ENGINEERING DESIGN 1

Topic 10 :Design for Safety, Risk

& Reliability

Ir Dr Zainal Fitri B Zainal Abidin

Sept 2016
Introduction

2
Classification of Societal
Hazards

3
 Regulation as a Result of
Risk
• The following federal organizations have a
major role to play in regulating technical risk:
• USA
• Malaysia
– Consumer Product Safety Commission
(CPSC) – Drug Control Authority (DCA)
– Environmental Protecting Agency (EPA) – Environmental Protecting Agency
– Federal Aviation Agency (FAA) (EPA)
– Federal Highway Administration (FHA) – Malaysian Communication &
– Federal Railway Administration (FRA) Multimedia Commission (MCMC)
– Nuclear Regulatory Commission (NRC) – Board Of Engineers Malaysia (BEM)
– Occupational Safety and Health – Occupational Safety and Health
Administration (OSHA) Administration (OSHA)

4
Sample of US Federal Laws
Concerning Product Safety

5
 Standards
• The role that standards play in protecting public safety
was first shown in the United States in the middle of
the19th century.
• The explosion of steam boilers was an all-too-frequent
occurrence, until the ASME developed the Boiler and
Pressure Vessel Code that prescribed detailed
standards for materials, design, and construction.
• Other examples of public safety standards are:
– Fire safety and structural codes for buildings
– Codes for the design, construction, maintenance, and
inspection of elevators.

6
 Mandatory Versus Voluntary
Standards
• Standards may be mandatory or voluntary.
• Mandatory standards are issued by governmental agencies,
and violations are treated like criminal acts for which fines
and/or imprisonment may be imposed.
• Voluntary standards are prepared by a committee of
interested parties usually under the sponsorship of a
technical society or a trade association.
• A voluntary standard indicated the lowest safety level that an
industry intends to provide in the product it manufactures.
• A mandatory standards indicated the lowest safety level the
government will accept.
7
 Risk Assessment
• The assessment of risk is an imprecise process involving judgment and
intuition.
• The level of risk, as perceived by an individual or the public, can be
classified as:
– Tolerable risk:
• Indicates that people are prepared to live with the level of risk but
want to continue to review its causes and seek ways of reducing
the risk.
– Acceptable:
• Indicates that people accept the level of risk as reasonable and
would not seek to expend much in resources to reduce it further.
– Unacceptable:
• Indicates that people do not accept this level of risk and would
not participate in the activity or permit others to participate.
8
 Data On Risk

• Data on risk are subject to considerable uncertainty

and variability.
• In general three classes of statistics are available:

– Financial losses (chiefly from the insurance

industry)

– Health information

– Accident statistics 9
Fatality Rate

10
Probabilistic Approach to
Design
Basic Probability Using the
Normal Distribution

12
Standardizing Normal Distribution

13
Area Under Standardized Normal
Frequency Curve

14
 Sources Of Statistical Tables
• All statistical texts contain tables for the z distribution,
the confidence limits of the mean, and the t and f
distributions.
• Here we mention two convenient sources of statistical
tables and information:
– Microsoft spreadsheet program Excel provides
access to many special mathematical and statistical
functions.
– The NIST/SEMATECH e-Handbook of Statistical
Methods is the modern version of Experimental
Statistics (www.itl.nist.gov/div898/handbook).
15
Statistical Functions Available in
MS Excel

16
Variability in Material Properties

17
Typical Values of Coefficient Of
Variation

18
 Safety Factor
• An important concept in risk and reliability analysis is that
hazards are controlled, mitigated, or removed by barriers.
• The property of a material that is used to build a component
can be considered a barrier.
• This situations is considered in a class of problems called
Stress-Strength model.
• This model assumes that the barrier fails if the stress
exceeds the resistance of the material to the stress,
measured in terms of some material property like yield
strength.
• The use of a safety factor is the oldest and simplest stress-
strength model.
19
Safety Factor

20
Component Of Safety Factor

21
Estimating the Contribution
from the Material

22
Estimating the Contribution
from the Load or Stress

23
Estimating the Contribution
from Tolerances (Geometry)

24
Estimating the Contribution
from Failure Analysis

25
Estimating the Contribution
from Reliability

26
 Safety Factor Consideration
In Engineering Practice
Three safety factor that are required are :

1) The design loads.

2) The material properties.
3) The performance criteria.

27
Reliability-Based Safety Factor

28
Mean & Standard Deviation of
Independent Random Variables x and y

29
Value of z to give Different Levels of
Probability of Failure

30
Reliability Theory
Reliability

32
Failure Rate

33
 Form of Failure Curves

Three-stage curve typical of Failure curve more typical of

electronic equipment mechanical equipment

 The general failure curve is the summation of three competing

processes:
(1) An early failure process
(2) A random failure process
(3) A wearout process 34
Definitions

35
Average Failure Rates For A Variety
of Components and Systems

36
Constant Failure Rate

37
 Constant Failure Rate Example

If a device has a failure rate of 2 × 10−6

failures/h, what is its reliability for an operating
period of 500 h? If there are 2000 items in the
test, how many failures are expected in 500 h?
Assume that strict quality control has eliminated
premature failures so we can assume a
constant failure rate.

38
 Constant Failure Rate Solution

Re (500) = exp(-2x10−6 x 500) = e−0.001 = 0.999

N𝑠 = N0 R(t) 2000(0.999) = 1998
N𝑓 = N0 - N𝑠 = 2 failures expected

If the MTBF for the device is 100,000 h, what is the

reliability if the operating time equals 100,000 h?
1
t=Ť=
λ
R(t)= 𝑒 −𝑡/Ť = 𝑒 −1 = 0.37
• We note that a device has only a 37 percent chance
of surviving as long as the39MTBF.
 Constant Failure Rate Solution
• If the length of the constant failure rate period is 50,000 h, what is the
reliability for operating for that length of time?
R(50,000) = exp x (-2x10−6 x 5x 104 ) = 𝑒 −0.1 = 0.905

• If the part has just entered the useful life period, what is the probability it
will survive 100 h?
R(100) = exp (-2x10−6 x 102 ) = 𝑒 −0.0002 = 0.9998

• If the part has survived for 49,900 h, what is the probability it will survive
for the next 100 h?
R(100) = exp (-2x10−6 x 102 ) = 𝑒 −0.0002 = 0.9998

We note that the reliability of the device is the same for an equal period of
operating time so long as it is in the constant-failure-rate (useful-life) region.
Weibull Frequency Distribution

41
Weibull Frequency Distribution

42
43
 Weibull Plot For Life Of Ball
Bearings

C. Lipman and N. J. Sheth, Statistical Design and Analysis of Engineering Experiments, p. 41, 1974.
44
Reliability With A Variable
Failure Rate

45
System Reliability

46
System Reliability

47
 RELIABILITY QUESTIONS

A complex engineering design can be described by a reliability

block diagram as shown. In subsystem A, two components
must operate for the subsystem to function successfully.
Subsystem C has true parallel reliability. Calculate the reliability
of each subsystem and the overall system reliability.

48
 RELIABILITY Solutions

Answer :

49
 RELIABILITY Solutions

Answer :

50
 Maintenance and Repair
• An important category of reliability problems
deals with maintenance and repair of systems.
• If a failed component can be repaired while a
redundant component has replaced it in service,
then the overall reliability of the system is
improved.
• If components subject to wear can be replaced
before they have failed, then the system
reliability will be improved.
• Preventive maintenance is aimed at minimizing
system failure. 51
Maintainability

52
Availability

53
Design For Reliability
 Design for Reliability
• The design strategy used to ensure reliability can fall
between two broad extremes.
• The fail-safe approach is to identify the weak spot in the
system or component and provide some way to monitor
that weakness.
• Frequently an absolute worst-case approach is used, in it
the worst combination of parameters is identified and the
design is based on the premise that all can go wrong at the
same time.
– This is a very conservative approach, and is often leads
to overdesign!

55
 Design for Reliability (2)
• Two major areas of engineering activity determine the
reliability of an engineering system:

1. Provision for reliability must be established during the

design concept stage, carried through the detailed
design development, and maintained during the many
steps in manufacture.

2. Once the system becomes operational, it is imperative

that provision be made for its continued maintenance
during its service.

56
Reliability Activities Throughout
Design

57
 Causes of Unreliability
• Design mistakes:
– Among the common design errors are failure to include all important
operator factors, incomplete information on loads and environmental
conditions, erroneous calculations, and poor selection of materials.
• Manufacturing defects:
– Poor surface finish, decarburization crack in heat-treated steel.
• Maintenance
• Exceeding design limits:
– Exceeding limits of temperature, speed, etc.
• Environmental factors:
– Subjecting equipment to environmental conditions for which it was not
designed.

58
 Minimizing Failure
• A variety of methods are used in engineering
design practice to improve reliability:
– Margin of safety
– Derating
– Redundancy
– Durability
– Damage tolerance
– Ease of Inspection
– Specificity
59
Distribution of Defect in
Engineering Components

60
Failure Mode & Effect
Analysis (FMEA)
 FMEA
• Failure mode and effects analysis (FMEA) is a team-
based methodology for identifying potential problems
with new or existing designs.

• FMEA was first used to identify and correct safety

hazards.

• FMEA identifies the mode of failure of every component

in a system and determines the effect on the system of
each potential failure.

62
Factors in Developing
FMEA

63
Rating for Severity of Failure

64
Rating of Occurrence of Failure

65
Rating of Detection of Failure

66
Results Of A FMEA Analysis

67
 Creating a FMEA Chart
1. The design is reviewed to determine the interrelations of
assemblies and the interrelations of the components of each
subassembly.
2. Now look more broadly, and ask what are the
consequences to the system of each failure identified in
step1.
3. For each of the functions, list the potential failure modes.
4. For each of the failure modes identifies, describe the
consequences or effect of the failure.
5. Using the severity of failure table, enter the numerical value.

68
 Creating a FMEA Chart (2)
6. Identify the possible causes of the failure mode.
7. Using the occurrence of failure table, enter a value for
the occurrence of the cause of each failure.
8. Determine how the potential failure will be detected.
9. Using Table 14.14, enter a rating that reflects the
ability to detect the cause of the failure identified in step
8.
10. Calculate the risk priority number (RPN).
11. For each potential failure, determine the corrective
action to remove a potential design, manufacturing, or
operational failure.
69
Fault Tree Analysis

What is fault tree analysis?

70
 Fault Tree Analysis
• Fault tree analysis (FTA) is a systematic method to
identify undesired events (faults) in a system.
• A fault is when a system does something it is not
supposed to do or does not do something it is supposed
to do.
• Often these faults are reliability or safety issues.
• Fault tree analysis starts with the top undesired event
and develops in a tree-like fashion all potential causes
for that event.
• FTA is the ability to identify combinations of events that
can affect the top undesired event.

71
Fault Tree For The Failure Of A
Lawn Mower Engine to Start

72
Defects and Failure Modes

What are typical defects and

failure modes in engineering
design?

73
 Defects and Failure Modes
• Failures of engineering designs and systems are a result of
deficiencies in four broad categories:
– Hardware failure:
• Failure of a component to function as designed
– Software failure:
• Failure of the computer software to function as designed
– Human failure:
• Failure of human operators to follow instructions or
respond adequately to emergency situations.
– Organizational failure:
• Failure of the organization to properly support the system.
74
 Causes of Hardware Failure
1. Design deficiencies

2. Deficiency in selection of material

3. Imperfection in material due to manufacturing

4. Improper testing or inspection

5. Overload and other abuses in service

6. Inadequate maintenance and repair

7. Environmental factors 75
 Failure Modes
• The specific modes of failure of
engineering components can usually be
grouped into four general classes:
– Excessive elastic deformation
– Excessive plastic deformation
– Fracture
– Loss of required part geometry through
corrosion or wear

76
Failure Modes for Mechanical
Components

77
Examples of Failure Modes in
Components

78
 Importance of Failure
• It is a human tendency to be reluctant to talk about failure or
to publish much information about failures.
• Spectacular system failures, like the Tacoma Narrows bridge
or the O-ring seal on the space shuttle Challenger solid rock
booster, have caught the public’s attention.
• While the literature on engineering failures is not extensive,
there are several useful books on the subject.

For information on conduction failure analysis see

Techniques for Failure Analysis at www.mhhe.com/dieter.

79
 Design for Safety

What are methods for applying

design for safety?

80
 Design For Safety
• Safety may well be the paramount issue in product
design.
• Normally we take safety for granted, but the recall of an
unsafe product can be very costly in terms of product
liability suits, replaced product, or tarnished reputation.
• The product must be safe to:
– Manufacture
– Use
– Dispose of after use.
• A safe product is one that does not cause injury or
property loss.
81
 Aspects To Design for Safety

1. Make the product safe, design all hazards out of the

product.
2. If it is not possible to make the product inherently safe,
then design in protective devices like guards, automatic
cutoff switches, and pressure-relief valves, to mitigate
the hazard.
3. If step 2 cannot remove all hazards, then warn the user
of the product with appropriate warnings like labels,
flashing lights, and loud sounds.
4. Provide training and protective clothing or devices to
the user or operator of the equipment.

82
 Fail-Safe Design
• A fail-safe design seeks to ensure that a failure will either not affect the
product or change it to a state in which no injury or damage will occur.
• There are three variants of fail-safe design:
– Fail-passive design:
• When a failure occurs, the system is reduced to its lowest-energy
state, and the product will not operate until corrective action is
taken.
– Fail-active design:
• When failure occurs, the system remains energized and in a safe
operating mode.
– Fail-operational design:
• The design is such that the device continues to provide its critical
function even though a part has failed.

83
 Potential Dangers
• We list here some of the general categories of
safety hazards that need to be considered in
design:
 Acceleration/deceleration  Leaks or spills
 Chemical contamination  Life cycle factors
 Electrical  Materials
 Environment  Mechanical
 Ergonomic  Physiological
 Explosions  Pressure/Vacuum
 Fire  Radiation
 Human factors  Structural
 Temperature
84
 Guidelines for Design for
Safety
1. Recognize and identify the actual or potential hazards, and
then design the product so they will not affect its
functioning.
2. Thoroughly test prototypes of the product to reveal any
hazards overlooked in the initial design.
3. Design the product so it is easier to use safely than to use
unsafely.
4. If field experience turns up a safety problem, determine the
root cause and redesign to eliminate the hazard.
5. Realize that humans will do foolish things, and allow for it
in your design

85
 Guidelines for Design for
Safety (2)
6. There is a close correspondence between good
ergonomic design and a safe design.
7. Minimize the use of flammable materials, including
packaging materials.
8. Paint and other surface finishing materials should be
chosen to comply with EPA and OSHA regulations for
toxicity to the user and for safety when they are burned,
recycled, or discarded.
9. Think about the need for repair, service, or
maintenance.
10. Electrical products should be properly grounded to
prevent shock.
86
 Safety Design Principles
5 principles (key elements) for safe design:
• Principle 1: People with control Safe design is
everyone’s responsibility – ensuring safe design
rests with all parties influencing the design of a
building or structure.
• Principle 2: The life cycle Safe design employs
life cycle concepts – applying to every phase in
the life cycle of a building or structure, from
conception through to redevelopment and
demolition.
87
 Safety Design Principles
5 principles (key elements) for safe design:
• Principle 3: Risk management Safe design
implements risk management – through
systematically identifying, assessing, and
controlling hazards.
• Principle 4: Knowledge and capability Safe
design requires knowledge and capability –
which should be either demonstrated or
accessed by any person influencing design.

88
 Safety Design Principles
5 principles (key elements) for safe design:
• Principle 5: Information transfer Safe design
relies on information – requiring effective
documentation and communication between
everyone involved in the life cycle of a building,
product or structure.

89
Integrating Design & Risk Management

90
Integrating Design & Risk Management

91
Integrating Design & Risk Management

92
 Safety Hierarchy

Hierarchy of hazard control is a system used in industry

to minimize or eliminate exposure to hazards. It is a widely
accepted system promoted by numerous safety
organizations. This concept is taught to managers in
industry, to be promoted as standard practice in the
workplace. Various illustrations are used to depict this
system, most commonly a triangle..

93
 Safety Hierarchy
"Safety Decision Hierarchy"
(Manuele, 2003)
• Eliminate hazards and risks
through system design and
redesign,
• Reduce risks by substituting less
hazardous methods or materials,
• Incorporate safety devices (fixed
guards, interlocks),
• Provide warning systems, Apply
administrative controls (work
methods, training, etc.), and
• Provide personal protective
equipment.

94
Safety Hierarchy

95
 Safety Hierarchy
Elimination
Eliminating the hazard—physically removing it—is the most effective hazard
control.[4] For example, if employees must work high above the ground, the hazard
can be eliminated by moving the piece they are working on to ground level to
eliminate the need to work at heights.

Substitution
This pesticide contains DDT, an effective substitution would be to
replace it with a green pesticide.
Substitution, the second most effective hazard control, involves
replacing something that produces a hazard (similar to
elimination) with something that does not produce a hazard—for
example, replacing lead based paint with acrylic paint. To be an
effective control, the new product must not produce another
hazard. Because airborne dust can be hazardous, if a product
can be purchased with a larger particle size, the smaller product
may effectively be substituted with the larger product.[4]
96
 Safety Hierarchy
Engineered controls
The third most effective means of controlling hazards is engineered controls.
These do not eliminate hazards, but rather isolate people from hazards.[2]
Capital costs of engineered controls tend to be higher than less effective controls
in the hierarchy, however they may reduce future costs.[5] For example, a crew
might build a work platform rather than purchase, replace, and maintain fall
arrest equipment. "Enclosure and isolation" creates a physical barrier between
personnel and hazards, such as using remotely controlled equipment. Fume
hoods can remove airborne contaminants as a means of engineered control.[4]

97
 Safety Hierarchy
Administrative controls
This sign warns people that there are explosives in
Walker Lake, however it cannot prevent people from
swimming in it.
Administrative controls are changes to the way people
work. Examples of administrative controls include
procedure changes, employee training, and installation
of signs and warning labels (such as those in the
Workplace Hazardous Materials Information System).[2]
Administrative controls do not remove hazards, but limit
or prevent people's exposure to the hazards, such as
completing road construction at night when fewer people
are driving.[4]

98
 Safety Hierarchy
Personal protective equipment
Main article: personal protective equipment
Personal protective equipment (PPE) includes gloves, respirators, hard hats,
safety glasses, high-visibility clothing, and safety footwear. PPE is the least
effective means of controlling hazards because of the high potential for damage
to render PPE ineffective.[4] Additionally, some PPE, such as respirators,
increase physiological effort to complete a task and, therefore, may require
medical examinations to ensure workers can use the PPE without risking their
health.

99
 Warning Labels
• With rapidly escalating costs of product liability,
manufacturers have responded by plastering their products
with warning labels.
• Warnings should supplement the safety related design
features by indicating how to avoid injury or damage from
the hazards that could not be feasibly designed out of the
product without seriously compromising its performance.
• Attention is achieved by printing:
– Danger
– Warning
– Caution
100
THE END