actions
Video: The Traditional 4M + New Categories
Man Machine Materials Method + Management Environment
MANAGEMENT
MACHINE / EQUIPMENT
MAN / PEOPLE
PROBLEM
STATEMENT
ENVIRONMENT
METHOD / PROCEDURE
MATERIALS / INPUT
Key Recommendations for an Effective Ishikawa Diagram
•
•
•
•
•
•
•
•
•
•
Correct definition of the problem and its statement.
Invite key stakeholders who live, know and understand the process where the
Improvement will be implemented.
Brainstorm all possible causes of the problem and write them in sticky notes (one
cause, one note).
Avoid “One Word” sticky notes. Try to write full sentences.
Avoid writing explanations: “look for causes, do not explain the problem”.
Avoid writing justifications: “look for the origin of the problem, not justify your actions”.
Avoid judging the current situation or causes: “Bad procedure”, “Inadequate operation”
Avoid writing causes as a “lack of your solution”: “Lack of training”, “not enough time”
Check that all participants understand all written causes in the same way.
You can decide to use the 4M rule or create your own categories for your diagram.
Pie Chart
Sales (million USD)
Generally useful to
show percentage
or proportional
data and when
trying to compare
parts of a whole.
D; 5; 4%
C; 15; 11%
B; 30; 23%
A; 80; 62%
A
B
C
D
Bar graphs
Bar or column
100% Stacked column
%
100%
Jan Feb Mar Apr May Jun Jul Aug Sep
Jan Feb Mar Apr May Jun
Jul
Aug Sep
Radar Chart (Spider chart)
Skill 1
10
Test result Z
100
Skill 2
10
5
5
Test result Y
50
100
Skill 3
5
10
50
0
5
C
C-
10
Performance
evaluation
Skill 4
A
A+
Personality test 1
Graphical method to displaying
multivariate data in a 2 dimentional
char of 3 or more radial axis
representing quantitative and
qualitative variables.
Very useful to view one process or
person vs different variables when
a balance between all the variables
is desired.
Boxplot (Box & Whisker)
Graphical representation of a data distribution
(center, width and outliers) based on:
• Minimum or smallest value of data set
• First or Lower Quartile (Q1)
• Second Quartile or Median (Q2)
• Third or Upper Quartile (Q3)
• Maximum or largest value of data set
• “Outliers” (value that lies at an abnormal distance
from other values or statistically outside current data
set distribution).
• Note: For symetric distributions, statistical limits of
the distribution (the whiskers) are calculated as
follows:
• Q3 + 1.5 IQR or largest value
• Q1 – 1.5 IQR or smallest value
Where IQR = Q3 – Q1 (the Box)
Scatter Plot Graph
n=50
Analyses possible
relationship between two
variables:
• Dependent (Y) –
response variable (effect)
• Independent (x) - cause
Y
X
n=50
Y
Y
Y
X
n=50
n=50
X
IMPORTANT:
Linear correlation between two variables does not necessarily imply a causal relationship
X
The following Scatter Plot graph shows the relationship between First Call Resolution
and Net Promoter Score. The higher the number of issues resolved at the very first
attempt (FCR), the more likely the customer is going to recommend the company to
other potential customers (NPS).
In this graph, NPS represents
the response variable (effect)
and FCR represents the
Independent variable (cause)
Run Charts (line graph / time series)
A line graph representing the observed data (measurements) made over the time to monitor
the behavior of a process or system variable.
• Measured data come from a result (Lagging variable).
• Measured data come from process controls (Leading variables)
LAGGING = Result or effect: Profits, Total process time, CSAT, resolution rates
LEADING = Cause or origin: Cost of waste, stoppages, accuracy rate, errors
Run Chart
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
1
2
3
4
5
6
7
8
9
10
11
12
13
Note: This are dynamic graphs as behavior over the time can be analized
14
15
Run Charts (Call for Action)
Possible out of control situations.
40.00
OUTLIER
SHIFT (BIAS)
40.00
30.00
30.00
20.00
20.00
10.00
10.00
0.00
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
TREND
RUNS
40.00
40.00
30.00
30.00
20.00
20.00
10.00
10.00
0.00
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Run Charts (Call for Action)
Possible out of control situations.
CYCLES
40.00
TREND & DISPERSION
40.00
30.00
30.00
20.00
20.00
10.00
10.00
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.00
1 2 3 4 5 6 7 8 9 1011 12131415
DISPERSION
DIFFERENT SOURCES /
STRATIFICATION
40.00
30.00
40.00
20.00
30.00
10.00
20.00
0.00
10.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Control Limits vs Specification Limits
Control Limits (Statistical Control Limits)
•
•
•
Based on process performance. Represents the “Voice of the Process”.
Determine process “center” and “width” estimated with central tendency and dispersión measures.
Variation is a consequence of natural causes (chance).
Lower
Control
Limit (LCL)
Lower
Specification
Limit (LSL)
Upper
Control
Limit (UCL)
Upper
Specification
Limit (USL)
Specification Limits (Technical standards, targets, norms, customer requirements)
•
•
Provided by customers, sector and or regulations.
Define the “Must be” of “Shall comply” specifications which define a conforming vs non conforming service.
Process Variation vs Specifications
How capable is my process to comply with customer specifications?
Specification limits
1
BIAS
Deviation vs Center
2
Control limits (process)
DISPERSION
3
ABNORMALITY
(Assignable cause)
Process Capability
The ability of the process to meet or comply with
customer specifications or requirements.
Car
width
Parking
space
width
Car
width
Parking
space
width
Control limits
(process)
Specification
limits
Control limits
(process)
Car vs Parking Space
Specification
limits
SL
PL
Process
Capability.
PL: Process Control Limits
SL: Specification Limits
-4 -3 -2 -1 0 1 2 3 4
SL
PL
-4 -3 -2 -1 0 1 2 3 4
Video: Process Control
n=100
Statistical Process Control (SPC)
•
•
•
Control graphs were developed
by Walter A. Shewhart an
published in 1924.
A tool for process performance
and indicators analysis and
control.
Separates variation caused by:
• Natural or random causes
(by chance).
• Assignable or special
cause (abnormal variation)
Assignable
cause
7
6
5
p 4
(%) 3
2
1
0
Natural
cause
Asignable
cause
0
2
4
6
8
1
0
1
2
1
4
1
6
1
8
2
0
2
2
2
4
2
6
2
8
days
días
Assignable
cause
Natural
cause
Assignable
cause
Control Graphs
Control Graphs are run charts (time series) with 3 additional lines:
1. Expected value, or mean value: Central Control Limit (CL)
2. Natural variation or statistical process limits indicating the threshold at
which the process output is considered statistically “stable / non
stable” or “under control / out of control”: Upper and Lower Control
Limits (UCL & LCL)
n=100
7
6
5
p 4
(%) 3
2
1
0
UCL
CL
LCL
0
2
4
6
8
1
0
1
2
1
4
days
días
1
6
1
8
2
0
2
2
2
4
2
6
2
8
Control Charts: Basic types
Continuous data (measure)
•
R (Range)
–
•
Analize and control of sample means and its
variation over time.
Analize and control individual data over time
Useful when measuring cost is high and/or long
time interval between
np (defective units)
–
Analize and control of constant size sample
variation over time.
–
•
–
•
Analize and control sample variation over time
(variable sample size > 8)
–
•
Analize and control non conforming products or
services over time with variable sample sizes.
Good/No Good Criteria (“pass / no pass”)
c (Number of defects or errors)
s (Standard Deviation)
–
Analize and control non conforming products or
services over time with constant sample size.
Good/No Good Criteria (“pass / no pass”)
p (fraction defective)
–
x (Individual data)
–
–
•
•
Xbar or 𝒙 (Mean)
–
•
Discrete data (count)
Analize and control number of defects or errors
per constant size lots or samples.
u (Number of defects per unit)
–
Analize and control number of defects or errors
per variable size lots of samples.
Call for Action: Possible out of control situations
1.
2.
3.
4.
5.
6.
Points out of statistical
control limits.
Run: 7 o more points
on same side of
average.
10 out of 11
consecutive points on
same side of average
(12 of 14, 14 of 17, 18
of 20)
Trends.
Cycles.
2 of 3 consecutive
points (3 of 7) in the
last third of the control
limits.
Chart Selection Tree
Conventional process vs. Continuous
Improvement approaches
IDENTIFY A
PROJECT
STANDARDIZATION
Conventional approach.
PROJECT
IDENTIFICATION
VERIFICATION
OF RESULTS
Continuous Improvement
approach
DEFINITION
OF THE PROBLEM
& GOALS
IMPLEMENT THE
SOLUTION
IMPLEMENTATION
OF
SOLUTIONS
CAUSE
ANALYSIS
DEFINE THE
SOLUTION
SOLUTIONS
ANALYSIS
Video: Golden Rule for Improvement
1.
Who is being affected?


2.
Customer focus.
Containment / Corrective actions.
What process caused the problem? / What processes generate the
problem?


3.
Apply process Improvement and problem solving steps.
Corrective and preventive actions.
What skills do people need to know to prevent recurrence of the
problem?


Actions to develop and improve people skills.
Actions to replicate to other Business units / processes.
The DMAIC Cycle
DMAIC Methodology Overview
DEFINE THE PROBLEM
The customer and the goals
MEASURE KEY ASPECTS
Current processes and relevant data
ANALYZE THE DATA
Understand the cause and it’s effects
IMPROVE AND OPTIMIZE
Review existing policies and procedures
CONTROL FUTURE RESULTS
Anticipate potential deviations and compensate
Improvement process & DMAIC
Problem solving steps
DMAIC 6 Sigma
1. Quantitative selection of the problem
DEFINE.
a)
Validate Business opportunity and project CTQ (Critical to
Quality) identification (“Y”)
b)
Team chartering and project storyboarding / “Project
Charter”
c)
Analyze current process (SIPOC and “As is Process” map)
2. Understanding of current situation
MEASURE.
a) Collect data.
b) Validate measurement system (MSA)
c) Assess current Process Capability.
3. Cause analysis and root causes definition
ANALIZE.
a) Cause Analysis (“X´s”)
b) Validate causes
4. Solutions analysis
IMPLEMENT AND IMPROVE.
a) Generate potential solutions and validate pilot solutions.
b) Assess their impact.
5. Implementation of solutions
6. Verification of the impact of solutions
7. Standardization and control
8. Conclusions and next steps
CONTROL.
a) Evaluate final impact of solutions.
b) Standardize, normalize and document.
c) Final evaluation and conclusions.
Define the Problem
STEP 1:
Validating Business Opportunity and Project
CTQ (Critical To Quality) Identification
STEP 2:
Problem statement
Project Storyboarding and Team Chartering
STEP 3:
SIPOC and As-Is Process Mapping
Business
Problem
Measuring Key Aspects
STEP 4:
Identify Possible Project X's and
Data Collection Strategy
Go and “walk” the process
STEP 5:
Validate Measurement System
STEP 6:
Determine Process Capability
Business
Problem
Analyze the Data
STEP 7:
Identify possible causes of the problem
Identify Vital Project X’s (root causes) and
validate them
Statistical
Problem
Statistical
Solution
Improve and Optimize
STEP 8:
Generate Potential Solution and Assess their
Failure Mode (risks and assumptions).
STEP 9:
Validate Pilot Solution(s).
Assess secondary effects.
STEP 10:
Process Control and Risk Analysis.
Reduce risk adding required actions.
Statistical
Solution
Business
Solution
Control Future Results
STEP 11:
Execute control Plan and Finalize
Documentation.
STEP 12:
Communicate to Business Results and sign
off to close the CI Project.
Statistical
Solution
Business
Solution
VOC Methodologies
There are several ways to gather information.
Observations
Focus Groups
Suggestions
Interviews
Surveys
Distinct Categories
VOC has 4 distinct categories – the AICP
Associate
Investor
Customer
Process
SIPOC
High Level Process Map
VOC to CTQ
NEED
DRIVERS
Taste
Critical To Quality
Requirements
Not acidic
Rich flavor
Customer says:
“I want coffee that
that is good.”
Temperature
< 176° F
Cost
Difficult to quantify
> 140° F
< $3.50/cup
Easier to quantify
What is a problem statement?
•
•
The difference between my current situation compared to the a desired condition
(reference).
A Problem is a deviation (distance) from my target, goal, especification, promise
or requirement.
Problem statement:
Problem is a
Deviation from
desired
condition
“Bad service”
vs
“20 out of 100 claims are not
solved before 24 hours (our
standard)”
What is a problem statement?
•
•
The difference between my current situation compared to the a desired condition
(reference).
A Problem is a deviation (distance) from my target, goal, especification, promise
or requirement.
Problem statement:
Problem is a
Deviation from
desired
condition
“Bad service”
vs
“20 out of 100 claims are not
solved before 24 hours (our
standard)”
Problem statement
YES
NOT SO GOOD
•
A problem shall be understood as a “Difference
between the desired or mandatory condition
(reference point) and the current situation”.
•
A description of the undesired condition, the
effects and symtoms of the problem, or a
justification of past actions.
•
Magnitude of the problem = Size of the gap
between current situation and reference point.
•
Magnitude of current condition without
comparing it to the reference point.
•
“Solving problem is reducing differences and
achiving our desired condition”.
•
A company has many problems never
addressed with a disciplined approach.
•
Good statements:
•
“Bad statements”:
–
Non conforming calls in B division exceed our
standard by 1 minute.
–
Sales dropped 8% compared to last Quarter.
–
Last month, we missed our target by $500,000
dollars.
–
Many calls take too long.
–
Low sales.
–
Bad attitude of our sales personnel.
Difficulty (problem or solution) vs. Impact
Simple / High Impact
Complex / High Impact
PRIORITY 1
PRIORITY 2
EVERYONE
MANAGERS (TOP & MIDDLE)
PRIORIDAD 3
Difficulty / Impact
LEARNING PROJECT
Simple / Low Impact
WATCH OUT:
NEED FOR ANOTHER
METRIC TO DECIDE?
Complex / Low impact
R. Hirata, 2013 ©
My Project Charter
•
•
Project or Improvement title (problem
statement).
Project justification.
–
–
–
•
•
Type of project:
–
–
•
M
A
I
Current project goal (increase or reduce)
Big “Y” de Qualfon (Region / Country).
Target statement
Time frame.
Difficulty and Impact Matrix criteria.
DMAIC or other type of project.
Scope: Process or department :
Time
D
Contribution to a higher level goal or target.
Project Scope.
“Big “Y” of Qualfon (Corporate or Regional).
C
End
Target or tangible results.
–
–
Program
Step
Project objective.
–
–
•
•
•
Documentation and closing
Members (in case it applies)
Role
Name
Department
Short Improvement registration
Before
Impact / Benefits
After
Process
Mapping
3 Ways to Maximize Process Maps
1. Process maps
2. Process flowcharts
3. Deployment flowcharts
Maps and Flowcharts
Utilizing a graphic display of steps to
indicate how each step is to be carried out.
Data Types
Measuring and identifying the results
Qualitative Data
Subjective and not
measured or
counted objectively
Ej: Type of claim,
description of a
process
Continuous
Data
Measured
Quantitative Data
Objective and
measured or
counted
Discrete
Data
Counted
Ej. Number of
claims, process lead
time
Sampling
Measuring through Data Collection Strategies
The process of gathering data by selecting a small
number of elements from larger, defined group.
2 kinds of Sampling:
1. Probability Sampling
2. Non-probability Sampling
Sample
Population
NEVER USE SAMPLING WHEN
events and products are unique and
cannot be replicated.
Sampling
Measuring through Data Collection Strategies
2 general sampling methods.
Sample
Target
population
(goal)
Total
population
(theoretical)
Probability Sampling.
•
•
•
All individuals have the samne
probability of being included in a simple.
All samples have the sme probability of
being selected.
The results are useful to generalize
about the population and other samples.
Non Probability Sampling.
•
•
Not all individuals have the same
probability of being selected.
They are not useful to generalize about
the population or other samples.
Probability Sampling
Methods for probability sampling
SIMPLE RANDOM SAMPLING
Every unit has the equal chance
of being selected
STRATIFIED RANDOM SAMPLING
Stratum / groups are created. Then
random sampling on each group is
implemented
SYSTEMATIC SAMPLING
Divide the elements into groups.
Select every “n”th unit from
each group
CLUSTER SAMPLING
Random Sample of selected clusters
Non Probability Sampling
Methods of Non Probability Samplinb
Volunteers /
Judgment
Convenience /
Intentional
Based on convenience and
ease of access to data.
Based on people willing
to participate
(ex. Customers in this cafeteria,
exit polls on site X)
(ex. Check out survey at a
hotel)
Sample or
population
Based on relevant
characterstics needed for
the research .
Based upon respondent
referrals of others with like
characteristics
(ex. 20 Asian origin people on
Mexico City)
(ex. Customers with similar problem
or situation)
Quota
Snowball
Solution Parameters
Decision Statement
Explains your purpose and
intent.
Solution Criteria
Lists 6 to 12 criteria to
solve the problem.
Classify Solution Criteria
Identifies “musts” and
”wants”.
Refine Solution Criteria
Discuss. Understand.
Restate.
SCAMPER Tool
Refine Solution Criteria
Assumption Busting
Benchmarking
Creative
Brainstorming
Brain-writing
Techniques
Idea Generation
Modified
Brainstorming
Musts and Wants
• Identify how solutions will perform against the
solution criteria
• Assess based on the best available information
• Best projections
• Guide for data gathering
Cost Benefit Analysis
Weigh the real cost of a
potential solution and
compare against
expected benefits.
Identify associated (and
potential) risks. Are they
worth it? Can it be
mitigated?
Failure Mode & Effect Analysis
FMEA identifies all probable
failures and prioritizes those
for focused attention by using a
scoring model based on
severity (S), occurrence (O),
and detect ability (D).
RPN = S*O*D
Terms to remember:
• Failure mode
• Effect
• Cause
• Current controls
• Severity
• Occurrence
• Detection
• RPN
Pilot Solution
Implementation
When to and why pilot?
• The scope is large, e.g. enterprise-wide
• The change would have far-reaching unintended consequences
• Implementation is costly
• Change can be difficult to reverse
Pilot Requirements
Key steps involved in conduction pilots
1.
2.
3.
4.
5.
6.
7.
8.
Strong top management leadership
A steering committee (pilot team)
Briefing the pilot team
Pilot planning for issueless execution
Cooperation of affected employees
Employee training for pilot execution
Pilot implementation on the shop-floor
Debriefing after pilot implementation and extend to a second area, if required.
Control & Implementation
Control and Implementation Plans
Standardization
Documentation
• What are the steps in
the process?
• Procedures
improvement.
• Who does these steps in
the process and when?
• Forms and templates
improvement.
• Where more detailed
work instructions can be
found?
• Metrics, indicators and
measurement
improvement.
• Improve protocols for
decision making.
Monitoring Plan
• Monitoring to verify
improvements are in
place and working
• Monitoring to detect
new deviations and
problems.
• Assure the problem will
not appear again.
Response Plan
A response plan helps
identify the next steps on
what needs to be done if
one detects a change in
the process while
monitoring
What was done?
What was not done?
Next Steps?
Implementation
Control and Implementation Plans – Response
Key Elements
•
•
•
•
•
•
•
•
Clear Objectives
Pilot Learning's Incorporated
Implementation Milestones
Resource Needs
Influence Strategy
Implementation Budget
Process Control Plan
Process Documentation
Review & Sign-Off
When defect reduction is demonstrated over a pre-agreed amount of time, the
project is “turned off”, only to be “turned on” when the old process is used again.
The key in this step is the “on-off” idea.
Thank you