ASSIGNMENT OF APPLIED STATS(MA-241)

Submitted by: SAJJAD AHMAD

REG. (2022-ME-157)
Submitted to: SIR DR. SHABIR

DEPARTMENT OF MECHANICAL ENGINEERING,

UNIVERSITY OF ENGINEERING AND TECHNOLOGY,
LAHORE

1
 Problem Brief
 What is reliability evaluation of an engineering system?
 What is Reliability Logic Diagram (RLD) or Reliability Block Diagram (RBD)?
 Illustration with examples.
 Describe classification of engineering systems:
i. Series System
ii. Parallel System
iii. Series-Parallel System
 Derive a general mathematical formulae for system reliability in each case mentioned above.

Literature Review
 Abstract:
Many objects consist of more components. The mutual arrangement of the individual elements
influences the resultant reliability. The formulae are shown for the resultant reliability of series
arrangement, as well as for parallel and combined arrangement. The possibility of reliability
increasing by means of redundancy is explained, and also the principle of optimal allocation of
reliabilities to individual elements.

 Introduction:
Many processes are carried out as a system of numerous elements combined in different
manners. Resultantly, the reliability of system, that is, probability of failure free operation
depends on reliability marking of its elements. Depending on manner of arrangement of
elements, reliability may increase or decrease which will be illustrated in upcoming sections.

Reliability evaluation is an essential factor to calculate and consider in dealing with engineering
systems in order to predict risks and tackle them beforehand if possible.

For same purpose, a pictorial arrangement of elements with marked reliability is known as RLD
or RBD.

Here we will focus on reliability calculation of various sequential engineering systems and will
discuss measures to optimize system reliability.

2
  Element:

From reliability point of view, an element is any component or object that is considered in the
investigated case as a whole and is not decomposed into simpler objects.

 Reliability Evaluation:

Reliability evaluation of an engineering system involves assessing the likelihood of the system
performing its intended functions satisfactorily over a specified period under stated conditions.
It's a critical aspect of engineering design and operation, especially in fields like mechanical
engineering where system failure can have significant consequences.

 RLD or RBD:

A Reliability Block Diagram (RBD), also known as a Reliability Logic Diagram (RLD), is a
graphical representation used in reliability engineering to model the reliability of complex
systems. It depicts the system as a series of blocks, each representing a component or subsystem,
and the relationships between these blocks in terms of their reliability and dependencies. RBDs
are particularly useful for analysing the reliability of systems with multiple components and
paths.

Figure 1 RLD or RBD

3
  Classification of Engineering Systems:
 Series System:

From reliability point of view, a series system is one which fails if any of its sub-system or
element fails. It is represented as continuous connection in RLD indicating non-functionality of
any component will block the path to desired results.

Figure 2 Series System

 Parallel System:

From reliability point of view, a parallel system is one which only fails when all of its
components fail simultaneously. It is represented as side-by-side connection of elements or sub-
systems. It indicates that it is possible to achieve desired results regardless if any of any
component’s failure.

Figure 3 Parallel System

4
  Series-Parallel System:

This system is interconnection of members in a mixed manner of series and parallel connections
at various linkages. Such systems are resolved by assuming smaller simpler connections and
replacing them as a single component having combined effect. In this way they are gradually
simplified and solved as a whole system.

Figure 4 Series-Parallel System

 Derivation:
 Principle:

The principle in calculation of system’s reliability lies in the previously studied statistical fact
that probability of simultaneous occurrences of various independent events is equal to product of
independent probability of each event.

P(Ai,1 ∩ Ai,2 ∩ Ai,3 ∩ Ai,4 ∩ Ai,5 ∩ Ai,6 ∩ … Ai,n) =

P(Ai,1) . P(Ai,2) . P(Ai,3) . P(Ai,4) . P(Ai,5) . P(Ai,6) . … . P(Ai,n)

 In case of Series System Reliability, by definition, system will produce results if all parts
succeed. That is probability of simultaneous success of each element is to be determined.
 In case of Parallel System, by definition, system will fail if all elements fail
simultaneously. It means we will calculate probability of simultaneous failure of each
element which is complement of probability of failure free operation. The product of

5
 these failure probabilities will give equivalent failure probability of system, complement
of which will be Probability of failure free operation of parallel system.

 Series System:

Equivalent Reliability of Series system is calculated by considering following point:

 It will work given that all elements and sub-systems are functioning simultaneously.
 Probability of simultaneous occurrence of events is given as product of individual
occurrence probability.
 Hence Reliability Probability of series system is given as product of Reliability marking
of each element which is constituting the system.

Req = R1 * R2 * R3 * … * Rn = ∏ Rj
 Considering that reliability might depend on time in more realistic cases, so, in that cases,
the equation becomes:

Req(t) = R1(t) * R2(t) * R3(t) * … * Rn(t) = ∏ Rj(t)

 Knowing that Failure is Complement of reliability in this case. So:

F = 1-R
Feq = 1 - Req = 1 – (R1 * R2 * R3 * … * Rn) = 1- ∏ Rj
Feq = 1- ∏ ( 1- Fj)
 Considering Time dependence:

Feq(t) = 1- ∏ ( 1- Fj(t))

 Parallel System:

Equivalent Reliability of Series system is calculated by considering following point:

 It will fail given that all elements and sub-systems fail simultaneously.
 Probability of simultaneous occurrence of events is given as product of individual
occurrence probability.
 Hence Failure Probability of series system is given as product of Failure Probability of
each element which is complement of reliability marking of each of them.

Feq = F1 * F2 * F3 * … * Fn = ∏ Fj
Req = 1 – Feq = 1 - ∏ Fj
Req = 1 - ∏ (1 - Rj)

6
  Considering that reliability might depend on time in more realistic cases, so, in that cases,
the equation becomes:

Req(t) = 1 - ∏ (1 - Rj(t))

 Series-Parallel System:

As mentioned earlier, series-parallel systems are solved by identifying solicit parallel or series
connections and replacing them with equivalent simple component to gradually simplify the
system and solve it as a whole.

Problem 1
 Objective:
Determine the system’s reliability by applying the network reduction technique or otherwise.

Figure 5 Problem 1

 Solution:
Approach:

At first sight, the system seems rather complex and it isn’t suggested to try solving it from either
of its terminal ends. But if we look at component 2 and 3 they are definitely in parallel and their

7
combination will be in series with component 4 and so on. In this way we can easily evaluates
this system’s reliability by reducing its network gradually.

Step 1:

Combining Component 2 and 3 (Parallel Linkage):

For Parallel System:

Req = 1 - ∏ (1 - Rj)

R2-3 = 1 – {(1 – R2)*(1 – R3)}

R2-3 = 1 – {(1 – 0.8)*(1 – 0.9)}

R2-3 = 0.98

Step 2:

The reduced system calculations are below:

Now we can see that component 2-3 and component 4 are in series linkage:

For series:

Req = ∏ Rj

R2-3-4 = (R2-3) * (R4)

R2-3-4 = (0.98) * (0.8)
R2-3-4 = 0.784

Step 3:

The reduced system calculations are below:

Now Component 2-3-4 and Component 5 are in Parallel:

Req = 1 - ∏ (1 - Rj)

R2-34-5 = 1 – {(1 – R2-3-4)*(1 – R5)}

8
 R2-3 = 1 – {(1 – 0.784)*(1 – 0.8)}

R2-3-4-5 = 0.9568

Step 4:

The reduced system calculations are below:

Now, component 1 and component 2-3-4-5 are in series:

Req = ∏ Rj

R1-2-3-4-5 = (R2-3-4-5) * (R1)

R1-2-3-4-5 = (0.9568) * (0.9)
R1-2-3-4-5 = 0.86112

Step 4:

The reduced system calculations are below:

Now, component 1-2-3-4-5 and component 6 are in parallel:

Req = 1 - ∏ (1 - Rj)

Rsystem = 1 – {(1 – R1-2-3-4-5)*(1 – R6)}

Rsystem = 1 – {(1 – 0.86112)*(1 – 0.9)}

Rsystem = 0.986112

Fsystem = 0.013888

 Comments:
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: 98.6112% success or 1.3888% failure probability which ensures smooth
workflow with minimal chances of unreliability incidents.

9
 Problem 2
 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.

Figure 6 Problem 2

 Solution:
General Equation:

 Insights:

 It is a series-parallel system.
 Here Component 1,2,3,4 are in series and Component 5,6,7,8 are also in series.
 But considering them as a group they are parallel to each other.
 We will first solve both group individually in series linkage.
 Then we’ll calculate their equivalent parallel linkage reliability i.e: reliability of whole
system.

 Derivation:

For component 1,2,3,4 in series:

Req = ∏ Rj

R1-2-3-4 = (R1) * (R2) * (R3) * (R4) ___________ (i)

10
For component 5,6,7,8 in series:

Req = ∏ Rj

R5-6-7-8 = (R5) * (R6) * (R7) * (R8) ___________ (ii)

Now, for both R1-2-3-4 & R5-6-7-8 in parallel:

Req = 1 - ∏ (1 - Rj)

Rsystem = 1 – {(1 – R1-2-3-4)*(1 – R5-6-7-8)} ___________ (iii)

Substituting equation (i), (ii) in equation (iii):

Rsystem = 1 – [{1 – (R1 * R2 * R3 * R4)}*{1 – (R5 * R6 * R7 * R8)}] ___________ (iv)

Applying the condition that all elements have same reliability to equation (iv):

R 1 = R2 = R3 = R4 = R5 = R6 = R7 = R8 = R

Rsystem = 1 – [{1 – (R4)}*{1 – (R4}]

Rsystem = 1 – (1 - R4)2 ___________ (v)

Calculation:

Put R = 0.8 in equation 5 (Given)

Rsystem = 1 – (1 – (0.8)4)2

Rsystem = 1 – (0.5904)2

Rsystem = 0.6514

 Comments:
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.

11