Transcript Chapter6

Managing Quality
Introduction



What: quality in operations
management
Where: Quality affects all goods and
services
Why: Customers demand quality
What is Quality



High quality products
Low quality products
What does quality mean to you?
American Society for Quality

“The totality of features and
characteristics of a product or service
that bears on its ability to satisfy stated
or implied needs”
User-Based Definition



“Quality lies in the eye of the beholder”
Higher quality = better performance
Higher quality = nicer features
Manufacturing-Based
Definition


Quality = conforming to standards
“Making it right the first time”
Product-Based Definition

Quality = a measurable variable
Our Definition

Quality: The ability of a product or
service to meet customer needs
Implications of Quality



Company Reputation
Product Liability
Global Implications
Global Implications




National Quality Awards:
US: Malcolm Baldridge National Quality
Award
Japan: Deming Prize
Canada: National Quality Institute
Canada Awards for Excellence
Canada Award Winners 2000




Aeronautical and Technical Services
British Columbia Transplant Society
Delta Hotels
Honeywell Water Controls Business Unit
Quality and Strategy



Differentiation
Cost Leader
Response
Quality and Profitability
Sales Gains
•Improved Response
•Higher Prices
•Improved Reputation
Improved Quality
Increased Profits
Reduced Costs
•Increased Productivity
•Lower Rework, Scrap
•Lower Warranty Costs
Costs of Quality




Prevention Costs
Appraisal Costs
Internal Failure
External Costs
International Standards


ISO 9000
Establish quality management
procedures




Documented processes
Work Instructions
Record Keeping
Does NOT tell you how to make a
product!
Total Quality Management



TQM – Total Quality Management
Quality emphasis throughout an
organization
From suppliers through to customers
W. Edwards Deming
Deming’s 14 Points






Create consistency of purpose
Lead to promote change
Build quality into the product, stop depending
on inspections to catch problems
Build long-term relationships based on
performance instead of awarding business on
the basis of price
Continuously improve product, quality and
service
Start training
Deming’s 14 Points








Emphasize leadership
Drive out fear
Break down barriers between departments
Stop haranguing workers
Support, help and improve
Remove barriers to pride in work
Institute a vigorous program of education and
self-improvement
Put everybody in the company to work on
transformation
TQM Concepts






Continuous Improvement
Employee Empowerment
Benchmarking
Just-In-Time
Taguchi
Knowledge of Tools
Continuous Improvement
Act
Plan
Check
Do
Continuous Improvement



Kaizen
Zero Defects
Six Sigma
Employee Empowerment



Involve employees in every step of
production
High involvement by those who
understand the shortcomings of the
system
Quality circle
Benchmarking



Pick a standard or target to work
towards
Compare your performance
Best practices in the industry
Just-In-Time



Produce or deliver goods just when they
are needed
Low inventory on hand
Keeps evidence of errors fresh
Taguchi Concepts



Quality robustness
Quality Loss Function
Target-oriented Quality
TQM Tools







Check Sheet
Scatter Diagram
Cause and effect diagram (fishbone)
Pareto Chart – 80-20 Rule
Flow Charts
Histogram
Statistical Process Control
Inspection


Attribute Inspection
Variable Inspection
Inspection







At supplier’s plant
Upon receipt of goods from supplier
Before costly processes
During production
When production complete
Before delivery
At point of customer contact
Source Inspection


Employees self-check their work
Poka-yoke
Statistical Process Control




Apply statistical techniques to ensure
processes meet standards
Natural variations
Assignable variations
Goal: signal when assignable causes of
a variation are present
Statistics




Mean
Standard deviation
Natural variation
Assignable variation
Taking Samples
Central Limit Theorem
Central Limit Theorem
As sample size
gets
large
enough,
sampling distribution
becomes almost
normal regardless of
population
distribution.
X
X
Population and Sampling
Distribution
Three population distributions
Distribution of sample means
Beta
Mean of sample means  x
x
Standard deviation of
 x 
the sample means
n
Normal
Uniform
 3 x  2  x  1 x
x
  x  2  x  3 x
(mean)
95.5% of all x fall within  2  x
99.7% of all x fall within  3 x
Central Limit Theorem
Sampling
distribution of the
means
Process
distribution of
the sample
xm
( mean )
Central Limit Theorem
Summary





Mean
Standard Deviation
95.5% within +/- 2σ
99.73% within +/- 3σ
This means that, if a point on the chart
falls outside the limits, we are 99.73%
sure that the process has changed
Central Limit Theorem
Summary
Properties of normal distribution
99.7% of al l x fall
within  3
x
95.5% of al l x fall
within  2
x
x
x 
In Control vs Out Of Control



In control and producing within control
limits
In control, but not producing within
control limits
Out of control
In Control vs Out Of Control
Frequency
Lower control limit
(a) In statistical control and
capable of producing within
control limits. A process with
only natural causes of
variation and capable of
producing within the
specified control limits.
Upper control limit
(b) In statistical control, but not
capable of producing within control
limits. A process in control (only
natural causes of variation are
present) but not capable of
producing within the specified
control limits; and
Size
Weight, length, speed, etc.
(c) Out of control. A process out of
control having assignable causes of
variation.
Setting Limits





Mean of samples means x bar
Standard Deviation of process σ
Standard Deviation of sample means σx
= 
n
Upper Control Limit (UCL) = x  z x
Lower Control Limit (LCL) = x  z x
Making X-Bar Control Charts





Mean (x-bar) chart
Standard Deviation is difficult to
calculate, so we calculate a Range R –
the difference between the biggest and
smallest values in the sample
Value of A2 from chart on page 204
UCL = x  A2R
LCL = x  A2R
Making R Control Charts




Plot the range on the chart
D3 and D4 from chart on page 204
UCL = D4R
LCL = D3 R
What X-Bar and R Charts Tell
Us
Summary: Steps to Create
Control Charts




Collect 20 to 25 samples of n=4 or n=5 from
a stable process and compute the mean and
range for each sample
Compute overall means (X-bar and R-bar),
UCL and LCL
Graph sample means and ranges on control
charts
Investigate points that indicate process is out
of control
Control Charts for Attributes




So far we have been using control
charts for variables: size, length, weight
What about attributes: defective or not
defective
We can measure percent defective – pchart
We can measure count defective – cchart
P-Chart



p-bar = mean fraction defective in the
sample
z = number of standard deviations (2 or
3)
σP = standard deviation of sampling
distribution = p 1  p 
n
P-Chart Continued


UCL = p  z p
LCL = p  z p
C-Chart




Controls number of defects per unit of
output
Average count c-bar
UCL = c  z c
LCL = c  z c
Patterns to Look For
Process Capability

We need a summary measure to tell us
if the process is capable of producing
within the design limts
 Upper Specification Limit  x
C pk  minimum of 
, or
3

x  Lower Specification Limit 

3

where x  process mean
  standard deviation of the process population
What does Cpk Tell Us?
Cpk = negative number
Cpk = zero
Cpk = between 0 and 1
Cpk = 1
Cpk > 1
Acceptance Sampling





Used to control incoming lots of purchased
products
Take random samples of batches (“lots” of
finished product
More economical than 100% inspection
Quality of sample used to judge quality of all
items in lot
Rejected lots returned to supplier or 100%
inspected
Operating Characteristic Curve




Each party wants to avoid costly mistake of
rejecting a good lot
Operating Characteristic (OC) curve describes
how well an acceptance plan discriminates
between good and bad lots
Producer’s Risk α – Probability good lot
rejected
Consumer’s Risk β – Probability bad lot
accepted
Quality Levels


Acceptable Quality Level (AQL) –
Poorest level of quality we are willing to
accept (ie 20 defects per 1000 = 2%)
Lot Tolerance Percent Defective –
Quality level of a lot that we consider
bad – we reject lots of this or poorer
quality (ie 70 defects per 1000 = 7%)
OC Curve
100
95
 = 0.05 producer’s risk for AQL
75
Probability of
Acceptance
50
25
= 0.10
10
Consumer’s
risk for LTPD
0
0
1
Good
lots
2
AQL
3
4
5
6
Indifference zone
7
Percent
Defective
8
LTPD
Bad lots
Average Outgoing Quality
(AOQ)


Sampling plan replaces all defective items
encountered
Determine true percent defective in lot
AOQ 
(Pd )(Pa )(N  n )
N
Pd = true percent defective of the lot
Pa = probability of accepting the lot
N = number of items in the lot
n = number of items in the sample