Reliability Engineering - 123 - 07 10 2013

‫وما توفيقي إال باهلل‬
Reliability Engineering Best Practice
Quality Improvement
Dr. Attia Gomaa

Industrial Engineering Professor & Consultant
Engineering and Science Services - American University in Cairo
TQM 2013
Reliability Engineering Best Practice
:Contents Zero
Failures
• Quality & Reliability Overview
• Reliability Analysis
• RAM Analysis
• Quality & Reliability Tools
• Failure Analysis Tools
• Quality & Reliability Improvements
• Product Value Analysis
• Case Studies
TQM Reliability Engineering Best Practice 2

Rules of the Course
This is an Open Discussion Course:
– Share Knowledge ‫– تبادل المعرفة‬

– Share Experiences Team Approach
‫– تبادل الخبرات‬
– Share Best Practices ‫– تبادل التجارب الجيدة‬
– Share Questions ‫– تبادل األسئلة‬
Remember … You can’t do it alone!
Let us be a teamwork.
Quality & Reliability Overview
Systematic
Main Targets: Approach
Failure Mistake Error

What is Quality?
The degree to which a product or service meets specified requirements.
The totality of features and characteristics of a product or service that bear on its
ability to satisfy stated or implied needs. ISO 8402 / ISO 9000
Dimensions of Quality
Fit to Use
Fit to Customer
Fit to Standard
Fit to Price
Doing Right Thing Right First Time and every Time

Fit to Price Fit to Standard Fit to Customer Fit to Use
Cost Saving Reliability Flexibility Efficiency
For example; Product: Car
• Price • Brand • Manual / Auto. • Performance
• • • •
• • • •
• • • •
• • • •
• • • •
• • • •
For example; Product: Car
• Price • Brand • Manual / Auto. • Performance
• Energy • Failure-free • Options • Duty
• Spare Parts • Error prevention • Size •
• • Fault detection • Colors •
• • Safety System • Shape •
• • Fully Certified • •
• • Longest Warranty • •
Example:
Product: ---

Dimensions of Quality  Customer Satisfaction
Efficiency Flexibility Reliability Cost Saving

Mistake-free
Failure-free
Most Energy Efficient
High Performance Maintenance-free
More Safe
What-if analysis
Longest life
Fully Certified
More Options Longest Warranty
Robust Design
Intelligence Energy Cost
Maintenance Cost
Cost Rates
"Quality” is closely linked to “Reliability”

How to select a reliable product?
Data Show Projector Product
- Meeting Room Applications
- Education
- Seminar
• High resolution Customer
• Brightness Needs
• Easy to use
• Compact body
What is • Low price
Reliability ?
4 ～ 6 hrs/day Hours of
20 ～ 30 hrs/week operation
Failure
4000 to 5000 LE Price Range
Specifications
TQM
?
Reliability Engineering Best Practice 10
Reliability – Critical to quality
Generally defined as;
the ability of a product to perform as expected over time
Probability that no failure will occur in a given time interval
Probability of failure-free operation for a specific time period
The probability a unit will perform its intended function for a specific
period of time (MIL-STD-721B).
Reliability is a dynamic aspect of Quality.
R(t)
R (t )  P(t | t  T ) , T 0
1
0
t
Reliability is defined as the probability that an item will perform a required

function without failure under stated conditions for a specified period of
time.
A statement of reliability has four key components:
• Probability  for example; a reliability goal of .9995
• Required function  For example, a pump’s required function

might be moving at least 20 gallons per minute.
• Stated conditions  working condition, environmental conditions,

maintenance conditions, etc.
• Specified period of time  For example, a pump might be designed to

function for 10,000 hours.

Reliability Statement
Reliability
Application Time
Based Based
Operating Probability
Based Based

Example:
Reliability Statement
Time Probability Operating Application
Conditions
Product: ---

Example:
Product:
Reliability Policy Quality Policy
Comments:

A reliable product
It is robust and mistake-free
One that does what the customer wants, when the customer wants to do
it
Desig
n for R
eliabil
i ty
Reliability cannot be tested into a product
It must be designed and manufactured into it
Testing only indicates how much reliability is in the product
Reliability Dimensions
4 Rights
Quality Performance Safety Time
Types of Reliability:
• Inherent reliability – predicted by product design
• Achieved reliability – observed during use
Redundancy:
The use of backup components to increase reliability
What affects Reliability?
5 Factors
Design Maintenance
Test Operation
Installation

Reliability Trade-off
Cost
Customer Price Limit
Optimal
Reliability
Reliability
Failure:
Device does not operate according to its specification.
Failure: Situation in which a product, part, or system does not perform

as intended
Types of Failures:
• Functional failure – failure that occurs at the start of product life due to
manufacturing or material detects.
• Reliability failure – failure after some period of use.
Probability of Failure
Failure in Failure in Failure in Failure in

Safety Durability Serviceability Compatibility
Failure due to Failure due to

Human Error Natural Causes

Quality Engineering
Systematic Approach To Achieve Certain Quality at Low Cost
Systematic Approach For Long Term Continuous Improvements
Quality Improvement
TQM Quality Engineering Best Practice 21

Quality Management System
Whole-System Process Approach to Quality
Quality
Policy
Top
Planning
Middle
Control
Operational
Inspection Site
Feedback

Quality Management System
Whole-System Process Approach to Quality
Quality
Quality Policy
Assurance Planning
Quality Planning
Control
Quality Control
QA/QC Form
Inspection
Feedback ITP Checklist
(QA/QC) Quality Assurance/Quality Control (Inspection and Test Plan) checklist

Quality Assurance
Quality Planning
Quality Control
Inspection

Reliability Engineering
Systematic Approach To Improve Products Reliability at Low Cost
Systematic Approach For Long Term Continuous Improvements
Quality Improvement

Quality Improvement
Reliability Improvement
Design for Rel.
Reliability
Analysis

Reliability Engineering
... -> Defects -> Faults -> Errors -> Failures ->...
Reliability Availability Safety Security
Fault (Defect) Fault Fault (Error) Fault (Failure)

Avoidance Removal Tolerance Prediction

Purpose of the Reliability Group
Determine the weaknesses in a design

AND correct them
before the device goes to the field
Areas Covered by Reliability

• Electrical
• Mechanical
• Software
• System, etc.

Reliability Importance
Warranty costs measured in million dollars for several large

.American manufacturers in 2006 and 2005
TQM )www.warrantyweek.com
Reliability (
Engineering Best Practice 29
Long Term Continuous Improvements
Global Leaders in their fields
Dr. Joseph M. Juran W. Edwards Deming

Father of Quality Father of QC
(1904 – 2008) (1900 – 1994)

Problem-solving process
Deming Wheel, 1950
Plan-Do-Check-Act (PDCA) Circle

Six Sigma Improvement Methodology – D-M-A-I-C, 1987 Motorola
Define
Control
Measure
ACT PLAN
CHECK DO
Improve
Analyze

Basics:
Probability and Statistics
• Mean, Variance, Standard Deviation
Production & Design Techniques
• Production Technology • Range, Mode, Histogram, Probability
• Product Design • Probability density function (PDF)
• Parts Design • Probability Distributions
• Material Selection & Control • Statistical process control (SPC)
• Process Design • Sample size, Control Charts
• Technical Standards • Out-of-control, Process Capability
• Stress-strength Analysis • Sigma Level, 3-Sigma, 6-Sigma
• Stress-life distributions • Confidence level, Confidence interval
• Robust Design • Operating characteristic (OC) curve
• Re-Engineering • Hypothesis test, Degrees of freedom
• Human Factors, etc. • Type I error, Type II error

• Experiment Design, ANOVA,
• F distribution, Chi Square Distribution, etc.

Sheet #1:
1. What is Quality?
2. What is Reliability?
3. Define and describe quality and reliability and how they relate to each other.
4. Define and describe the quality dimensions to select a reliable product.
5. Demonstrate how reliability engineering techniques and methods improve programs,
processes, products, and services. (Synthesis).
6. Demonstrate how reliability professionals can apply their techniques and interact
effectively with marketing, safety and product liability, engineering, manufacturing,
logistics, etc. (Analysis).
7. Integrate reliability engineering techniques with other development activities (e.g.,
concurrent engineering). (Synthesis).
8. Use liability and consequence limitation objectives to determine reliability acceptance
criteria, and identify development and test methods and verify and validate these criteria.
(Application)
9. Determine the impact of failures in terms of service and cost (both tangible and
intangible) throughout a product’s life-cycle. (Analysis)

Reliability Analysis

Measure of Reliability
• Failure Rate  = number of failures within a given time period.
• MTBF (mean time between failure) = the amount of time between failure
of a repairable product = 1/.
• MTTF (mean time to failure) = the amount of time between failure of a

non-repairable (i.e., must be replace) product = 1/.
Reliability R(t)  f (MTBF)
R(t) = exp(-λt) ≈ 1 – λt
Failure Rate =  = 1 / MTBF
[when λt is less than 0.001]
R(t) = e -t
In this formula, the failure rate, λ, is assumed to be constant with the age
of the component.

Measure of Reliability
• Probability of reliability = 1 – Probability of failure
• Average total number of failures (t)
• Failure Rate (t)

Number of failures per time unit,
= number of failures within a given time period.
= (number of failures within the testing time period)  ( total testing time period for
all components in the sample)
• MTBF (mean time between failure)

= the amount of time between failure of a repairable product
1
•
MTBF 
MTTF (mean time to failure)  (t )
= the amount of time between failure of a non-repairable (i.e., must be replace)
product
1
MTTF 
 (t )
Description of Parameters
t1 t2 t3 t4 t5 .......time
 Failure Event of an Item r =Number of Failures
Repairable Items:
t
i
i
Mean Time Between Failures =
r
Non Repairable Items:
t
i
i
Mean Time To Failure =
r
Example : MTBF
Item has a MTBF of 4380 hours.
• What is the failure rate?
• What is the failure rate per million hours?
• What is the Sigma Level?
• What are your recommendations for 6-sigma?

Example : MTBF
Item has a MTBF of 4380 hours.
λ = 1/MTBF
λ = 1/4380 failures per hour
λ = 0.000228 failures per hour
λ = 228 failures per million hours
Sigma Level: 5
Design for 6-Sigma?

Example : MTBF
10 power supplies are put on test, to be terminated after
each has completed 1000 hours of operation. Two power
supplies fail, one at 420 hours and the other at 665 hours.

• What is the Sigma Level?

Example : MTBF
10 power supplies are put on test, to be terminated after
each has completed 1000 hours of operation. Two power
supplies fail, one at 420 hours and the other at 665 hours.

Eight units completed 1000 hours.
Total test time = 8(1000) + 420 + 665 = 9085 hours
λ = number of failures/total test time

λ = 2/9085 = 0.000220 failures per hour
λ = 220 failures per million hours
Sigma level: 5
Estimated MTTF = 1/0.00022 = 4545
Design for 6-Sigma?
Example : MTBF
10 Lamps are put on test, to be terminated after each has
completed 10000 hours of operation.
3000 * 1 items
4200 * 2 items
6650 * 3 items
8000 * 2 items
9000 * 1 items
9800 * 1 items
• What is the failure rate? 0.000151 f/hr

• Estimated MTTF = 1/0.000151 = 6617  > 6600 hr
• What is the failure rate per million hours? 151 F/M.hr
• What is the Sigma Level? 5.25

Example:
10 components were tested. The components (not repairable) failed as
follows:
 Component 1 failed after 75 ours
 Component 2 failed after 125 hours

• What are your recommendations for 6-sigma
• What is the probability that it will reliable during the first
200 hours of operations?
• What are your recommendations for Reliability > 80%
during the first 200 hours ?

Example:
10 components were tested. The components (not repairable) failed as
follows:
 Component 1 failed after 75 ours
Solution: 325 525

130
Five failures, operating time = 3805 hours 125 5 x 525
75
Failure Rate =
 = 5 / 3805 = 0.001314
MTTF = 3805 /5 = 761 Hours

Design for 6-Sigma?
What is the probability that it will reliable during the first 200 hours of operations?
Rt = e – λt e – (200) (0.001314) = % 76.89 =
What are your recommendations for Reliability > 80% during the
first 200 hours ?

MTBF = MTTF ??
An assumption:
Failed item returns to “As Good As New” status after
repair or renewal.
note: Time To Repair is not considered.
UP TIME
DOWN

How is  measured?
In order to determine the failure rate, we collect a random sample of
components and determine the time of failure for each component and
determine the estimated failure rate using the following equation:
 = (number of failures within the testing time period) 
Example:
( total testing time period for all components in the sample)
• 10 components were tested. The
components (not repairable) failed as
follows: 325 525
 Component 1 failed after 75 ours 130
125 5 x 525
 Component 2 failed after 125 hours 75
Solution:
Five failures, operating time = 3805 hr
Failure Rate =
 = 5 / 3805 = 0.001314

Example:
Total working hours = 100,000 hours Number of failures = 20
MTTF = 100,000 / 20 = 5,000
• Failure Rate λ = 20/100,000 = .0002 per hour
• What is the probability that it will reliable during the first 200
hours of operations?
Rt = e – λt e – (200) (0.0002) = % 96.08 =
R >= 99% = MTTF  Select the best item

Reliability Growth Example
70000
60000
50000
MTBF (Hours)
40000
30000
20000
10000
0
1997 1998 1999 2000
Year of Report
Quality Reliability Analysis 50

Failure Rate Curve: Bathtub Curve
Infant
Middle Late
mortality
1
Failure Rate
Early Increasing
Constant
Failure Failure Rate
Failure Rate (Wear Out )
0 Time t
Infant
Middle Late
mortality
1
Failure Rate
Early Increasing
Constant
Failure Failure Rate
Failure Rate (Wear Out )
0 Time t
Example:
14 random sample of light bulbs Hours of Operation
Before Failure Sample
35 1
90 2
100 3
150 4
Based on this information, the following: 700 5
1010 6
- Failure rate 1040 7
- Failure rate curve
1190 8
- Estimated MTTF  (Life Time)
- Failure rate per million hours 1200 9
- Sigma Level 1250 10
- Recommendations for 6-sigma 1270 11
1280 12
1300 13
1450 14
Hours of Operation
What is the failure rate for the testing Before Failure Sample
period of 450 hours? 35 1
What is the MTTF? 90 2
100 3
450 hours 150 4
700 5
900 hours 1010 6
1040 7
What is the failure rate for the testing
period of 900 hours? 1190 8
What is the MTTF? 1200 9
1250 10
1270 11
1280 12
1350 hours 1300 13
What is the failure rate for the testing 1450 14
period of 1350 hours?
What is the MTTF?
What is the failure rate for the
testing period of 450 hours?
Hours of Operation
What is the MTTF?
Samples 1, 2, 3, & 4 failed within 35 1
450 hours. 90 2
100 3
The other 10 survived.
150 4
= (number of failures within the testing 700 5
time period)  ( total testing time period for 1010 6
1040 7
1190 8
 = (4)  (35+90+100+150+10*450)
1200 9
=0.000821 failures/hour.
1250 10
1270 11
MTTF = 1/ = 1/0.000821 1280 12
= 1,218 hours. 1300 13
1450 14
Hours of Operation
What is the MTTF?
Samples 1, 2, 3, 4 & 5 failed within 35 1
900 hours. 90 2
100 3
The other 9 survived.
150 4
1040 7
1190 8
 = (5)  (35+90+100+150+700+9*900)
1200 9
=0.000545 failures/hour.
1250 10
1270 11
MTTF = 1/ = 1/0.000545
1280 12
= 1,835 hours.
1300 13
1450 14
Hours of Operation
What is the MTTF?
Samples 1, to 13 failed 35 1
90 2
within 1350 hours.
100 3
Only sample 14 survived. 150 4
1040 7
 = (13)  (35+90+100+150+700+ 1190 8
…1300+ 1*1350) 1200 9
=0.001087 failures/hour. 1250 10

1270 11
MTTF = 1/ = 1/0.001087 1280 12
= 920 hours. 1300 13
1450 14
Mean Time To Failure Rate  Period: Testing
Failure MTTF failures/hr Bath-Tub Time
(hours) Function
Infant
1,218 0.000821 mortality 450 hours
1,835 0.000545 Middle 900 hours
920 0.001087 Late 1,350 hours
Why did we use the MTTF and not the MTBF?

For example: Bearing / Gear Life Cycle
1
Failure Rate
1 3
Reliability ‘by the book’=R(t)= t
MTBF
e
2
Early Constant Wear Out

Life Failure Rate
0 Time t
Distributions:
Weibull
Β <1
Failure Rate
Exponential Β >1
Normal
Β =1 Lognormal
Wear-in Random Wear-Out
0 Time t
Product Failure Rate Over Time:
Early Life Failures and the Bathtub Curve
Inherent Failure Rate Latent Defects
Early Life Wear-out

Failures
Monthly Steady
Failure State
Rate
Time

Failure Patterns
Pattern A:
is the well known bathtub curve it begins with high
incidence of failure (known as infant mortality) followed by
constant failure then by wear out zone .
Early Life Failure

(Infant Mortality) Wearout
Useful Life (End of Life)
Failure rate
(Constant Failure Rate)
Time

FAILURE CHARACTERISTICS CURVES

For example: Bearing / Gear Life Cycle
1
Failure Rate
1 3
MTBF
e
2
Early Constant Wear Out

Life Failure Rate
0 Time t
Distributions:
Weibull
Β <1
Failure Rate
Exponential Β >1
Normal
Β =1 Lognormal
0 Time t
Case Study: Bearing Failure Rate
1
MTBF
e

Critical Systems / Equipment
• Safety Effect • Critical Systems

• Process Effect
• Critical Equipments
• Financial Effect
• Critical Components
• Critical Parts
• Critical Failure
For Critical Failures

MTBCF
Mean Time Between Critical Failures

MTTCF
Mean Time To Critical Failure

Critical Failures
Moving towards System Design
A System Failure resulting in (temporary or permanent) Mission

Termination.
X
COMPUTER A simple
configuration
X SUBSYSTEM of parallel hot
COMPUTER Redundancy.
A Failure: any computer failure
A Critical Failure: two computers failed
68
Mean Time to Failure: MTTF
 
MTTF   tf (t )dt   R (t )dt
0 0
1 n
MTTF   ti
1 n i 1
2
2 is better than 1?
R(
t)
0
Time t
Basic Calculations
Suppose n0 identical units are subjected to a test. During the
interval (t, t+∆t), we observed nf(t) failed components. Let ns(t)

be the surviving components at time t, then the MTTF, failure
density, hazard rate, and reliability at time t are:
n0
t i
n f (t )
MTTF  i 1
, fˆ (t ) 
n0 n0 t
ˆ (t )  n f (t ) , ˆR(t )  P (T  t )  ns (t )
r
ns (t )t n0
Example:
200 light bulbs were tested and the failures in 1000-hour intervals are
Failures in the Time Interval (Hours)

interval
100 0-1000
40 1001-2000
20 2001-3000
15 3001-4000
10 4001-5000
8 5001-6000
7 6001-7000
200 Total

Calculations
Failures Time Interval Time Failure Density Hazard rate
in the (Hours) Interval f ( t ) x 104 h (t ) x 104
interval
100 100
100 0-1000 0-1000 200  103
 5.0
200  103
 5.0
40 1001-2000
20 2001-3000
15 3001-4000
40 40
10 4001-5000 1001-2000 200  103
 2.0  4.0
100  103
8 5001-6000
7 6001-7000 20 20
2001-3000  1.0
60  103
 3.33
200 Total 200  103
…… …….. ……
7 7
6001-7000 200  103
 0.35
7  103
 10

Failure Density vs. Time
×10-4
1 2 3 4 5 6 7 x 103
Time in hours

Hazard Rate vs. Time
10-4×
× 7 6 5 4 3 2 1
103
Time in Hours

Calculations
Failures Time Interval Time Interval Reliability R (t )

in the (Hours)
0-1000 200/200=1.0
interval
100 0-1000
40 1001-2000 1001-2000 100/200=0.5
20 2001-3000
15 3001-4000 2001-3000 60/200=0.33
10 4001-5000
8 5001-6000
7 6001-7000 …… ……
200 Total
6001-7000 0.35/10=.035

Reliability vs. Time
x 7 6 5 4 3 2 1
103
Time in hours

Example
In a water pump station, during a period of 100,000 working hours,
the bearing failed 40 times. In this case,
MTBF = 100,000 / 40 = 2500 hours

The failure rate may be calculated as follows :
λ = 40/100,000 = 0.0004
For a 2 working hours, t = 2 , the reliability then becomes –

R(t) = exp(-λt) ≈ 1 – λt [when λt is less than 0.001]
So R(t) = 1 – 0.0004 x 2 = 1 – 0.0008 = 0.9992
or, expressed as a percentage,

R%(t) = 0.9992 x 100% = 99.92%
Mechanical Seal

MTBF = 100,000 / 20 = 5,000
• What is the probability that it will reliable during the first 200 hours of
operations?
Rt = e – λt
e – (200) (0.0002) =
% 96.08 =
Frequency, Reliability and Probability
Failure Reliability Failure Rate Component

Probability R=e(-mt) 
P=1-R (faults/year)
0.45 0.55 0.6 Control
Valve
0.25 0.75 0.29 Controller
p = 1- e-t
where p is the annual probability of occurrence,

 is the annual frequency and t is time period
(i.e., 1 year).

Example:
The chart below shows operating time and breakdown time of a machine.
20.2 2.5 6.1 7.1 24.4 4.2 35.3 1.8 46.7
Operating time Down time

a) Determine the MTBF.
b) What is the system reliability for a mission time of 20 hours?
Solution:
Total operating time = 20.2 + 6.1 + 24.4 + 4.2 + 35.3 + 46.7 = 136.9 hours
 = 4 / 136.9 = 0.02922
Therefore;
 = MTBF = 1/  = 34.22 hours
R = e-t t = 20 hours
R= e-(0.02922)(20) = 55.74%

Reliability Modeling / Redundancy
Reliability is the probability that no failure will occur in a given time interval
Series
Item 1 Item 2 Item 3 Item n
Active
Parallel Item 1
Item 2 Redundancy can get you

more reliability but usually
Standby at a cost, size, or weight
Parallel Item 1 penalty
Item 2

System Reliability
• Series System: A B
RS = RA*RB
A
• Parallel System:
RS = 1-(1-RA)*(1-RB) B
• Backup System:
B
RS = RA + RB(1-RA)
A

System Reliability Analysis
Series Systems
1 2 n
RS = R1 R2 ... Rn
Parallel Systems
1
)Redundancy(
2
RS = 1 - (1 - R1) (1 - R2)... (1 - Rn)

Series or chain structure: Rs = R1 * R2 * R3 * …..
Rs =
R1 R2 R3
0.95*0.85*0.90
0.95 0.85 0.90
= 0.7 = 70%
Parallel structure: Rs = 1 –(1-R1)* (1-R2)* (1-R3) * ..…
R1 = 0.95 Rs =
1- (1-0.95) *
R2 = 0.85
(1-0.85) * (1-0.90)
R3 = 0.90 = 0.99925 = 99.925%

Rs =
R1 R2 R3
0.95*0.85*0.90
0.95 0.85 0.90
= 0.7 = 70%
Rs >= 80?
Rs >= 85?
Rs >= 90?

R2A
0.85
R1 R3
0.95 R2B 0.90
0.85
Rs = 83%
Rs >= 80?
Rs >= 85?
Rs >= 90?
R2A R3A
0.85 0.90
R1
0.95 R2B R3B
0.85 0.90
Rs = 91.9
Rs >= 80?
Rs >= 85?
Rs >= 90?
Lamp 1 Lamp 2
.90 .80 .90 x .80 = .72
.80 Lamp 2 (backup)
.90 .90 + (1-.90)*.80 = .98

Lamp 1

.70 Lamp 3 (backup for Lamp 2)
.80 Lamp 2 (backup for Lamp1)
1 – P(all fail)
.90 1-[(1-.90)*(1-.80)*(1-.70)] = .994
Lamp 1

Determine the reliability of the system shown
.90 .92
.98 .90 .95
The system can be reduced to a series of three components
.98 .90+.90(1-.90) .95+.92(1-.95)
.98 x .99 x .996 = .966

Series-Parallel System
C
RA RB RD
RC
A B D
C
RC
 Convert to equivalent series system
RA RB RD
A B C’ D
RC’ = 1 – (1-RC)(1-RC)
Reliability Examples:
RA = 0.72, RB = 0.96, RC = 0.87, RD = 0.91
A
C D
A B
C D
A
RT = 0.898

Reliability Examples:
An electrical part has 2 components set up in parallel. The

component reliabilities are .95 and .98. What is the reliability of
the part?
Rs = 1-(1-.98)(1-.95) = .999
An electrical part has 3 components set up in series. The

component reliabilities are .95, .98 and .90. The first and third
components have backups with reliabilities of .92 and .89,
respectively.
What is the reliability of the part?

Reliability Example
Motor Pump
R1 R2 R3
99. 80. 90. Rs
Reliability of the process is
Rs = R1 x R2 x R3 = .99 x .80 x .90 = .713 or 71.3%

Question :
Four subsystems have the following reliabilities:
RA = RB = .90 and RC = RD = .95.
The four subsystems are connected as shown.
Discuss the system reliability.

Question :
RA = RB = .90 and RC = RD = .95.
Since A & B are parallel, R(AB) = 1 – { (1-Ra) * (1-Rb)} = 0.99 = 99 %

Since C & D are parallel, R(CD) = 1 – { (1-Rc) * (1-Rd)} = 0.9975 = 99.75%
Since AB & CD are series, R(ABCD) = R(AB)*R(CD) = 0.987525 = 98.75%
System reliability (which means the probability of zero failure) = 98.75%

This means,
How to improve the system relibility?

• Improve the maintenance program.
• Improve the mistake proofing (Poka-Yoke)
• Improve the manpower skills through advanced training plan.

Question :

RA = RB = .90 and RC = RD = .95.
The four subsystems are connected as shown
RA=RB = 15 $/unit RC=RD= 20$/unit
Unit Cost = 95 $/unit Price= 100 $/unit
Competitors 
Rs = 99.2% Price= 110 $/unit
After  AB+1
Rs = 99.6% Cost = 110 $/unit
Profit = 0
Value Added = Price – Material Cost = 110 –(15*3+20*2) = 25
Value Added Ratio = 25/110 = 22%
System Reliability:
What is the reliability of this system?
R = [.95 + .92(1-.95)] • [.98] • [.90 + .89(1-.90)]

= (.996) (.98) (.989) = .965 = 96.5%

Case Study
Motor Pump
MTTF (hour/failure)
Pump Coupling Motor
3571 6667 9090
1- Calculate the failure rate for each part
2- MTTF for the system
2- After 1000 working hours, Calculate the reliability

Motor Pump
(t) = Failure Rate (failure/hour)

MTTF = Mean Time To Failure =1/
Pump Coupling Motor
3571 6667 9090 MTTF (hr)

0.00028 0.00015 0.00011 Failure rate (f/hr)
MTTFs = 1 / (1 + 2 + 3 ) = 1/(0.00054) = 1852 hour

Motor Pump
(t) = Failure Rate (failure/hour)

MTTF = Mean Time To Failure =1/
R(t) = Reliability = e ^- (.t)
Pump Coupling Motor
0.00028 0.00015 0.00011 Failure rate (f/hr)
3571 6667 9090 MTTF (hr)
e ^ -(0.28) e ^ -(0.15) e ^ -(0.11) Reliability after 1000 hr
= 0.75578 = 0.86071 = 0.89583
MTTFs = 1 / (1 + 2 + 3 ) = 1/(0.00054) = 1852 hour
Rs = R1 * R2 * R3 = 0.5827 = 58.27%
Input A B C Output
R = RA x RB x RC
 ( 1  2  3 ... n ) t
= Rs (t) e
Example
• Systems expected to operate for 1000 hours. It consists of 4 subsystems
in series, MTBFA = 6000 hours, MTBFB = 4500 hours, MTBFC = 10,500
hours, MTBFD = 3200 hours. Determine overall reliability.
 A = 1 /MTBFA = 1/6000 = 0.000167

 B = 1/MTBFB = 1/4500 = 0.000222
 C = 1/MTBFC = 1/10500 = 0.000095
 D = 1/MTBFD = 1/3200 = 0.000313
 Therefore; R = e-(0.000797)(1000) = 0.4507

Parallel Network A
• Consider two units A and B in B
Example parallel. The systems fails only
when A and B failed. C
Fs(t) = Fa(t) Fb(t)
= [1-Ra(t)][1-Rb(t)]
A
= 1-Ra(t) Rb(t) + Ra(t) Rb(t)
B Rs(t) = 1- Fs(t)
= Ra(t) + Rb(t) – Ra(t) Rb(t)
• If A and B are constant failure rate units, then:
a t
• Ra(t) =
e Rb(t) =
e  b t

1 1 1
And Rs(t) =  Rs (t )dt    s = MTBF
0
a b a  b
Component Reliability:
ABC Company produces 3000 computer chips per day. Three
hundred chips are tested for 500 operating hours. Two chips
failed after 50 hours; two failed after 100 hours; one failed at 300
hours; one failed at 400 hours.
What is the MTBF?
Number of failures = 6
Total operating time = 2(50) + 2(100)
+ 1(300) + 1(400) + 294(500) = 148,000 hours
MTBF =148,000/6= 24666.67 hours=1027.78 days=2.85 years
What is the probability that a chip produced will fail within 2
years?
MTBF = 2.85 years T = 2 years
R(T) = e-(2/2.85) = e-(.7018) = .4957
F(T) = 1 - .4957 = .5043
Reliability Tables
T/MTBF e -)T/MTBF( T/MTBF e -)T/MTBF( T/MTBF e -)T/MTBF( T/MTBF e -)T/MTBF(
0.05 0.95123 2.05 0.12873 4.05 0.01742 6.05 0.00236

0.10 0.90484 2.10 0.12246 4.10 0.01657 6.10 0.00224
0.15 0.86071 2.15 0.11648 4.15 0.01576 6.15 0.00213
0.20 0.81873 2.20 0.11080 4.20 0.01500 6.20 0.00203
0.25 0.77880 2.25 0.10540 4.25 0.01426 6.25 0.00193
0.30 0.74082 2.30 0.10026 4.30 0.01357 6.30 0.00184
0.35 0.70469 2.35 0.09537 4.35 0.01291 6.35 0.00175
0.40 0.67032 2.40 0.09072 4.40 0.01228 6.40 0.00166
0.45 0.63763 .2 45 0.08629 4.45 0.01168 6.45 0.00158
0.50 0.60653 2.50 0.08208 4.50 0.01111 6.50 0.00150
0.55 0.57695 2.55 0.07808 4.55 0.01057 6.55 0.00143
0.60 0.54881 2.60 0.07427 4.60 0.01005 6.60 0.00136
0.65 0.52205 2.65 0.07065 4.65 0.00956 6.65 0.00129
0.70 0 49659. 2.70 0.06721 4.70 0.00910 6.70 0.00123
0.75 0.47237 2.75 0.06393 4.75 0.00865 6.75 0.00117
0.80 0.44933 2.80 0.06081 4.80 0.00823 6.80 0.00111
0.85 0.42741 2.85 0.05784 4.85 0.00783 6.85 0.00106
0.90 0.40657 2.90 0.05502 4.90 0.00745 6.90 0.00101
0.95 0.38674 2.95 0.05234 4.95 0.00708 6.95 0.00096
1.00 0.36788 3.00 0.04979 5.00 0.00674 7.00 0.00091

Reliability Tables
T/MTBF e- )T/MTBF( T/MTBF e- )T/MTBF( T/MTBF e- )T/MTBF( T/MTBF e- )T/MTBF(

1.05 0.34994 3.05 0.04736 5.05 0.00641 7.05 0.00087
1.10 0.33287 3.10 0.04505 5.10 0.00610 7.10 0.00083
1.15 0.31664 3.15 0.04285 5.15 0.00580 .7 15 0.00078
1.20 0.30119 3.20 0.04076 5.20 0.00552 7.20 0.00075
1.25 0.28650 3.25 0.03877 5.25 0.00525 7.25 0.00071
1.30 0.27253 3.30 0.03688 5.30 0.00499 7.30 0.00068
1.35 0.25924 3.35 0.03508 5.35 0.00475 7.35 0.00064
1.40 0.24660 3.40 0.03337 5.40 0 00452. 7.40 0.00061
1.45 0.23457 3.45 0.03175 5.45 0.00430 7.45 0.00058
1.50 0.22313 3.50 0.03020 5.50 0.00409 7.50 0.00055
1.55 0.21225 3.55 0.02872 5.55 0.00389 7.55 0.00053
1.60 0.20190 3.60 0.02732 5.60 0.00370 7.60 0.00050
1.65 0.19205 3.65 0.02599 5.65 0.00352 7.65 0.00048
1.70 0.18268 3.70 0.02472 5.70 0.00335 7.70 0.00045
1.75 0.17377 3.75 0.02352 5.75 0.00318 7.75 0.00043
1.80 0.16530 3.80 0.02237 5.80 0.00303 7.80 0.00041
1.85 0.15724 3.85 0.02128 5.85 0.00288 7.85 0.00039
1.90 0.14957 3 90. 0.02024 5.90 0.00274 7.90 0.00037
1.95 0.14227 3.95 0.01925 5.95 0.00261 7.95 0.00035
2.00 0.13534 4.00 0.01832 6.00 0.00248 8.00 0.00034

Transistor Amplifier Failure Rate
+V
R1 RC
T1
CO
RS CB RL
AC
R2 RE
CE
0 V
AC RS CB R1 R2 T1 RC RE CE CO RL PS

System Reliability
Static Reliability = R = (95.10)*(98.61)*(93.50)*(99.10) = 86.89
Fuel Delivery Engine Transmission Steering

95.10 98.61 93.50 99.10

When a number of parts are in series, and all must work
for the entire network to be operational:
Network MTBF = 1/(1/MTBF1 + 1/MTBF2 + … 1/MTBFn)
Example:
Three parts are in series. MTBF1=10000, MTBF2=20000, MTBF3=13000.
MTBF MTBF MTBF

=10000 =20000 =13000
Network MTBF = 4407

When a number of parts are in parallel, and at least one must work for the
entire network to be operational:
Network MTBF = MTBF1 x MTBF2 x … MTBFn
Example:
Three parts are in parallel. Each has an MTBF of 10000.
Network MTBF = 1.0E+12 (which is 1000000000000)
MTBF=10000
MTBF=10000
MTBF=10000

Production Line - 20 Similar Furnaces
Average downtime = 120 hour /year

Total Time = 12 month * 22 day * 16 hour = 4224 hour/year
Average equipment availability

= Available time / Total time
= (Total time – total down time) / Total time
A = (4224 – 120) / 4224 = 0.97 %
1 2 N=20
RS = R1 R2 ... Rn=20
= 0.97^20 =

Overall System Reliability
100 –
n=1
Reliability of the system (percent)
80 – n=1
0
60 –
n=
40 – n 50
=
10
n 0
=
20 – 20
n= 0
n
=
30
0
40
0
| | | | | | | | |
0 –100 99 98 97 96
Average reliability of each component (percent)

Horrible !
Do not bother me
with these statistics.
I am the Professor
and I do not trust
any statistics not
performed by me.
Dr. Attia
Gomaa
Let us try to make it easy

( ee‫اء هللا‬ee‫نش‬e‫) إ‬
Operating characteristic (OC) curve for sampling plan.
λ0 = Expected failure rate λ1 = Maximum acceptable failure rate

α = Producer’s (type I) risk β = Consumer’s (type II) risk

Fixed-Time Test Plans
Ten units are on test.

The units are not replaced when they fail (non-replacement).
One unit fails at (t1=685 hours), and a second unit fails at (t2=1690) hours.
The test is ended at (t = 2500 hours) with no additional failures.
What is the total accumulated test time?
T = 685 + 1690 + (8)(2500) = 22,375 hours

Attribute test
A reliability estimate can be obtained from an attribute test. The test could be
described as testing (n units) and recording (r failures). The estimate of the
reliability of the unit for the conditions of test is
150 sensors used in the airbag system on an automobile are tested at a

force that should be detected. Two of the sensors failed to detect the force
and therefore are classified as failures.
What is the estimate of the reliability of the sensor for detecting that force?
n = 150 r=2
Rˆ = (150 – 2) / 150 = 0.987

Zero Failure Test
A point estimate of reliability can not be made if the test results in zero
failures. A lower confidence limit can be found for a zero-failure test.
The lower confidence limit on the reliability value RL for a test of n units
with zero failures at a confidence value of C is:
In the above equation (1 – C) is referred to as the significance of the test and

may be replaced by α .
One hundred and fifty sensors used in the airbag system on an automobile are
tested at a force that should be detected. None of the sensors failed to detect
the force, therefore no point estimate of the reliability of the sensor is possible.
What is the lower 90 percent confidence limit on the reliability of the sensor for
detecting the force?

This equation can be solved for the number of units necessary to test without
failure to show a given reliability at a given confidence level.
The compliance requirement is to demonstrate that the airbag sensor has a

minimum reliability of 0.98 with a confidence of 0.90.
How many sensors need to be tested if the test results in zero failures?

Question :

RA = RB = .90 and RC = RD = .95.
The four subsystems are connected as shown
RA=RB = 15 $/unit RC=RD= 20$/unit
Unit Cost = 95 $/unit Price= 100 $/unit
Reliability  Zero failure Test at confidence limit 90%

System Rel. R = 98.5
After  AB+1 Rs = 99.6%

Sample Size
A production lot of 200 units contains 1.5% defectives.

If a sample of 30 is tested,
what is the probability of finding two or more of the defective units?
The hyper-geometric distribution gives the probability of finding exactly (d)

defective units in a sample of n units.
d = Number of defective units in the sample
D = Number of defective units in the population
n = Number of units in the sample
N = Number of units in the population

Hazard & Reliability Functions
A total of 283 products are tested for 1100 hours. The number of failures in
each 100-hour block of time is recorded.

RAM Analysis

Excellent
RAM
Continuous Improvement
Good
Normal
Bad
Reliability
Safety
RAMS
Availability
Maintainability Resource Utilization

(Cost Reduction)
You can only control what you measure!
KPIs  Improve System Performance (RAMS+R)
Effectiveness Efficiency
Resource
Reliability Availability Maintainability Safety
Utilization
- Facility - Facility - MTTR - Overdue - Maintenance
Reliability Availability Ratio for Cost Variance
- MTTM critical items
- Equipment - Equipment - Process
Reliability Availability - PM Downtime - Critical Losses
Near-Miss
- MTBF - Item - CM Downtime Rate - Maintenance
Availability Cost Rate
- MTBCF - Process - Accident
- Downtime Rate - Performance
- CM/PM - Efficiency
- - Back-log - Lost Time
- CM/PM for (overdue) Ratio Injury (LTI) - Maintenance
critical Eq. - Labor
(High) - Utilization
-
Overall Indicators:
Overall Equipment Effectiveness (OEE)
Reliability is the Probability of
failure-free operation for a RAM
specific time period
Maintainability is the probability Reliability Maintainability

that a system that has failed will
be restored to operation
effectiveness within a given
time.
Availability is the probability

that a system will available to
work for a given period and Availability
under given operating
conditions.
Reliability R(t)  MTBF
Failure Rate =  = 1 / MTBF R(t) = exp(-λt) ≈ 1 – λt
R(t) = e-t [when λt is less than
0.001]
In this formula, the failure rate, λ, is assumed to be
constant with the age of the component.
Maintainability M(t)  MTTR
Maintenance Rate =  = 1 / MTTR
M(t) = 1 - e-t
Availability A(t)  MTBF & MTTR

Availability:
• Operational availability • Inherent availability

MTBF MTBF
AO  AO 
MTBM  MDT MTBF  MTTR
MTBM = mean time between maintenance

MTD = mean down time
MTBF = mean time between failures
MTTR = mean time to repair

Product Availability Calculation
0.99 0.999 0.9999 0.99999 0.999999

0.01
0.1
M TTR
10
100
1000
1,000 10,000 100,000 1,000,000
MTBF

Reliability/Maintainability Measures
Reliability R(t)
(Failure Rate) = 1 / MTBF
R(t) = e-t
Maintainability M(t)
(Maintenance Rate) = 1 / MTTR
M(t) = 1 - e-t

Equipment Availability & Reliability Analysis
Total (calendar) time (TT)
A Planned (target) time (PT) PM
R
Available time (AT) CM
U
Operating (used) time (OP) S/B Downtime
PM = Planned downtime A= Equipment Availability = AT/TT
CM = Corrective downtime = Available time / Total Time

= (Total Time – Downtime) / Total Time
S/B=Standby time
= (OP + S/B) / (OP+S/B+PM+CM)
OP = Operating time
R= Equipment Reliability = AT / PT
AT = Available Time = (OP+S/B) (OP + S/B + CM)
PT = Planned Time U= Equipment Utilization = OP / AT
Availability & Reliability
Total time (Total Working Hours) (T0) = 400
A
Planned (Target) time (T1) = 350 PM
R 50
Available time (T2)= 320 CM
30
Planned (PM) downtime = 50 Unplanned (CM) downtime = 30

Total down time = 50 + 30 = 80
A= Availability = T2/T0 = 320 / 400 = 80.0 %

R= Reliability = T2/T1 = 320 / 350 = 91.4%
Process Process Total Unplanne Planned Total
Reliabilit Availabilit downtim d downtime Working
y y e downtime Hours
91.4 % 80.0 % 80 30 50 400
Case Study:
As shown in the following figure, a water Station has 6 pumps:
P1
P2 P5 1 Running +
2 Running + 1 Standby
2 Standby
P3
P6
P4
Average Average Annual time

Unplanned downtime Planned downtime (Hour)
600 200 8700

Calculate and analyze the system reliability & availability.
ump availability & reliability
Total time (T0) = 8700/2 = 4350

A
Planned time (T1) = 4150 t1=200
R
Available time (T2)= 3550 t2=600
Planned downtime = t1=200 Unplanned downtime = t2=600

Total down time = 200 + 600 = 800
A= Availability = T2/T0 = 3550 / 4350 = 0.8161= 81.61%

R= Reliability = T2/T1 = 3550 / 4150 = 0.8554 = 85.54%
Pump Pump Total Average Average Average

Reliability Availability downtime Unplanned Planned Annual time
% % downtime downtime (Hour)
85.54 % 81.61 % 800 8700/2= 4350

600 200

System Availability analysis
0.8161
0.8161
0.8161
0.8161
0.8161
0.8161
System Availability = 97.87 %

System Reliability analysis
0.8554
0.8554
0.8554
0.8554
0.8554
0.8554
System Reliability = 97.87 %

Case Study:
As shown in the following figure, a water Station has 6 pumps;
(3 Running & 3 Standby)
P1
P5
P2
P3
P6
P4

Case Study
Motor Pump
6 Pumps  3 Running & 3 Standby

Annual running time = 4000 RH (for each pump)
Total running time during 3 years = 4000 * 3 = 12000 RH (for each pump)
Top 5 Failures
Out of Coupling Motor Pump Mech. Critical Item
balance Bearing Bearing Seal
1 2 4 6 10 Total Number of
Failures
2 6 12 18 40 Total Repair Time
(hour)
1200 1200 12000 12000 12000 Total running time
Brain Storming
Motor Pump Mech. Seal Critical Item
Bearing Bearing
4 6 10 Total Number of Failures
12 18 40 Total Repair Time (hour)
12000 12000 12000 Total running time
MTBF
Failure Rate = = 1 / MTBF
MTTR
Maintainability=  = 1 / MTTR
Availability =
MTBF / (MTBF + MTTR)
Reliability for t = 4000 RH
R(t) = e-t

Motor Pump Mech. Seal Critical Item
Bearing Bearing
4 6 10 Total Number of Failures
12 18 40 Total Repair Time (hour)
12000 12000 12000 Total running time
12000/4 12000/6 12000/10 MTBF
= 3000 = 2000 = 1200
10/4 = 3 18/6 = 3 40/10= 4 MTTR
(3000)/ (2000)/ (1200)/ Availability =
(3000+3) (2000+3) (1200+4) MTBF / (MTBF + MTTR)
= 99.999% = 99.85% =99.67%
R(t) = e-t

Brain Storming
Out of Coupling Critical Item
balance
1 2 Total Number of Failures
2 6 Total Repair Time (hour)
1200 1200 Total running time
MTBF
MTTR
Availability =
MTBF / (MTBF + MTTR)
R(t) = e-t

Quality & Reliability Tools

Define
Control
Measure
ACT PLAN
CHECK DO
Improve
Analyze

- Problem statement
Product / Process
-
Reference
- Teamwork
- Current Situation, etc.
Define
-
Control Incomplete Information
Measure
ACT PLAN Information Sheet
Information Collection
CHECK DO
- Improve Analyze -
-
-
-
-
-
-
-
-

Define Stage: Improve Stage:
• Product / Project / Process Define
Quality Improvement Tools:
• Problem Statement & Main Objectives • Quality Function Deployment
• Quality Dimensions / Reliability Dim. • Service Quality Analysis
• Process Plan / Method • Root Cause Analysis
• Quality Policy / Rel. Policy • Failure Mode Effect Analysis
• Quality Assurance vs. Quality Control • HAZOP Study
• QA/QC Check List / ITP Check list • Robust Design
Measure Stage: • Visual Control Tools:
• Information Sheets
(5 S, Poka-Yoke, Jidoka, etc.)
• Information Collection
• Feasibility Analysis
Analysis Stage:
• Value Analysis
• Seven QC tools
• Action Plan
• Normal Distribution
• Process Control Charts
• Process Capability Analysis
Control Stage:
• Sigma Level
• Key Performance Indicators (KPI)
Define Stage:
• Product / Project / Process Define
• Problem Statement & Main Objectives
• Quality Dimensions
• Process Plan
Quality Control Tools • Quality Policy
• Quality Assurance vs. Quality Control
• QA/QC Check List
Measure Stage:
• Information Sheets
• Information Collection
Analysis Stage:
• Seven QC tools
• Normal Distribution
• Process Control Charts
• Process Capability Analysis
• Sigma Level
oject / Process Define
Project: Gas Pipe Line – 20 inch – 120 km From Site X to Site Y
Process: Electric Arc Welding
Problem Statement:
Apply Six Sigma problem solving methodology to insure that the welding process
is achieving Six Sigma quality levels of less than 3.4 defects per million for all
critical welding parts.
Main Objectives:
Produce welding parts that meet the customer requirements and maximize the
resource utilization.
STANDARDS / CODES:
API 1104, Welding of pipelines and related facilities.
Quality Concept:
The degree to which a product or service meets specified requirements.
(Fit to use + Fit to standard + at minimum cost)
Quality Dimensions:
Quality dimensions are the parameters and specifications for customer
requirements & needs.
In arc welding for example;

Standards / Codes Qualitative Dimensions Quantitative Dimensions
API 1104 • Welding profile • Thickness
• Yield strength
• Toughness

Process Plan:
Process plan defines the key steps & stages which must be takes to achieve
the target process.
Process Map
SIPOC & Measurement Instruments
Requirements Requirements Requirements
Input Boundary Output Boundary
S I P O C
Suppliers Inputs Process Outputs Customers
Measures Measures
Measures

Process Map
Process Chart

Process Chart
52
plates 520 102
Raw sheets tank
Materials
Cutting Welding
Supplier
3 day 8.1 hr 52.0 hr
Welding
1581 rods
32 kg
Finishing
20.4 hr
Symbols:
101
Operations Product Description: tank
A welded water tank
open on one side.
Inspection
500 Painting
500 25.3 hr
Transportation
100
500
tank
Delay
Order
due date
Storage 21 day
(All dimensions are in mm) Final
Products

Process Map (SIPOC Chart)
• Production • Work Order • Products • Storage

Planning • Process Sheet • Right Quantity • Market
Department • Right Quality
• Drawings • Customer
• Raw Materials • Right Time
Process map
• Tools
(flow chart)
Preparation of
Quality Control
Receive Shop drawings
Submittal to
W/O & Drawings
the Engineer
Drawings Checking
for Approval
Quantity Completion of
Survey the W/o
Elements
Detailed Procurement of
Planning Resources Execution
of Works

Quality Policy
One Page Quality Management Roadmap.
It should be understood to all level of organization.
Quality Quality
Objectives Guidelines / Rules
Quality
Standards
• Customer Satisfaction • Quality culture

• Continuous Improvement • All Staff Responsibility
• Six Sigma Target • ISO 9001:2008 • Staff Training
• • Technical Standards • Leaders & Teamwork
• • Quality circles
Quality Policy
Our goal is to ensure customer satisfaction and minimize rejects.
The company achieves this goal by establishing, documenting,

implementing and maintaining an effective and efficient Quality
Management System that meets the requirements of ISO 9001:2008.
Quality improvement is a continuous process and it is primary

responsibility of management. All Staff must consistently strive for a six
sigma target.

Quality Assurance vs. Quality Control
Quality Assurance Quality Control
An overall management A series of analytical

plan to guarantee the measurements used to assess
integrity of data the quality of the analytical data
(The “system”) (The “tools”)
QA and QC are powerful techniques

• Quality Assurance (QA)  System oriented
• Quality Control (QC)  Product oriented
QA/QC or ITP Check List:
Quality Assurance / Quality Control (Inspection and Test Plan) checklist is
used to ensure the quality of a product / process. It defines set of tasks &
inspections to be performed in different stages of the product/process.
QA / QC Checklist for Arc Welding Process:
Responsibility Status Item Phase
Check Metal Sheets (type, size, defects, etc)
Check Welding Electrode (type, grade, humidity, etc)
Check Welding Machine & Tools Before
Welding
Check Welding Conditions (angle, amp, temp., area, etc)
Check Process (cleaning, area, safety, etc.)
Check Welding Procedure
Check Welding Positions During
Welding
Check Worker Efficiency (skills, safety, etc)
Visual Inspection (profile, size, surface defects, etc) After
Inspect Welding Quality Welding
Final Inspection Final(
Quality Documentation )Inspection
Comments:

Quality Control System
Before During After
Continuous
improvement

QC Throughout Production Systems
Inputs Conversion Outputs
,Raw Materials
Production Products and
Parts, and
Processes Services
Supplies
Control Charts Control Charts

and Control Charts and
Acceptance Tests Acceptance Tests
Quality of
Quality of Quality of
Partially Completed
Inputs Outputs
Products

Statistical Process Control (SPC)
A methodology for monitoring a process to identify special causes of
variation and signal the need to take corrective action.
• Seven QC tools:
1. Flowchart
2. Check sheet
3. Histogram
4. Pareto Chart
5. Scatter diagram
6. Control chart
7. Cause-and-effect diagram
• Focus: Quality control

7 Basic tools of Quality Control 2- Check sheet ( Defect Frequency)
1- Flowchart (Process sheet)
4- Pareto Chart (Sorting)
3- Histogram
6- Scatter diagram
5- Cause-and-effect diagram
7- Control chart

PARETO ANALYSIS Sorting
Defect Defect Defect Level Cumulative Ratio Defect Defect
Frequency Code Ratio % % Freq. Code
100 A 21.20 21.2 120 F
40 B 38.87 17.7 100 A
10 C 56.54 17.7 100 E
60 D 67.14 10.6 60 D
100 E 76.38 9.2 52 H
120 F 83.43 7.1 40 B
40 G 90.53 7.1 40 G
52 H 94.73 4.2 24 I
24 I 98.23 3.5 20 K
20 K 99.99 1.8 10 C
566 Total ≈ 100% 100 566 Total

PARETO ANALYSIS Sorting
Defect Defect Defect Level Cumulative Ratio Defect Defect
Frequency Code Ratio % % Freq. Code
100 A A1 21.20 21.2 120 F
40 B A2 38.87 17.7 100 A
10 C A3 56.54 17.7 100 E
60 D A4 67.14 10.6 60 D
100 E A A5 76.38 9.2 52 H
120 F B1 83.43 7.1 40 B
40 G B2 90.53 7.1 40 G
52 H B B3 94.73 4.2 24 I
24 I C4 98.23 3.5 20 K
20 K C C5 99.99 1.8 10 C
566 Total ≈ 100% 100 566 Total

PARETO ANALYSIS
21.2
F
17.7 17.7
A E
10.6
9.2
D
H
7.1 7.1
B G
4.2
3.5
I K
1.8
C
Defect

PARETO ANALYSIS Class B
K C
I
Class A G
B
H
D Class C
Defect
Cause-and-Effect Diagram or Fishbone diagram
Working Conditions Manpower Skills Method
Welding
Defect
Environmental Machine Materials

Defect Analysis
• Defect analysis for the last two months (May and Jun 2013)
Ratio (%) QTY (Ton) Defect

38% 47,367 Color Deviation
33% 41,979 Hazy
11% 13,682 Pinhole
6% 7,650 Bits
4% 4,904 Contamination
3% 3,644 Bad Flow
2% 2,850 Mechanical
2% 2,358 Gloss Deviation
1% 1,592 Bad Profile
100% 126,026 Total

PARETO Analysis
A B C

Normal Distribution:
n
x i  x  x
i
2
x i 1 
n n 1
Beta
Normal

Uniform
x
What is Sigma……
Sigma (σ) is the Greek symbol used to represent standard deviation,
a measure of the variation from the mean in a normal distribution.
That is a measure of dispersion around the mean.
 =   (X– X) 2
(n-1)
X
X
X
X X XX
X
X
XXXX
X XXX
XX
X
X
Degree of
 = 1.4 variation  = 0.7
Case Study: Process Data: 30 Units
Final Product
Product Dimensions
8.5 8.4 8.5 8.6 8.8
8.2 to 8.6 mm 8.4.2
8.4 8.5 8.4 8.5 8.5
Mean:
X =  [X1 + X2 + … Xn] / n
8.6 8.3 8.5 8.4 8.5
 =   (X– X) 2
(n-1) 8.5 8.6 8.4 8.5 8.5
8.6 8.4 8.5 8.6 8.4

Process  Mean = 8.5033 (Shift)
 = 0.1016 (Variance) 8.5 8.6 8.5 8.4 8.7

What is Sigma……
Sigma is a statistical unit of measure which reflects process capability.
C B A Process
10 10 10 Mean
9-11 8-12 6-14 Range
0.25 1 4 Variance
0.50 1 2 Sigma

Mean & Standard Deviation

3 Sigma and 6 Sigma Quality
Lower Upper
specification specification
1350 ppm 1350 ppm
1.7 ppm 1.7 ppm
Process
mean
+/- 3 Sigma
LSL +/- 6 Sigma USL

LSL USL
Brainstorming
Process Yield % Defects % DPPM Total Total
Sigma Level Defects Production
1 68.27 1,000,000
2 95.45 1,000,000
3 99.73 1,000,000
30,000 1,000,000
6,000 1,000,000
Process Control Chart
Stop the process; look for assignable cause
UCL
Good
CL
Good
LCL
Stop the process; look for assignable cause

PROCESS CONTROL CHART
Walter Shewhart
Developer of
Control Charts in
the late 1920’s

How to check the control chart
① ② ③
④ ⑤ ⑥ UC
L
LCL

Welding Process Rates: Defect Frequency:
Number of Total Week Defect Defect #

Rejected Joints #
Frequency Description
1
2 Incomplete Joint A
3 Improper Fusion B
4
Overlap C
5
6 Insufficient Penetration D
7
Pits E
8
9 Porosity F
10 Vertical Crack G
11
Under Bead Crack H
12
13 Irregular Weld I
Total Others K
Quality Control Tools:
The quality information for pipe line welding is as follows:
Number of Total Week Number of Total Week
Rejected Joints # Rejected Joints #
120 1100 11 120 1000 1
120 1200 12 110 1100 2
100 1200 13 100 1000 3
120 1300 14 110 1200 4
120 1300 15 105 1100 5
120 1300 16 100 1000 6
130 1400 17 100 1100 7
130 1400 18 110 1000 8
140 1500 19 120 1000 9
140 1500 20 110 1100 10
Based on this information, discuss the following:
1- Frequency Analysis 2- Control Chart
3- Sigma Level 4- Summary report
Welding Process Comment R% Rejected Production Week
Production rate Max. 12.0 120 1000 1
10.0 110 1100 2
1000 – 1500
10.0 100 1000 3
Joint/week 9.17 110 1200 4
Average = 1180 9.55 105 1100 5
10.0 100 1000 6
Rejected ratio
9.09 100 1100 7
8.33 – 12.00% 11.0 110 1000 8
Average = 9.84 % 12.0 120 1000 9
Range = 12-8.33 = 3.67 10.0 110 1100 10
10.9 120 1100 11
Standard Dev. = 0.943
10.0 120 1200 12
Quality Rate = 100 – 9.85 Min. 8.33 100 1200 13
= 90.15 % 9.23 120 1300 14
9.23 120 1300 15
1 S  68.27%
9.23 120 1300 16
X  90.15 9.29 130 1400 17
2 S  95.45% 9.29 130 1400 18
3 S  99. 73% 9.33 140 1500 19
9.33 140 1500 20
Sigma Level ≈ 1.81
9.84 116.25 1190 X’
(without 1.5 S Shift)
0.943 12.13 164.012 S
Sigma Level ≈ 2.80 8.3 100 1000 Min
TQM
(with 1.5 S Shift) Quality Engineering 12.0 140
Best Practice 1500 Max
183
Frequency Analysis
Frequency Class Limit
1 7.8-8.5
2 8.5-9.2
8 9.2-9.9
5 9.9-10.6
2 10.6-11.3
2 11.3-12.0
Total = 20

e t
e for men
B ve
Frequency ro
I mp
20
15
10
0
9 10 11 12
Rejected %
Mean 
x = 9.85
n
Standard deviation   ( mean  x ) 2

= 0.961
n 1

X’ = 9.84 S=0.943
UCL = x’ + 3 S = 12.67 UCL = x’ - 3 S = 7.0
Rejected % Upper Control Limit  mean + 3 stdev
Mean
Lower Control Limit  mean + 3 stdev
Week
Comment: All the points are within the control limits.
Rejected %
Out of chart
Fair
Good
Very good X
Excellent
Week

Process Capability Ratios
)Actual Performance( / )Desired Performance(
Process performance is
not necessarily centered
between the spec limits The shaded areas
This curve is the represent the
distribution of data percentage of off-spec
from the process
production
Voice of Customer
Voice of Process
The capability index is defined as:

Cp = (allowable range)/6 =T (Tolerance)/ 6= (USL - LSL)/6

Process Capability:
It is a method of measure used to determine if the process is capable to
produce the required quality level & customer specifications or not.
Process capability is the ability of the process to meet the desired quality
LSL USL
(Not capable!)
Cp < 1.0
X
Cp = 1.0
(Capable) X
(High Capable) Cp > 1.0

X
LSL USL
Process Capability 2.500.05
Assigned Tolerances
2.45 2.55
Good
CPK>1
-3 +3
Measured Variation
Poor
CPK<1
-3 +3
Process Capability: C pkVariations
(a) (b) (c)
Cpk = 1.0 Cpk = 1.33 Cpk = 2.0
LSL USL LSL USL LSL USL
(d) (e) (f)
Cpk = 1.0 Cpk = 0.60 Cpk = 0.80
LSL USL LSL USL LSL USL

Process Capability & Six Sigma
Nominal
value
Cpk < 1 Not Capable
Cpk > 1 Capable at 3 Six sigma
Cpk > 1.33 Capable at
4 Four sigma
Cpk
Cpk >
>=1.67
2 Capable
Capable at at
6
5 Three sigma
Process
Capability &
Design For 6 Lower Upper
Sigma specification Mean specification

Question:
The upper and lower specification limits for a component are 0.150 cm. and 0.120 cm.,
with a process target of .135 cm. The process standard deviation is 0.004 cm. and the
process average is 0.138 cm.
Discuss the process capability index
Customer  USL = 0.15 LSL= 0.12 Target Mean = M = 0.135
Process  S = 0.004 Mean = x` = 0.138 (X’ > M, the process mean is shifted)
Cp = (USL – LSL) / 6 S = (0.15-0.12)/6*0.004 = 1.25 > 1
Cpu = (USL – M) / 3 S = (0.15 – 0.135)/3*0.004 = 1 =
Cpl = (M -LSL) / 3 S = (0.135-0.12)/3*0.004 = 1.5 >1
Cpl > Cpu, since the process mean is not centered.
Process capability index = Lower value (Cp, Cpu, Cpl) = 1
This means, the process is barely capable to achieve the desired.
How to improve the process capability?
Width of landing
strip 1/2 Width
of landing
strip
25
50
If pilot always lands 100

25
within 1/2 the landing strip
width, we say that he has
Six-sigma capability.
Process Capability & Design For 6 Sigma
500.2 mm
USL = 50.2
Mean = 50.0
LSL = 49.8
Target  Design For 6 Sigma  Process Cap. = 2
USL = 50.2
Target Process Control Chart

12 S = 0.40 UCL = 50.1
6 S = 0.20
Mean = 50.0 6 S = 0.2  S=0.033
LCL = 49.9
S = Process Stand. Dev.

LSL = 49.8
Design for 6 sigma,

Reliability Improvements
Failure Mistake Error

How to Make A System Reliable
Maintenance Product Manufacturing

?
System System
Component determination
overview design
Component Reliability
design or testing & data ?
purchase analyses

Improving Reliability
• Quality Policy • Redundancy/backups
• Safety Policy
• Component design improvement
• Testing Policy
• System design improvement
• Maintenance Policy
• Production / Assembly Techniques
• Failure Repair Policy
• Training Policy
• Motivation Policy
• User Education Policy

Core Objectives
A= Availability = (Total Time –Total Down Time )/Total Time = MTBF /(MTBF+MTTR)
MTBF = Mean Time Between Failure = Operating hours / Total no. of failures
MTTR= Mean Time To repair = Total repair time / Total no. of failures

How to Minimize MTTR?
Fault Detect
Return Repair
Minimize Mean Time to Repair (MTTR):

Problem Diagnoses
Problem Troubleshooting
Spare Parts Analysis & Control

How to Maximize MTBF?
– More effective maintenance practices
– Predictive maintenance – based on condition monitoring
– Proactive maintenance – controlling root cause of failure
– Improved lubrication ……
Goal is to improve uptime – managed productivity
Maximize Mean Time Between Failure

System / Machine Breakdown
Critical Parts
Part Failure Analysis
Failure Mode Effect Analysis
Condition Based Maintenance

Core Objectives  Maximize RAMS
• Maximize Mean Time Between Failure (MTBF)
– System Breakdown
– Machine Breakdown
– Critical Parts
– Part Failure Analysis
– Condition Based Maintenance
• Minimize Mean Time to Repair (MTTR)

– Problem Diagnoses
– Problem Troubleshooting
– Spare Parts Analysis & Control
• Minimize the Risk of Failures

– Safety Criticality Analysis
– Failure Mode Effect Criticality (FMECA) Analysis
– Problem Solving Tools

How To Improve the System Effectiveness?
Equipment Availability
 Maintenance scheduling Maintenance Cost
 Labour utilization
 Response / repair time
 Maintenance planning
 Spares availability
 Contractor management
Equipment Reliability
 Effectiveness Operating Costs
 Equipment efficiency
 Quality of work/labour
 Operating efficiency (Utilization)
 Feedback to operations
Safety
 Equipment safety
 Personnel safety
Long Term Continues Improvements  Long Term Project
Long Term Plan / Medium Term Plan / Short Term Plan

Error Types:
Design errors: these types of errors are the result of inadequate design. for example
the placement of controls and displays so far apart that an operator is unable to use
them in an effective manner.
Operator errors: these errors are the result of operator mistakes and the conditions that
lead to operator errors include lack of proper procedures ,complex tax ,poor training
and operator carelessness.
Assembly errors: these errors occur during product assembly due to humans.
Inspection errors: these errors occur because of less than %100 accuracy of
inspectors.
Maintenance errors: the errors occur in the field due to oversights by the maintenance
personnel.
Installations errors: these errors occur due to various reasons including using the
wrong installation related blueprints or instructions.
Handling errors: these errors basically occur because of inadequate storage or

transportation facilities.

Causes for the occurrence of human errors:
• Poor motivation of involved personnel.
• Poor training or skill of concerned personnel.
• Poor equipment design.
• Inadequate or poorly written equipment operating and

maintenance procedures.
• Poor job environment: poor lighting ,high/low temperature ,high
noise level ,crowded work space, ,etc.
• Inadequate work tools.
• Complex tasks.
• Poor work layout

Human error occurrence classification:
• decision error: occur when the wrong decision is made after considering the
situation.
• Action error: are the result of no action ,incorrect action ,or the performance of
correct action on the wrong object when required.
• Transmission error: occur when information that must be passed or to others is not
send ,sent incorrectly ,or sent to the wrong destination.
• Checking error: occur when system require checks, the incorrect checks are
made ,checks are omitted ,or correct checks are made on the wrong object.
• Diagnostic errors :are the result of misinterpreting the actual situation when an
abnormal event occurs.
• Retrieval errors :occur when required information either from an individual, an

individual memory,or from any other reference source is not received or the incorrect
information is recieved.

• Causes of Unreliability – Improper design
– Improper materials
– Manufacturing errors
– Assembly and inspection errors
– Improper testing
– Improper packaging and shipping
– Improper start-up
– User abuse
– misapplication
• Random Failure Causes – Insufficient design margins

– Misapplication: overstress
– Use in wrong environment
– Predictable failure levels
– Intermittent failures
– Inherent manufacturing failures

Quality & Reliability Improvement Tools
Improve Stage: Product Development Tools:
Process Development Tools: • Reliability Analysis
• Root Cause Analysis • Quality Function Deployment
• Failure Mode Effect Analysis • Service Quality Analysis
• Visual Control Tools: • HAZOP Study
(5 S, Poka-Yoke, Jidoka, etc.) • Robust Design

• Value Engineering
Feasibility Study Tools:

• Value Analysis
• Action Plan

Process Development Tools
• Root Cause Analysis

• Failure Mode Effect Analysis
• Visual Control Tools:
(5 S, Poka-Yoke, Jidoka, etc.)

What is Root Cause Analysis (RCA)?
Root Cause Analysis is a systematic approach to get to the true
root causes of the problems.
WHY-WHY TECHNIQUE
Problem
Why's 5
You Can
Root Cause Control
Root Cause Analysis:
It is a method based on why-why technique to find the root cause of a
problem, and then define the remedy to prevent this problem in future.
“What happened analysis?”
Root Cause Analysis (RCA)
Material Working
Conditions
Method of Environmental
Work
Machine & Manpower

Tools Skills

Cause & Effect Diagram (C&E): “Ishikawa Diagram or Fishbone diagram”
It is used to present the priorities of root causes for a certain problem/defect.
Working Conditions Manpower Skills Method
Welding
Defect
Environmental Machine Materials

Cause-and-Effect Diagram or Fishbone diagram
Method Manpower
Drill Tired
Overtime
Slow Too many
defects
Wood Old
Steel Not Maintained
Material Machinery
Cause Effect
Failure Mode Effect Analysis (FMEA)
What If A tool used to evaluate
potential failure modes and their causes.
What If Problem
Analysis?
Cause Failure
Effect
Mode
Ishikawa
Failure Mode Effect Cause
Proactive Action (Risk Plan)

Failure Mode Effect Analysis:
It is a method based on risk assessment to predict the different failure modes
for the process, and then define the remedy to prevent this modes in future.
“What If analysis?”
FMEA for Welding Process
Proactive Action Cause Effect Failure Mode
(QA/QC)
- Overlap Slow Speed
-
- Incomplete Joint High Speed

-
- Irregular weld Inconstant
- Speed

5 S, Japanese Tool, 1970
5 Pillars of Visual Control & Workplace Organization
“Productivity Improvement Tool”
A Systematic Approach for Visual Control & Workplace
Organization in order to improve safety, quality and productivity
1.
Sort
5. 2.
Sustain
5S Set in order
4. 3.
Standardize Shine

What's 5'S?
Sometimes called 5S + 1 or “6S”

Safety First ! (in everything we do!)
Before After

Before After

Shigeo Shingo
Poka-Yoke (Mistake-Proofing)
An approach for mistake-proofing processes
using automatic devices or methods to avoid
simple human or machine error.
Work to achieve defect prevention

not defect detection A Strategy for
Performance
• Defect warning system
Excellence
• Make it impossible to make mistakes
• Initial term was baka-yoke, which means
‘fool-proofing’
• Improve quality and reliability
• Can eliminate both human and mechanical
errors

Poka-Yoke Examples:

Process for Mistake Proofing
Describe Defect and Defect Rate
Determine Defect Location
Study the Process Flow
Observe the Process
Identify Errors and Determine Cause
Determine Prevention Method
Test/Re-test in Extreme Conditions

Jidoka Examples:
Temperature
Alarm
Vibration Alarm

Jidoka Examples:

Product Development Tools
• Quality Function Deployment
• Service Quality Analysis
• HAZOP Study
• Robust Design
• Value Engineering

Desig
n for R
eliabil
i ty

Reliability
The probability that no failure will occur in a given time
interval
A reliable system is one that meets the specifications
The probability a unit will perform its intended function
for a specific period of time (MIL-STD-721B).
R(t)
R (t )  P(t | t  T ) , T 0
1
0
t
• Probability of zero failure
• The ability of the system to delver services as specified
• Probability that the product, service or part will function as expected
Reliability R(t)  f (MTBF)
R(t) = exp(-λt) ≈ 1 – λt
Failure Rate =  = 1 / MTBF
[when λt is less than 0.001]
R(t) = e-t
In this formula, the failure rate, λ, is assumed to be constant with the age of the
component.
• MTBF (mean time between failure) 1

MTBF 
= the amount of time between failure of a repairable product  (t )
• MTTF (mean time to failure) 1

= the amount of time between failure of a non-repairable (i.e., must be MTTF 
replace) product
 (t )

Example:
• 10 components were tested. The components
(not repairable) failed as follows:
 Component 1 failed after 75 ours Determine the
 Component 2 failed after 125 hours MTBF
Solution: 325 525
130
Five failures, operating time = 3805 hours 125 5 x 525
75
Failure Rate =
 = 5 / 3805 = 0.001314
What is the probability that it will reliable during the first 200 hours of operations?
Rt = e – λt e – (200) (0.001314) = % 76.89 =

Example:
MTBF = 100,000 / 20 = 5,000
• What is the probability that it will reliable during the first 200 hours
of operations?
Rt = e – λt e – (200) (0.0002) = % 96.08 =

• Probability of zero failure
• Probability that the product, service or part will function as
expected
Rs =
R1 R2 R3
0.95*0.85*0.90
0.95 0.85 0.90
= 0.7 = 70%
Parallel structure: Rs = 1 –(1-R1)* (1-R2)* (1-R3) * ..…

R1 = 0.95 Rs =
1- (1-0.95) *
R2 = 0.85
(1-0.85) * (1-0.90)
R3 = 0.90 = 0.99925 = 99.925%

Question :
RA = RB = .90 and RC = RD = .95.
Since A & B are parallel, R(AB) = 1 – { (1-Ra) * (1-Rb)} = 0.99 = 99 %

Since C & D are parallel, R(CD) = 1 – { (1-Rc) * (1-Rd)} = 0.9975 = 99.75%
Since AB & CD are series, R(ABCD) = R(AB)*R(CD) = 0.987525 = 98.75%
System reliability (which means the probability of zero failure) = 98.75%

This means,
How to improve the system relibility?


Quality Function Deployment (QFD)
QFD is a strategic quality planning tool to translate the
customer requirements to design indicators & target quality
plan, (Design for 6 Sigma) - 1970
A technique for identifying customer requirements and matching

them with engineering design and performance parameters
QFD is a planning tool used to fulfill customer expectations. It can

be applied to manufacturing and service industries.
More than 1000 International Company:
• Toyota, 1977 Ford, 1978
• IBM, 1982 Boeing, 1983
• Nokia, 1986

QFD: Quality Function Deployment
It is a tool used to translate the customer requirements to design indicators &
target quality plan; as “strategic quality planning tool”
SMART
Quality Plan
FEATURES
PROCESS
PROCESS
FEATURES
FEATURES
PRODUCT
PRODUCT
FEATURES
CUSTOMERS
NEEDS
CUSTOMERS
NEEDS
CUSTOMERS

House of Quality Example
You’ve been assigned

temporarily to a QFD team.
The goal of the team is to

develop a new camera
design.
Build a House of Quality.
© 1984-1994 T/Maker Co.

Customer Needs:
Importance Customer Needs #
60% Light weight 1
40% Easy to use 2
25% Reliable 3
20% Easy to hold steady 4
Focus on 20% No double exposures 5
Customer Satisfaction 15% Color correction 6

15% Reduce Size 7
10% Improve Power Eff. 8
10% Low Cost 9
5% Product Colors 10
5% Product Assembly 11
3% Others 12
Reference, Benchmarking, Standards

Engineering Characteristics (Product Design) – Rule 80/20
Class How? Importance Customer #

Needs
A - Aluminum Parts 3 Light weight 1

B - Auto Focus 2 Easy to use 2
B - Auto Exposure (flash) 1 Reliable 3
C - Auto Film advance
C - Auto Photograph
C - Ergonomic design
(Redesign Structure)

Customer Requirements & Importance (Rule 80/20)
Customer Customer
Requirements Importance
Light weight 3
Easy to use 2
Reliable 1
Target Values
H- High relationship M- Medium relationship L- Low Relationship

Engineering Characteristics (Product Design) – Rule 80/20
Customer Customer Aluminum Auto Auto

Requirements Importance Focus Exposure
Parts
Light weight 3
Easy to use 2
Reliable 1
Target Values

Relationship Matrix
Relationship between customer

attributes & engineering
characteristics (‘rooms’)

Requirements Importance Parts Focus Exposure
Light weight 3 
Easy to use 2  
Reliable 1   
Target Values

Relationship

Light weight 3 
Target Values

Target Values (Priority System)  (A, B, C, D)
100
10
C A
Budget Flexibility %
50
D B
0
0 50 100
Technical Flexibility %
Target Values (Priority System)
Target values for engineering

characteristics (‘basement’);
key output
H

Light weight 3 
Target Values A B B

Components of Quality: Service-based
Service Quality (ServQual) Analysis
Customer Satisfaction
Customer Expectations Customer Demands
+ Price
Service Quality (ServQual)
– Tangibles: appearance of physical facilities, equipment,
personnel, and communication materials
– Reliability: ability to perform the promised service dependably

and accurately
– Responsiveness: willingness to help customers and provide

prompt service
– Assurance: knowledge and courtesy of employees and their

ability to convey trust and confidence
– Empathy: the caring, individualized attention provided to the

customers
Service Quality (ServQual) Analysis:

Model of Service Quality
Consumer expectations
Gap 1
Management perception of
consumer expectations
Gap 2
Service quality specifications Gap 5
Gap 3
Delivery of specifications
Gap 4
External communications

HAZOP Analysis
(Hazard Operation Study) - “Hazard and Operability Study”
HAZARD  NEAR-MISS  INCIDENT  ACCIDENT

HAZOPS Analysis Process
Parameter*
Guide Words
None Prevention/ Action

Deviation Cause Consequence
More Of Mitigation
Item
Less Of
Part Of
Consequence,
As Well As
Likelihood,
Other.
& Risk Ranking
.
*Covering every parameter relevant to the system:

flow, pressure, temperature, viscosity, components, etc.

HAZOP Terms
• Guide-words:
• (NO, LESS, & MORE).
• Deviation:
• (flow, pressure, and temperature).
• Possible causes:
• (equipment and instrumental failure).
• Consequences:
• (problems results).
• Actions:
• (problems solutions)
HAZOP Procedure
1. Divide the system into sections.

2. Choose a unit.
3. Describe the design intent.
4. Select a process parameter.
5. Apply a guide word.
6. Determine causes.
7. Evaluate Consequences/problems.
8. Recommend action.

HAZOP Work Sheet
Guide Word Deviation Cause Consequence Action
e.g. e.g. Potential Consequences Action need

cause from of the cause it to Solve
No No Flow the deviation and the
Low Flow deviation itself problem
Low
High High Flow
e.g. “more” + “temperature” = “higher temperature”

Case Study – Shell & Tube Heat Exchanger
Using relevant guide works, perform HAZOP study on shell
& tube heat exchanger
Process
fluid
Cooling
water
HAZOP on Heat Exchanger – Answer 1
Action Consequences Causes Deviation Guide Word
High Temperature of Pipe blockage Less flow of Less
Temperature process fluid cooling water
Alarm remains constant
Low Temperature of Failure of More cooling More
Temperature process fluid cooling water flow
Alarm decrease valve
Install high Bursting of tube Failure of More More of
pressure alarm process fluid pressure on
valve tube side
Proper Contamination of Leakage of Contaminatio Contamination
maintainance process fluid tube and n of process
and operator cooling water fluid line
alert goes in
Proper Less cooling and Hardness of Corrosion of Corrosion
maintainence crack of tube cooling water tube

HAZOP on Heat Exchanger – Answer 2
Action Consequences Causes Deviation Guide
Word
Install Temperature Process fluid Failure of inlet No cooling water NONE
indicator before and after temperature is not cooling water flow
the process fluid line lowered accordingly valve to open
Install TAH
Install Temperature Output of Process Failure of inlet More cooling MORE
indicator before and after fluid temperature too cooling water water flow
process fluid line low valve to close
Install TAL
Installation of flow meter Process fluid Pipe leakage Less cooling LESS
temperature too low water
Install check valve Product off set Failure of Reverse process REVERSE
(whether it is crucial have process fluid fluid flow
to check?) inlet valve
Proper maintenance and Outlet temperature Contamination Process fluid CONTAMI

operator alert too low in cooling water contamination NATION

‫لمتماسك‬ee‫ ا‬e‫لتصميم‬ee‫ ا‬Robust Design
DFMA – Design For Manufacture and Assembly
“Robust means product & process insensitive to noises”
Goals: Genichi Taguchi, 1985
‫ميم ويعزى‬eee‫ى بالتص‬eee‫نظام جودة يع‬
• Improve Design Flexibility
‫ز‬ee‫ي ويرك‬ee‫ي تاجوتش‬ee‫ث اليابان‬ee‫للباح‬
• Improve Product Quality ‫اسية‬eee‫ف األس‬ee ‫ين الوظائ‬eee‫ى تحس‬ee ‫عل‬
‫هل‬eeeeee‫ج والعمليات وبهذا يس‬eeeeee‫للمنت‬
• Improve Product Performance
‫ا اإلنتاج‬e‫ميم وتكنولوجي‬e‫ة التص‬e‫مرون‬
• Minimize Manufacturing Variation ‫ي‬e‫لوب فعال ف‬e‫د اس‬e‫ ويع‬،‫ه‬e‫ة ب‬e‫المرتبط‬
‫ين الجودة‬ee‫ف وتحس‬ee‫ض التكالي‬ee‫تخفي‬
• Minimize Assembly Errors ‫ر‬eee‫ت تطوي‬eee‫ض وق‬eee‫ى خف‬eee‫ة إل‬eee‫إضاف‬
• Minimize Total Cost .‫المنتج‬
http://www.isixsigma.com/library/content/c020311a.asp
www.amsup.com/robust_design/
Taguchi Method
• The Robust Design method can also be called the Taguchi Method, and was
pioneered by Dr. Genichi Taguchi.
• Robust Design greatly improves engineering productivity by consciously
considering the noise factors: environmental variation during the product's
usage, manufacturing variation, and component deterioration.
• The cost of failure in the Robust Design method helps ensure customer
satisfaction.
• Robust Design focuses on improving the fundamental function of the product
or process.
• Robust Design facilitates flexible designs and concurrent engineering. It is the
most powerful method available to reduce product cost, improve quality, and
reduce development interval
(http://www.isixsigma.com/library/content/c020311a.asp).
Process design + Process control = Integrated approach

Product Value Analysis
Value engineering is a systematic evaluation of the value-

chain to improve the product & reduce the cost.
Function analysis:
Balance
Value Engineering
Problem Solving Methodology
ce
Qu
an
rm
ali
rfo
ty
Pe
Life Cycle Cost (LCC)

The Value Expression
Cost Function
V = F/C
Function + Quality + Performance
Value =
Cost + Time
FUNCTION
VALUE =
COST
To Increase Value
F F F F F
C C C C C
REDUCE INCREASE MAINTAIN

Value Engineering and Value Analysis
• VA is the application of a set of techniques to an existing
product with a view to improving its value. It is a remedial
process.
• VE is the application of exactly the same set of techniques to a

new product at the design stage, project concept or preliminary
design when no hardware exists to ensure that bad features are
not added.
Planning & Construction

Design
Value Engineering Value Analysis

Product Innovation =
Insight + Ideas + Impact

Product Evolution

Cost Reduction of a Bracket through Value Engineering

Design Simplification
(a) Original design (b) Revised design (c) Final design
Assembly using One-piece base & Design for

common fasteners elimination of push-and-snap
fasteners assembly

Definition
phase
Implementation Information
Phase phase
Presentation
Value Eng. Function
phase Main 8 Steps Analysis
phase
Development Creativity
phase phase
Evaluation
phase
Feasibility Study Tools
• Value Analysis
• Action Plan

easibility Analysis
A feasibility study leads to a decision:

• Go ahead
• Do not go ahead
• Think again
Conducting a Feasibility Study

• Planning the study
• Collecting the data
• Analyzing the data
• Writing the report
• Presenting the report

Value Analysis - From the Customer Point of View:
A manager is trying to decide between four air condition

systems (A,B,C,D) for a computer lab; area 16 m * 18 m * 4 m;
18 Users, U-layout, maximum budget 60,000 LE.
Area Layout:
4m
16 m
18 m

Design Assumptions: (1 HP/ 8000 Btu)
• If Height ≤ 3 m Then 1 HP / 10 m2
• If Height > 3 m Then 1 HP / 25 m3
Design Factor:
Average temperature 1.00 High temperature 1.15
Very high temperature 1.30 Special cases 1.50
Cost Information (LE):
Air Condition Type Unit

D C B A HP
5600 5700 5800 6000 3

6500 6550 6700 7000 4
6800 7000 7200 7600 5
7000
TQM 7200 Quality Engineering
7500 Best Practice
8000 6 274
Check List For Customer Importance
Customer Importance Customer Main
5 4 3 2 1 Needs Parameters
- After sales service Maintainability

- Easy to replace any parts
- Spare parts availability
- Spare parts cost
- Low noise Quality &
- Improve reliability
Safety
- Good air distribution
- Improve air filtration
- Quick response
- Low water drops
- Power saving Power
- Auto switch off Consumption
- Easy to use (easy remote) Options
- Cooling & heating modes
- Small Size Design & View
- Light weight
- Look
Quality nice
Engineering Best Practice
TQM 275
Voice of the Customer:
Rating Customer Needs Main Parameters
4 - After sales service Maintainability
- Easy to replace any parts
- Spare parts availability
- Spare parts cost
3 - Low noise Quality & Safety
- Improve reliability
- Good air distribution
- Improve air filtration
- Improve Cooling Performance
- Quick response
- Low water drops
2 - Power saving Power Consumption
- Auto switch off
1 - Easy to use (easy remote) Options
- Cooling & heating modes
1 - Small Size Design & View
- Light weight
- Look nice
- Different Colors
Rating System:
Scale
10 8 6 0 Parameters
A B C,D - Maintainability
A B,C D - Quality & Safety
A B,C D - Power Consumption
A B C,D - Options
A B C,D - Design & View
Based on this information, discuss briefly the following:
From the Customer Point of View:
• Best number of units
• Best layout
• Best type selection
• Budget analysis

From the Customer Design Assumptions: (1 HP/ 8000 Btu)
Point of View: • If Height ≤ 3 m Then 1 HP / 10 m2
• If Height > 3 m Then 1 HP / 25 m3
Area = 16 * 18 = 288 m2 Volume = 16 * 18 * 4 = 1152 m3

Required Power = 1152/25 = 46 HP
Best Layout:
8 Units * 6 HP = 48 HP

Cost Information (LE):
Air Condition Type
D C B A
7000 7200 7500 8000 Unit 6 HP
8 8 8 8 Number
56000 57600 60000 64000 Total Cost
√ √ √ X Max. Budget
Technical Evaluation:
Air Condition Type Relative
D C B weight Parameters
6 6 8 4 Maintainability
6 8 8 3 Quality & Safety
6 8 8 2 Power Consumption
6 6 8 1 Options
6 6 8 1 Design & View
66
TQM 76 88 Engineering Best
Quality Technical
Practice Evaluation 279
Value Analysis:
Air Condition Type
D C B Parameters
66 76 88 Technical Evaluation
Best
56 57.6 60 Cost Evaluation
Best
1.17 1.32 1.46 Value Index = T/C
Best
Best Option:
According to the value index, Type B is the best option
Budget Analysis:
Product Type: B Unit Price = 7500 LE
Number of units = 8 Units * 6 HP
Total Budget = 8 * 7500 = 60,000 LE
Action Plan:
Responsible Finish Start Duration Budget Main Main Main
Activities Items Objective

References:
• J. Moubray, Reliability - centered Maintenance, Butterworth-Heinemann, Oxford,
1997.
• Charles E. Ebeling, An Introduction to Reliability and Maintainability Engineering,
Waveland Press, Inc., 2005.
• Elsayed, A. Elsayed, Reliability Engineering, Addison Wesley, 1996.
• MIL-P-1629 “Procedures for performing a failure mode, effects and criticality analysis
• Nicholls, David B., System Reliability Toolkit, DoD Reliability Information Analysis
Center (RIAC), 2005
• O'Connor, Patrick D. T., Practical Reliability Engineering, 4th ed., Wiley, 2002
• http://www.fmeca.com/
• Software from ReliaSoft Co.
- Weibull++6
- BlockSim6
- ALTA6 Pro
Lindo or CPLEX
– Problem formulation

Made in Egypt - ‫يمصر‬eee‫ ف‬e‫نع‬ee‫ص‬
‫اللهم أحفظ مصر من كل سوء – اللهم أمين‬
!Thank you for your attention
Useful Science For the

Suggestions Comments Sake of Allah
‫قَالُوا ُسب َْحانَ َك ال ِع ْل َم لَنَا إال َما‬

" ‫الح ِكي ُم‬
َ ‫نت ال َعلِي ُم‬َ ‫َعلَّ ْمتَنَا إنَّ َك َأ‬
(32 ‫ية‬T‫لبقرة آ‬TT‫ورة ا‬TT‫)س‬
Questions
Copyrights & More Information:

Dr. Attia Hussien Gomaa
Industrial Engineering Professor & Consultant
attiagomaa@yahoo.com or attiagomaa@aucegypt.edu
Tel: 01222738497 We’re here to help you!
Design For Reliability
Desig
n for R
eliabil
i ty

Design of Experiments (DoE)
Statistical tool for systematic testing
Main Factors Key Response

(Inputs) (Outputs)
- Material Type Process

- Speed
- Temperature - Accuracy
- Method - Surface Finish
- Tool Type
Inputs  Main Factors

Outputs 
Testing Tool Method Temp. Speed Material
Result Level
(E) (D) (C) (B) (A)
1
2
Example: DoE Life testing for a motor
Outputs 
Testing Coating Wind Wire Wire
Wire type
Result Material Method tension diameter Level
(A)
(E) (D) (C) (B)
Time To Fail Old WM1 90 50 Aluminum 1
New WM2 60 36 Copper 2
Main Factors
Run
Time To
Coating Wind Wire Wire Wire Order
Fail
Material Method tension diameter type (Day)
5,61 Old WM2 90 50 Al 1

4,59 New WM2 90 36 Al 2
6.29 New WM1 90 50 Al 3
22.65 New WM1 90 36 Co 4
20.29 Old WM2 60 36 Al 5
4.65 Old WM1 60 36 Co 6

High Low Standard
P Value 2 Value Coefficient Effect Term
CI CI error
Beta
Intercept
0.5702 1.1403 A:
-0.416 -0.832 B:
- - C:
- - D:
0.2949 0.5898 E:
0.5248 1.0496 AE
-0.3479 -0.6957 ABC
To maximize life of the motor choose the factor settings that maximize this
equation:
• A & E Set to high value (+ve effect)
•B Set to low value (-ve effect)
• C & D Do not significantly effect.

Example:
Number of variables = 3
3 Factors & 3 Level:
- A (Max., Nominal, Min.) (+1, 0 , -1) Total # of experiments:
- B (Max., Nominal, Min.) (+1, 0 , -1)
- C 2^f = 2^3 = 8
(Max., Nominal, Min.) (+1, 0 , -1)
Factor (C)
Outputs # Average gain for C=-1
C B A
13.80 0 0 0 0 13.7725
12.85 -1 -1 -1 1
Average gain for C=+1
13.01 -1 -1 1 2
14.52 -1 1 -1 3 13.86
14.71 -1 1 1 4
12.93 1 -1 -1 5
Slope =(13.86-13.7725)/2
13.09 1 -1 1 6 = 0.044
14.61 1 1 -1 7
14.81 1 1 1 8

Factor (B)
Outputs # Average gain for B=-1
C B A
13.80 0 0 0 0 12.9700
12.85 -1 -1 -1 1
Average gain for B=+1
13.01 -1 -1 1 2
14.52 -1 1 -1 3 14.6625
14.71 -1 1 1 4
Slope =(14.6625-12.97)/2
12.93 1 -1 -1 5
13.09 1 -1 1 6 = 0.85
14.61 1 1 -1 7
14.81 1 1 1 8

Factor (A)
Outputs # Average gain for A=-1
C B A
13.80 0 0 0 0
12.85 -1 -1 -1 1
Average gain for A=+1
13.01 -1 -1 1 2
14.52 -1 1 -1 3
14.71 -1 1 1 4
Slope =
12.93 1 -1 -1 5
13.09 1 -1 1 6 =
14.61 1 1 -1 7
14.81 1 1 1 8

Accelerated Testing
Accelerated Life Models Usually Consist of:
• A life distribution at each stress level (from Weibull Analysis)
• A life-stress relationship (from physics of failure or a statistical model
High
Stress 2
High
Stress 1
Use
Stress
Time (t) Stress

• Use engineering knowledge to choose a life-stress model
• Need enough data to find the model parameters
• Important role for simulation & expert judgement
Bathtub curve
h(t) DFR phase: Initial design, constant bug fixes
CFR phase: Normal operational phase
IFR phase: Aging behavior
(burn-in-period) (wear-out-phase)
CFR
(useful life)
DFR IFR
t
Decreasing failure rate Increasing fail. rate
Distributions:
Weibull
Β <1
Failure Rate
Exponential Β >1
Normal
Β =1 Lognormal
0 Time t
Stress-Strength Analysis:
Stress Strength
Fail Region
Increase design
margin
Reduce variation in
product strength
)Sigma Approach-6(

Stress-Strength Analysis:
Reduce effect of
usage /
environment

Reliability Block Diagram (RBD):

R1 R2 R3 R4
R1 = 0.98
R2 = 0.97 Rs = R1 x R2 x R3 x R4
R3 = 0.95 = 0.98 x 0.97 x 0.95 x 0.99
R4 = 0.99 = 0.894
Rs = Rin ; i = 1, 2, ……, n
0.99 = Ri1000
ln 0.99 = 1000 ln Ri
Ri = 0.99998995
 = 0.06
MTBF = 1000/0.06
λ t
R( t)  e
0.06(8)
R(8)  e
 0.619

Reliability Engineering - 123 - 07 10 2013

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reliability Engineering - 123 - 07 10 2013

Uploaded by

Copyright:

Available Formats

‫وما توفيقي إال باهلل‬

Reliability Engineering Best Practice

Dr. Attia Gomaa

TQM Reliability Engineering Best Practice 2

This is an Open Discussion Course:

– Share Knowledge ‫– تبادل المعرفة‬

– Share Questions ‫– تبادل األسئلة‬

Remember … You can’t do it alone!

Failure Mistake Error

TQM Reliability Engineering Best Practice 4

Doing Right Thing Right First Time and every Time

For example; Product: Car

• Price • Brand • Manual / Auto. • Performance

For example; Product: Car

• Price • Brand • Manual / Auto. • Performance

• Energy • Failure-free • Options • Duty

• Spare Parts • Error prevention • Size •

• • Fault detection • Colors •

• • Safety System • Shape •

TQM Reliability Engineering Best Practice 8

Efficiency Flexibility Reliability Cost Saving

"Quality” is closely linked to “Reliability”

Reliability is defined as the probability that an item will perform a required

A statement of reliability has four key components:

• Probability  for example; a reliability goal of .9995

• Required function  For example, a pump’s required function

• Stated conditions  working condition, environmental conditions,

• Specified period of time  For example, a pump might be designed to

TQM Reliability Engineering Best Practice 12

TQM Reliability Engineering Best Practice 13

TQM Reliability Engineering Best Practice 14

TQM Reliability Engineering Best Practice 15

Quality Performance Safety Time

TQM Reliability Engineering Best Practice 18

Customer Price Limit

Failure: Situation in which a product, part, or system does not perform

Failure in Failure in Failure in Failure in

Failure due to Failure due to

TQM Reliability Engineering Best Practice 20

Systematic Approach For Long Term Continuous Improvements

TQM Quality Engineering Best Practice 21

TQM Quality Engineering Best Practice 22

(QA/QC) Quality Assurance/Quality Control (Inspection and Test Plan) checklist

TQM Quality Engineering Best Practice 24

Systematic Approach For Long Term Continuous Improvements

TQM Reliability Engineering Best Practice 25

Design for Rel.

TQM Quality Engineering Best Practice 26

Reliability Availability Safety Security

Fault (Defect) Fault Fault (Error) Fault (Failure)

TQM Reliability Engineering Best Practice 27

Determine the weaknesses in a design

Areas Covered by Reliability

TQM Reliability Engineering Best Practice 28

Warranty costs measured in million dollars for several large

Dr. Joseph M. Juran W. Edwards Deming

(1904 – 2008) (1900 – 1994)

TQM Quality Engineering Best Practice 30

TQM Quality Engineering Best Practice 31

TQM Quality Engineering Best Practice 32

• Product Design • Probability density function (PDF)

• Parts Design • Probability Distributions

• Material Selection & Control • Statistical process control (SPC)