AP Statistics CHAPTER 17: CHAPTER 17: PROBABILITY MODELS Unit 4.
chapter 17
Transcript of chapter 17
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Chapter 17
Quality planning and control
Source: Archie Miles
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Quality planning and control
Operations strategy
Design Improvement
Planning and control
Operations management
Quality planning and control
The operation supplies …the consistent delivery of products and services at specification or above
The market requires … consistent quality of products
and services
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
The various definitions of quality
The transcendent approach views quality as synonymous with innate excellence.
The manufacturing-based approach assumes quality is all about making or providing error-free products or services.
The user-based approach assumes quality is all about providing products or services that are fit for their purpose.
The product-based approach views quality as a precise and measurable set of characteristics.
The value-based approach defines quality in terms of ‘value’.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Quality upQuality up
Profits upProfits up
Processing time down
Processing time down
Inventory down
Inventory down
Capital costs down
Capital costs downComplaint and
warranty costs down
Complaint and warranty costs
down
Rework and scrap costs
down
Rework and scrap costs
down
Inspection and test costs
down
Inspection and test costs
down
Productivity up
Productivity up
Service costs downService
costs down
Image upImage up
Scale economies up
Scale economies up
Price competition
down
Price competition
down
Sales volume up
Sales volume up
Revenue up
Revenue up
High quality puts costs down and revenue up
Operation costs down
Operation costs down
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Customers’ expectations
for the product or
service
Customers’ perceptions
of the product or
service
Gap
Perceived quality is poor
Perceived quality is good
Expectations > perceptions
Expectations = perceptions
Expectations < perceptions
Perceived quality is governed by the gap between customers’ expectations and their perceptions of the product or service
Gap
Perceived quality is acceptable
Customers’ expectations
for the product or
service
Customers’ perceptions
of the product or
service
Customers’ expectations
for the product or
service
Customers’ perceptions
of the product or
service
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
The operation’s domain
Management’s concept of the
product or service
The customer’s domain
Previousexperience
Word-of-mouth communications
Image of product or service
Customer’s own specification of
quality
Organization’s specification of
quality
The actual product or service
Customer’s expectations concerning a
product or service
Customer’s perceptions
concerning the product or service
Gap 1
Gap 2Gap 3
Gap 4
A ‘gap’ model of quality
Gap ?
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
The perception–expectation gap
Action required to ensure high perceived quality
Main organizational responsibility
Gap 1
Gap 2
Gap 3 Operations
Gap 4 Marketing
Ensure consistency betweeninternal quality specification andthe expectations of customers
Ensure internal specification meets its intended concept of design
Ensure actual product or service conforms to internally specified quality level
Ensure that promises made to customers concerning the product or service can really be delivered
Marketing, operations, product/service development
Marketing, operations, product/service development
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Quality characteristics of goods and services
Functionality – how well the product or service does the job for which it was intended
Appearance – the aesthetic appeal, look, feel, sound and smell of the product or service
Reliability – the consistency of performance of the product or service over time
Durability – the total useful life of the product or service
Recovery – the ease with which problems with the product or service can be rectified or resolved
Contact – the nature of the person-to-person contacts that take place
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Attribute and variable measures of quality
Attributes Variables
Defective or not defective?Measured on a continuous scale
Light bulb works or does not work Diameter of bulb
Number of defects in a turbine blade Length of bar
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Variablesthings you can measure
Attributesthings you can assess and accept or reject
Qualityfitness for purpose
Reliabilityability to continue
working at acceptedquality level
Quality
Quality of designdegree to which
design achieves purpose
Quality of conformancefaithfulness with which the
operation agrees with design
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Time
Som
e m
easu
re o
f op
erat
ions
per
form
ance
Some aspect of the performance of a process is often measured over time
Question:
“Why do we do this?”
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Time
Som
e m
easu
re o
f op
erat
ions
per
form
ance
Some aspect of the performance of a process is often measured over time
Question:“How do we know if the variation in process performance is ‘natural’ in terms of being a result of random causes, or is indicative of some ‘assignable’ causes in the process?”
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Time
Ela
psed
tim
e of
cal
lThe last point plotted on this chart seems to be unusually low.How do we know if this is just random variation or the result of some change in the process which we should investigate?
Some kind of ‘guidelines’ or ‘control limits’ would be useful.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
0.8 2.2 3.6
After the first sample
0.8 2.2 3.6
After the second sample
0.8 2.2 3.6
By the end of the first day
0.8 2.2 3.6
By the end of the second day
0.8 2.2 3.6Fitting a normal
distribution to the histogram of sampled
call times
Process control charting
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
The chances of measurement points deviating from the averageare predictable in a normal distribution
40 100 160Elapsed time of call (seconds)
Fre
quen
cy
68% of points
–2 standarddeviations
+2 standarddeviations
95.4% of points
–3 standarddeviations
+3 standarddeviations
99.7% of points
–1 standarddeviation
+1 standarddeviation
A standarddeviation
= sigma
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Time
Ela
psed
tim
e of
cal
lProcess control charting
If we understand the normal distribution, which describes random variationwhen the process is operating normally, then we can use the distributionto draw the control limits.
In this case the final point is very likely to be caused by an ‘assignable’ cause,i.e. the process is likely to be out of control.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
A P A P
A PA P
X
XX
X
Process variability
Scatter – precision: P
On/off target – accuracy: A
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Alternating and erratic behaviour – investigate!
UCL
C/L
LCL
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Suspiciously average behaviour – investigate!
UCL
C/L
LCL
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Two points near control limit – investigate!
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
UCL
C/L
LCL
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Five points on one side of centre line – investigate!
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
UCL
C/L
LCL
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Apparent trend in one direction – investigate!
UCL
C/L
LCL
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Process control charting
Sudden change in level – investigate!
UCL
C/L
LCL
In addition to points falling outside the control limits, other unlikely sequences of points should be investigated.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Low process variation allows changes in process performance to be readily detected
Time
Process distribution A
A
Time
Process distribution A
Process distribution B
A
B
Process distribution B
B
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
USLLSL
Process variation
Process variation
Process variation
Process variation
3 sigma process variation
= 66800 defects per million opportunities
4 sigma process variation
= 6200 defects per million opportunities
5 sigma process variation
= 230 defects per million opportunities
6 sigma process variation
= 3.4 defects per million opportunities
Process variation and its effect on process defects per million opportunities (DPMO)
USLLSL USLLSL USLLSL
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Percentage actual defective in the batch
Pro
babi
lity
of a
ccep
ting
the
batc
h
0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
1.0
Producer’s risk (0.05)
Consumer’s risk (1.0)AQL LTPD
0 0.060.050.040.030.020.01 0.07 0.08
In this ideal operating characteristic,the probability of accepting the batch
if it contains more than 0.04% defective items is zero, and the probability of
accepting the batch if it containsless than 0.04% defective items is 1
In this real operating characteristic (where n = 250 and c = 1), both
type 1 and type 2 errors will occur
Type 1 error
Type 2 error
Ideal and real operating characteristics
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Key Terms TestQualityConsistent conformance to customers’ expectations.
Quality characteristicsThe various elements within the concept of quality, such as
functionality, appearance, reliability, durability, recovery, etc.
Quality samplingThe practice of inspecting only a sample of products or
services produced rather than every single one.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Key Terms TestStatistical process control (SPC)A technique that monitors processes as they produce products or
services and attempts to distinguish between normal or natural variation in process performance and unusual or ‘assignable’ causes of variation.
Acceptance samplingA technique of quality sampling that is used to decide whether to
accept a whole batch of products (and occasionally services) on the basis of a sample; it is based on the operation’s willingness to risk rejecting a ‘good’ batch and accepting a ‘bad’ batch.
Control chartsThe charts used within statistical process control to record process
performance.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Key Terms Test
Process capabilityAn arithmetic measure of the acceptability of the variation of a
process.
Control limitsThe lines on a control chart used in statistical process control to
indicate the extent of natural or common-cause variations; any points lying outside these control limits are deemed to indicate that the process is likely to be out of control.
Quality loss function (QLF)A mathematical function devised by Genichi Taguchi that
includes all the costs of deviating from a target performance.
Slack, Chambers and Johnston, Operations Management 5th Edition © Nigel Slack, Stuart Chambers, and Robert Johnston 2007
Key Terms TestSix SigmaAn approach to improvement and quality management that
originated in the Motorola Company but was widely popularized by its adoption in the GE Company in America. Although based on traditional statistical process control, it is now a far broader ‘philosophy of improvement’ that recommends a particular approach to measuring, improving and managing quality and operations performance generally.
Zero defectThe idea that quality management should strive for
perfection as its ultimate objective, even though in practice this will never be reached.