solve quality management & control problem and analysis
nainasyu1a
qmc-3.pptxQuality Management & Control
Seung-Kuk Paik, Ph.D.
Systems and Operations Management
CSU, Northridge
What is Quality?
"Quality" can be defined in many ways.
1. Quality is defined as “FITNESS FOR USE”: How well a service or product performs its intended purpose.
2. Quality is also defined as “CONFORMANCE TO REQUIREMENTS”: How a service or product conforms to performance specifications.
What is Quality?
3. In a wider sense, "QUALITY" is often considered the degree of excellence whereby products and services may be ranked against each other on a relative basis for selected features and characteristics.
American Society of Quality (ASQ) has accepted the following definition:
QUALITY: The totality of features and characteristics of a product or service that bear on its ability to satisfy stated or implied needs.
What is Quality?
DIMENSIONS OF QUALITY
1. Performance - A product´s primary operating characteristics.
2. Features - Supplements to a product´s basic functioning characteristics
3. Reliability – Consistency of performance
What is Quality?
DIMENSIONS OF QUALITY
4. Durability - A measure of product life.
Serviceability - The speed and ease of repair
6. Aesthetics - Appearance of a product
7. Safety - Will the product perform its function without unnecessarily endangering the user?
Quality and Productivity
Historically, quality was viewed by some as a controlling activity which took place somewhere near the end of a production process, an after-the-fact measurement of production performance.
Efforts to obtain quality products increased the costs associated with making that product.
Thus, quality and productivity were viewed as conflicting; one was increased at the expense of the other.
Costs of Poor Process Performance
Defects: Any instance when a process fails to satisfy its customer.
Prevention costs are associated with preventing defects before they happen.
Appraisal costs are incurred when the firm assesses the performance level of its processes.
Internal failure costs result from defects that are discovered during production of services or products.
External failure costs arise when a defect is discovered after the customer receives the service or product.
7
Deming’s Chain Reaction
Quality and Costs
Costs decrease because of less rework, fewer mistakes, fewer delays, and better use of time and materials
Improve quality
Productivity improves
Stay in Business
Provides jobs and more jobs
Capture the market
Quality Improvement
DEMING´S 14 POINTS
1. Create constancy of purpose toward improvement of products
2. Adopt a quality philosophy
3. Cease dependence on mass inspection
4. End the practice of selecting suppliers on the basis of price alone
5. Improve constantly
6. Institute training on the job
7. Institute leadership
8. Drive out fear
DEMING´S 14 POINTS
9. Break down barriers between departments
10. Eliminate slogans and targets
11. Eliminate work standards that prescribe numerical quotas
12. Remove barriers that rob workers and managers of pride in workmanship
13. Institute a vigorous program of education and self-improvement
14. Put everybody in the organization to work to accomplish the transformation
Quality Improvement
Deming’s Process Improvement Cycle
Establish Expectations
Formalize improvements
Implement
Obtain Support
Compare
Adjust
Define Process
Pilot Assessment
Establish measurements
Flow Diagram
Cause and Effect Diagram
Benchmarking
Brainstorm
Check
Plan
Do
Act
1. It aids in developing methods for measuring and evaluating capabilities of a process.
2. SPC enables us to monitor a process and indicate when the process is in control and when corrective action is needed.
3. It increases efficiency by eliminating redundant or unnecessary activities.
Statistical Process Control
Statistical process control (SPC) is a system or a set of specific techniques for controlling and improving production and service processes.
Some Useful Characteristics:
12
Statistical Process Control
One of the most important tools of statistical process control is the process control chart.
Process Control Chart:
A time ordered plot of sample statistics obtained from an on going process used to distinguish between random and nonrandom variability
Purpose: to monitor process output to see if it is random
The essence of process control chart is to assure that the output of a process is random so that future output will be random
Upper and lower control limits define the range of acceptable variation
Causes of Variation
Two basic categories of variation in output include common causes and assignable causes.
Common causes are the purely random, unidentifiable sources of variation that are unavoidable with the current process.
Assignable causes of variation are any variation-causing factors that can be identified and eliminated, such as a machine needing repair.
14
Control Chart
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
UCL
LCL
Sample number
Mean
Out of control
Normal variation due to chance
Abnormal variation due to assignable sources
Abnormal variation due to assignable sources
Control Charts
A control chart is a graph of the performance of a process over time. For example, a chart might show the temperatures of successive meals during one night.
Superimposed on each graph are lines called upper and lower control limits (UCL and LCL) for the process.
The control limits are so drawn that if the process ever performs outside the control limits, something unusual (assignable cause) has almost certainly happened.
A centerline (usually the mean of the process parameter) is also drawn on the graph.
Thus, a control chart gives a visual answer to the question:
Is this process behaving the way it usually does or something has changed?
Appropriate Use of SPC
Use SPC if process is
Measurable
Repetitive
Prone to occasional drift or dramatic shift
Has quantitative output
Steps In Creating Control Charts
Take samples from the population and compute the appropriate sample statistic
Use the sample statistic to calculate control limits and draw the control chart
Plot sample results on the control chart and determine the state of the process (in or out of control)
Investigate possible assignable causes and take any indicated actions
Continue sampling from the process and reset the control limits when necessary
Control Chart for Attributes
Example: Consider a data entry operation that makes numerous entries daily. On each of 24 consecutive days, subgroups of 200 entries are inspected. The resulting raw data follows:
Day Number of Entries Insp Number Defective FD
1 200 6 .030
2 200 6 .030
3 200 6 .030
4 200 5 .025
5 200 0 .000
6 200 6 .030
7 200 14 .070
8 200 4 .020
9 200 0 .000
10 200 1 .005
11 200 8 .040
12 200 2 .010
13 200 4 .020
14 200 7 .035
15 200 1 .005
16 200 3 .015
17 200 1 .005
18 200 4 .020
19 200 0 .000
20 200 4 .020
21 200 0 .000
22 200 15 .075
23 200 4 .020
24 200 1 .005
Control Chart for Attributes
When the data consists of a series of fraction defectives, the appropriate control chart is a p-chart. A p-chart is suitable when a process outcome does not have a numerical value.
Centerline of a p-chart, p = Total Number of Defectives / (N*n)
where, N=Number of samples, and n=sample size, and
Total Number of Defectives is the aggregate of defectives from all samples, and Control limits, called 3-sigma limits are given by:
In the earlier example, =
UCL(p) = LCL(p) =
P-Chart
Day
Control Charts for Variables
Two common types of control charts
The X-bar( )control chart measures the mean of each sample
reveals any tendency for the process mean to drift or jump around over time
The R (range) chart measures the range of each sample
reveals any tendency of the process to behave more randomly or less randomly over time.
22
Constructing Control Charts
The followings steps are needed:
First, choose the sampling plan and sample size (n). Collect several samples (usually more than 20).
Calculate the range (R) and mean ( ) of each sample
Find the average of the means
Calculate average range ( ) of all the sample
Calculate upper and lower control limits for X-bar chart as below:
Calculate upper and lower control limits for R-bar
Factors for Computing Control Charts
Sample Size, n Mean Factor, A2 Upper Range, D4 Lower Range, D3
2 1.880 3.268 0
3 1.023 2.574 0
4 .729 2.282 0
5 .577 2.114 0
6 .483 2.004 0
7 .419 1.924 0.076
8 .373 1.864 0.136
9 .337 1.816 0.184
10 .308 1.777 0.223
Source: Special Technical Publication 15-C, American Society for Testing
Materials, “Quality Control of Materials,” pp. 63 and 72, 1951
Example for X-bar & R Charts
A restaurant manager is concerned about the temperature of a food
item being served. The manager never wanted the dish to be cooler
than 140 degrees F or warmer than 160 degrees F when it left the
kitchen to be served. The manager decided to do a capability study
over 10 days. Each day, she randomly picked three servings and
tested the temperature just as the dish was being served. The
following are the data that she collected.
Day Reading, 0F Range, 0F , 0F
1 150 160 155 10 155.0
2 140 150 155 15 148.3 =8.50F
3 145 150 150 5 148.3
4 150 150 155 5 151.6 =151.50F
5 130 155 150 25 145.0
6 140 140 145 5 141.6
7 150 150 150 0 150.0
8 155 155 160 5 156.7
9 160 160 160 0 160.0
10 150 160 165 15 158.3
Example of Initial Capability
R chart Limits:
Center Control Line = = 8.5 0F
chart Limits:
Center Control Line =
Initial Capability Study
UCL (160.2)
LCL (142.8)
Process not stable,
not in control
141.6
CL (151.5)
Initial Capability
UCL(21.8)
LCL(0)
25
Process not stable,
not in control
CL (8.5)
Capability Study of Improved Process
Day Reading, 0F Range,0F , 0F
11 151 147 150 4 149.3
12 151 150 147 4 149.3
13 150 152 151 2 151.0
14 150 156 151 6 152.3
15 150 148 152 4 150.0
16 148 151 155 7 151.3
17 152 157 149 8 152.6
18 147 151 150 4 149.3
19 150 150 156 6 152.0
20 150 155 151 5 152.0
= 5.0 = 150.9
R chart limits: LCL = 0, UCL = 12.9
chart limits: LCL = 145.8, UCL = 156.0
Improved Process
Process stable, and
in control
UCL(156)
LCL(145.8)
CL(150.9)
Improved Process
Process stable, and
in control
UCL(12.9)
LCL(0)
CL(5.0)
Process Capability
Process capability is the ability of the process to meet the design specifications for a service or product.
Nominal value is a target for design specifications.
Tolerance is an allowance above or below the nominal value.
A process in statistical control does not necessarily meet the design specification.
Just because a process is in control does not necessarily mean that it is within tolerances.
32
Process Capability
Lower Specification
Upper Specification
Process variability matches specifications
Lower Specification
Upper Specification
Process variability well within specifications
Lower Specification
Upper Specification
Process variability exceeds specifications
Range of Possible Solutions
Redesign the process so that it can achieve the desired output
Use an alternative process that can achieve the desired output
Retain the current process but attempt to eliminate unacceptable output using 100% inspection
Examine the specifications to see whether they really necessary or could be relaxed without adversely affecting customer satisfaction
Capability analysis is the determination of whether the inherent variability of the process output falls within the acceptable range of variability allowed by the specifications for the process output
Process Capability Analysis
Process Capability Ratio
Process capability ratio, Cp =
specification width
process width
Upper specification – lower specification
6
Cp =
1. For a process to be capable, it must have a ratio of at least 1.0
2. A ratio of 1.0 means that 99.73% of the output of a process
can be expected to be within the specifications.
3. The greater the ratio, the greater the probability that the output
of a machine or process will fall within design specifications.
Cp =
Upper Specification - Lower Specification
6s
Insurance claims process
Process mean x = 210.0 minutes
Process standard deviation s = .516 minutes
Design specification = 210 ± 3 minutes
= = 1.938
213 - 207
6(.516)
Process Capability Ratio Example
UCL
p
=
p
+
3
p
(
1
-
p
)
/
n
LCL
p
=
p
-
3
p
(
1
-
p
)
/
n
p
x
UCL
x
=
x
+
A
2
R
LCL
x
=
x
-
A
2
R
UCL
R
=
D
4
R
LCL
R
=
D
3
R
R
x
x
x
X-bar Chart
140
145
150
155
160
12345678910
Days
X-bar
Chart2
| 155 |
| 148.3 |
| 148.3 |
| 151.6 |
| 145 |
| 141.6 |
| 150 |
| 156.7 |
| 160 |
| 158.3 |
Sheet1
| X-bar | R |
| 155 | 10 |
| 148.3 | 15 |
| 148.3 | 5 |
| 151.6 | 5 |
| 145 | 25 |
| 141.6 | 5 |
| 150 | 0 |
| 156.7 | 5 |
| 160 | 0 |
| 158.3 | 15 |
Sheet1
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
| 0 |
Sheet2
Sheet3
R Chart
0
5
10
15
20
25
30
12345678910
Days
Range
Chart3
| 10 |
| 15 |
| 5 |
| 5 |
| 25 |
| 5 |
| 0 |
| 5 |
| 0 |
| 15 |
Sheet1
| X-bar | R |
| 155 | 10 |
| 148.3 | 15 |
| 148.3 | 5 |
| 151.6 | 5 |
| 145 | 25 |
| 141.6 | 5 |
| 150 | 0 |
| 156.7 | 5 |
| 160 | 0 |
| 158.3 | 15 |
Sheet1
Sheet2
Sheet3
X-bar Chart
140
145
150
155
160
12345678910
Days
X-bar
Chart4
| 149.3 |
| 149.3 |
| 151 |
| 152.3 |
| 150 |
| 151.3 |
| 152.6 |
| 149.3 |
| 152 |
| 152 |
Sheet1
| X-bar | R |
| 149.3 | 4 |
| 149.3 | 4 |
| 151 | 2 |
| 152.3 | 6 |
| 150 | 4 |
| 151.3 | 7 |
| 152.6 | 8 |
| 149.3 | 4 |
| 152 | 6 |
| 152 | 5 |
Sheet1
Sheet2
Sheet3
R Chart
0
2
4
6
8
10
12345678910
Days
Range
Chart5
| 4 |
| 4 |
| 2 |
| 6 |
| 4 |
| 7 |
| 8 |
| 4 |
| 6 |
| 5 |
Sheet1
| X-bar | R |
| 149.3 | 4 |
| 149.3 | 4 |
| 151 | 2 |
| 152.3 | 6 |
| 150 | 4 |
| 151.3 | 7 |
| 152.6 | 8 |
| 149.3 | 4 |
| 152 | 6 |
| 152 | 5 |
