University of Engineering and Technology

Lahore

Assignment
DEPARTMENT OF MECHANICAL ENGINEERING 2024

SUBMITTED BY : Jazab Mustafa

REG NO. : 2022-ME-101
SECTION : C
SUBJECT : Applied Engineering Statistics
COURSE CODE : MA-241
SUBMITTED TO : Dr. Shabbir
 Application of Probability in Reliability Engineering

Reliability:
In statistics and, reliability is the overall consistency of a measure. A measure is said to
have a high reliability if it produces similar results under consistent conditions.
It can be defined as:
“Reliability is the probability of a device performing its purpose adequately for the period
of time intended under the operating conditions encountered.” [1]
This definition breaks down into four basic parts:
 Probability
 Adequate Performance
 Time
 Operating Conditions
The first part, probability, provides the numerical input for the assessment of reliability
and also the first index of system adequacy. In many instances it is the most significant
index, but there are many more parameters calculated and used.
The other three parts adequate performance, time and operating conditions are all
engineering parameters, and probability theory is of no assistance in this part of the
assessment. Often, only the engineer responsible for a particular system can satisfactorily
supply information relating to these. [2]

Reliability Evaluation:
The process of determining whether an existing system / entity has achieved a specified
level of operational reliability (desired, agreed upon or contracted behavior). [3]
Reliability evaluation encompasses several key aspects, including failure analysis,
reliability modeling, testing, data collection and analysis, risk assessment, and the use of
reliability metrics. It involves identifying potential failure modes, developing
mathematical models to quantify reliability, conducting tests to validate performance,
analyzing data to identify trends, assessing risks associated with failures, and using
quantitative metrics to measure reliability.
Reliability evaluation is essential across various industries, including manufacturing,
aerospace, automotive, electronics, and healthcare, where dependable performance and
minimal downtime are critical requirements. It helps organizations identify weaknesses,
optimize maintenance strategies, and design more reliable products and systems.

Reliability Logic Diagrams:

A reliability logic diagram (RLD) or reliability block diagram (RBD) is a diagrammatic
method for showing how component reliability contributes to the success or failure of a
redundant. RBD is also known as a dependence diagram (DD). [4]
An RBD is drawn as a series of blocks connected in parallel or series configuration.
Parallel blocks indicate redundant subsystems or components that contribute to a lower
failure rate. Each block represents a component of the system with a reliability rate.
RBDs will indicate the type of redundancy in the parallel path. For example, a group of
parallel blocks could require two out of three components to succeed for the system to
succeed. By contrast, any failure along a series path causes the entire series path to fail.
Examples:
An RBD of a simplified computer system with a redundant fan configuration is shown
below. [5]

Another example of a simple RBD model for the air conditioning of a room is shown
below. [6]

Engineering Systems:
An engineering system is a complex arrangement of interconnected components,
processes, and functions designed to achieve specific objectives or perform specific tasks
within the field of engineering.
In practice, a system is frequently represented as a network in which the system
components are connected together either in series, parallel, meshed or a combination of
these.

Classification:
There are three common types of Engineering Systems:
1. Series System
2. Parallel System
3. Series-Parallel System
1. Series System:
The components in a set are said to be in series from a reliability point of view if they
must all work for system success or only one needs to fail for system failure.
Consider a system consisting of two independent components A and B connected in
series, from a reliability point of view. This arrangement implies that both components
must work to ensure system success.

Two Component Series System

Let RA, RB = probability of successful operation of components A and B respectively, and

OA, OB = probability of failure of components A and B respectively. Since success and
failure are mutually exclusive and complementary,
R A + QB=1 and R B +QB =1

The requirement for system success is that 'both A and B' must be working. We know that
reliability of a series system is given as:
R s=R A . RB

If there are now n components in series, above equation can be generalized to give:
n
R s= ∏ R i
i=1

This equation frequently is referred to as the product rule of reliability since it establishes
that the reliability of a series system is the product of the individual component
reliabilities.
2. Parallel System:
The components in a set are said to be in parallel from a reliability point of view if only
one needs to be working for system success or all must fail for system failure.
Consider a system consisting of two independent components A and B, connected in
parallel, from a reliability point of view as shown in Figure below:

Two Component Parallel System

In this case the system requirement is that only one component need be working for
system success. The system reliability can be obtained as the complement of the system
unreliability or by parallel reliability equation since 'either A or B or both' constitutes
success to give:
R P=1−Q A .Q B

R P=R A + RB −R A . RB

or for an n component system:

n
R P=1−∏ Qi
i=1

3. Series Parallel System:

The series and parallel systems discussed in the two previous sections form the basis for
analyzing more complicated configurations. The general principle used is to reduce
sequentially the complicated configuration by combining appropriate series and parallel
branches of the reliability model until a single equivalent element remains. This
equivalent element then represents the reliability (or unreliability) of the original
configuration.
 Problem No. 01
Objective:
Determine the system’s reliability by applying the network reduction technique or otherwise.

Solution:
Step 1:
Combining Component 2 and 3 (Parallel Linkage):

Req =1−∏ (1−R j)

R2−3=1−[ ( 1−R 2) ( 1−R3 ) ]

R2−3=0.98

Step 2:

Req =∏ R i

R2−3−4 =(R 2−3 )∙ (R 4 )

R2−3−4 =(0.98)∙(0.8)
R2−3−4 =0.78 4

Step 3:

Req =1−∏ (1−R j)

R2−3−4 −5 =1−[ ( 1−R2−3−4 ) ( 1−R 5 ) ]

R2−3−4 −5 =1−[ ( 1−0.784 ) ∙ ( 0.8 ) ]

R2−3−4 −5 =0.956 8

Step 4:

Req =∏ R i

R1−2−3−4−5=(R2−3−4 −5 )∙(R1)
R1−2−3−4−5=0.8611

Step 5:

Req =1−∏ (1−R j)

R system=1−[ ( 1−R 1−2−3 −4−5 ) ∙ ( 1−R 6 ) ]

R system=0.986 1

F system=1−R system

F system =0.139 0

Conclusion:
We implemented network reduction strategy and calculated overall system efficiency for
given arrangement of reliability marked elements. The result gained with this arrangement
and ratings is exceptional i.e: 98.61% success or 1.390% failure probability which ensures
smooth workflow with minimal chances of unreliability incidents.

Problem No. 02
Objective:
Find a general formula for the model's reliability, which can be used to assess the system's
reliability if each of its parts has a reliability of 0.8.
Solution:
Step 1:
For components 1, 2, 3, 4 in the diagram, we can use series equivalency:

Req =∏ R i
R1−2−3−4 =R 1 ∙ R2 ∙ R3 ∙ R 4

Also for components 5, 6, 7 and 8, we get:

Req =∏ R i
R5−6−7−8 =R 5 ∙ R6 ∙ R 7 ∙ R8

Step 2:
As both R1−2−3−4 and R5−6−7−8 are in parallel, so:

Req =1−∏ (1−R j)

R system=1−[ ( 1−R 1−2−3 −4 ) ∙ ( 1−R 5−6−7−8 ) ]

[
R system=1− {1−( R1 ∙ R 2 ∙ R 3 ∙ R4 ) } ∙ {1−( R5 ∙ R 6 ∙ R7 ∙ R 8 ) } ]
As the reliabilities of all the components are the same, then:
R1=R 2=R3=R 4 ¿ R 5=R6=R 7=R8 =R=0.8

R system=1−[ ( 1−R ) ∙ ( 1−R )]

2
R system=1−[ ( 1−R ) ]
2
R system=1−[ ( 1−0.8 ) ]

R system=0.6514

F system =0.348 6

Conclusion:
The system’s reliability is calculated to be 65.14% which is a significantly lower value with
risk occurrence probability of 34.86% which indicates frequent inspection of system under
analysis if it is made to run as it is.