QUALITY CONTROL
Quality Control
QUALITY CONTROL
Costs of Quality
• Internal Failure Costs
– Costs incurred to fix problems that are detected
before delivery to the customer.
• External Failure Costs
– Costs incurred to fix problems that are detected
after delivery to the customer.
• Appraisal Costs
– All product and/or service inspection costs.
• Prevention Costs
– Cost of preventing defects (Training, Quality
Planning,Process control, Quality Improvement)
QUALITY CONTROL
Key Contributors to Quality Management
Contributor Known for
Deming Analysis of variation
Juran Quality is fitness for use; quality trilogy
Crosby Quality is free; zero defects
Ishikawa Cause-effect diagrams; quality circles
Taguchi Statistical experiment design
QUALITY CONTROL
Quality Control
• Purpose: To assure that processes are
performing in the desired manner (conformance
quality). Phases of Quality assurance:
Inspection Corrective Quality built
before/after action during into the
production production process
Acceptance Process Continuous
sampling control improvement
The least The most
progressive progressive
QUALITY CONTROL
Inspection
• Monitoring the production process
– How Much/How Often
– Where/When
Inputs Transformation Outputs
Acceptance Process Acceptance
sampling control sampling
QUALITY CONTROL
How Much/How Often to Inspect
• Cost Trade-off:
– Cost of inspection
– Cost of passing defectives
• Low Cost-High Volume Items: Less
• High Cost-Low Volume Items: More
• Manual or labor intensive operations: More
QUALITY CONTROL
Where/When to Inspect in the Process
• Raw materials and purchased parts
• Finished products
• Before a costly operation
• Before an irreversible operation
• Before an operation that covers up defects
• On-site vs Centralized?
– On-site: Faster detection of errors & decisions
– Centralized: Maybe required by specialized
testing equipment
QUALITY CONTROL
Statistical Process Control
Sample Inspect
Process
of size n Sample
Outputs
IN CONTROL
Sample
ACTION ! Compare
Information
OUT OF CONTROL
Decision
Criteria
QUALITY CONTROL
Variations and Control
• All processes yield some variation in outputs
– Random variation: Natural variations in the output
of process, created by countless minor factors
– Assignable variation: A variation whose source
can be identified
Sample Statistics (e.g. Average, Range):
Represent the random variability as obtained
from the samples of the outputs.
Sampling Distribution: Distribution of sample
averages (Different from Process Dist.)
QUALITY CONTROL
Sampling Dist. Vs Process Dist.
• Central Limit Theorem: Sampling D. is Normal
even if the Process D. is not.
X Average of Sample Means = (Population Mean)
(Population Std.)
X Std. of Sample Means =
n (n = Sample Size)
Sampling
distribution
Process
distribution
Mean
QUALITY CONTROL
Control Charts
• Time-ordered plot of sample statistics
• Has Control Limits (LCL,UCL)
Sampling
distribution
Process
distribution
/2 Lower Mean Upper /2
control control
limit limit
• Limits are set typically at 2 3
• Type I error ( risk) : Concluding that the process is out of control when
it is not
QUALITY CONTROL
Control Charts
• Used to distinguish between random and non
random variation.
Abnormal variation Out of
due to assignable sources control
UCL
Mean
Normal variation
due to chance
LCL
Abnormal variation
due to assignable sources
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Sample number
QUALITY CONTROL
Control Charts for Variables
• Data is measured
= Population Std n = Sample Size
X Sample mean Std X Sample mean Aver.
n
Range = Highest - Lowest Value in the sample
R Average of Sample Ranges
z = Standard Normal Deviate
• Mean Chart: Measure central tendency
• Range Chart: Measure dispersion
QUALITY CONTROL
Mean Charts
UCL X z X
LCL X z X
• If is not known, use an estimate. (E.g. the sample standard deviation s). Or
UCL X A2 R
LCL X A2 R
QUALITY CONTROL
Range Charts
UCL D4 R
LCL D3 R
QUALITY CONTROL
Example
• Time to complete a manufacturing
operation. From experience, = 0.25 mins
Sample No Observations Average Range
1 10.6 10.4 9.3 10.2 10.125 1.3
2 9.98 10.25 10.05 10.33 10.1525 0.35
3 9.85 9.9 10.2 10.25 10.05 0.4
4 10.2 10.1 10.3 10.15 10.1875 0.2
5 10.24 10.3 10.2 9.9 10.16 0.4
6 9.98 9.9 10.2 10.3 10.095 0.4
Average 10.12833 0.508333
X = 10.128 X = = 0.125
n
R 0.508
QUALITY CONTROL
X = 10.128 X = = 0.125 R 0.508
n
• 3- Control Limits on the Mean:
UCL X z X = 10.128 + 3 (0.125) = 10.503
LCL X z X 10.128 3( 0.125) 9.753
UCL X A2 R = 10.128 + 0.73(0.508) = 10.499
LCL X A2 R 10.128 0.73( 0.508) 9.757
• 3- Control Limits on the Range:
UCL D4 R = 2.28 (0.508) = 1.159
LCL D3 R 0( 0.508) 0
QUALITY CONTROL
Mean and Range Charts
(process mean is
shifting upward)
Sampling
Distribution
UCL
x-Chart Detects shift
LCL
UCL
Does not
R-chart
detect shift
LCL
QUALITY CONTROL
Mean and Range Charts
Sampling
Distribution (process variability is increasing)
UCL
x-Chart Does not
reveal increase
LCL
UCL
R-chart Reveals increase
LCL
QUALITY CONTROL
Control Chart for Attributes
• p-Chart - Control chart used to monitor the
proportion of defectives in a process
– When observations can be placed into two
categories like (Good or bad) (Pass or fail)
• c-Chart - Control chart used to monitor the
number of defects per unit
– Used only when the number of occurrences per unit
of measure can be counted; nonoccurrences cannot
be counted.
• Scratches, chips, dents, or errors per item
• Cracks or faults per unit of distance
• Calls, complaints, failures per unit of time
QUALITY CONTROL
P-Chart
p Percent defect in the population
p(1 p)
p
n
UCL p z p
LCL p z p
• If p is not known, use an estimate
Total # Defects
p
Total # of Observations
QUALITY CONTROL
Example
• 6 samples of 15 parts taken from a process
Sample 1 2 3 4 5 6 Total
# of defects 3 1 0 2 0 3 9
Total # Defects 9
p 0.1
Total # of Observations 15( 6)
p(1 p) 0.1(1 0.1)
p 0.077
n 15
• 3- Control Limits
UCL 01
. 3 (0.077) = 0.331
LCL 01
. 3(0.077) 0131
. 0
QUALITY CONTROL
c-Chart
c Mean # of defects per unit in the population
UCL c z c
LCL c z c
• If c is not known, use an estimate
Total # Defects
c
Total # of Units Observed
QUALITY CONTROL
Example
• 8 sample units are taken from a process:
Sample Unit 1 2 3 4 5 6 7 8 Total
# of defects 3 2 4 5 1 4 2 3 24
Total # Defects 24
c 3
Total # of Units Observed 8
• 3- Control Limits
UCL 3 3 3 = 8.196
LCL 3 3 3 2.196 0