Course Title: Reliability

Course No: IPE- 434 Engineering and Maintenance

Management

Assignment on:
The Exponential Distribution

Submitted To: Submitted By:

Chowdury Md Luthfur Rahman Name: Nadim Md Maksud

Associate Professor Reg. No: 2018334027
Dept. of IPE, SUST Session: 2018-19
Date of Submission: 02-11-2023

Department of Industrial and Production Engineering

Shahjalal University of Science and Technology, Sylhet

 Contents
Introduction .....................................................................3
Reliability function: ........................................................3
Numerical 1 .....................................................................7
Numerical 2 .....................................................................7
Advantages ......................................................................8
Applications .....................................................................8
Conclusion .......................................................................8
 The Exponential Distribution

Introduction: The exponential, or strictly the negative exponential, distribution is probably

the most widely known and used distribution in reliability evaluation of systems. The most
important factor for it to be applicable is that the hazard rate should be constant, in which case it
is defined as the failure rate A. This is essentially the same requirement as that of the Poisson
distribution and it can be argued that the negative exponential distribution is only a special case of
the Poisson distribution, i.e., when considering the probability of the first failure. In practice the
negative exponential distribution has a much wider degree of significance than just that of first
failure and is extensively used in the analysis of repairable systems in which the components cycle
between operating or up states and failure or down states.
The negative exponential distribution is frequently used in system reliability evaluations, even
when the constant failure rate assumption is not confirmed.
Three justifications for this practice are:
➢ First, analytical techniques, especially for large systems, are highly complex unless
simplifications are employed. Assuming a constant failure rate and using the
exponential distribution simplifies the problem.
➢ Second, often, the data available for evaluation is limited and insufficient to confirm
the correct underlying distribution. Therefore, it's considered unrealistic to use a more
complex technique.
➢ Third, when the primary concern is limiting state values of system probability, the
underlying distribution becomes less significant, and the results are essentially the
same, regardless of the distribution used.
However, it's important to note that the third justification may not apply when dealing with time-
dependent values of probability, as the choice of distribution can significantly impact system
probability values in such cases.

Reliability function:
The following equations and mathematical relationships between the various reliability functions
do not assume any specific form of the functions and are equally applicable to all probability
distributions used in reliability evaluation.
Consider the case in which a fixed number of identical components are tested.
Let Ns(t) = number surviving at time t
Nf(t) = number failed at time t
i.e., Ns (t) + Nf (t) = No
At any time, t, the reliability, or survivor function R(t) is given by.
𝑵𝒔 (𝒕)
R(t) =
𝐍𝐨
(1)
𝐍𝐨−𝐍𝐟 (𝐭)
=
𝐍𝐨
𝐍𝐟 (𝐭)
= 1-
𝐍𝐨
(2)
Similarly, the probability of failure or cumulative failure distribution Q(t) is.
𝐍𝐟 (𝐭)
Q(t) =
𝐍𝐨
(3)
which follows the previous relationship given in Equation R(t) = 1-Q(t)

From Equations(2) and (3)

𝐝𝐑(𝐭) −𝐝𝐐(𝐭) −𝟏 𝐝𝐍𝐟 (𝐭)

= = .
𝐝𝐭 𝐝𝐭 𝐍𝐨 𝐝𝐭
As dt 0,
𝟏 𝒅𝑵𝒇 (𝒕)
f(t) = 𝑵𝒐 . 𝒅𝒕
at t = 0 ; failure density function = hazard rate
at t = 1 ; Number of components exposed to failure = No
Now,
No 𝑜𝑓 𝑓𝑎𝑙𝑢𝑟𝑒 𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 𝑡𝑖𝑚𝑒
Hazard rate λ =
No 𝑜𝑓 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡𝑠 𝑒𝑥𝑝𝑜𝑠𝑒𝑑 𝑡𝑜 𝑓𝑎𝑖𝑙𝑢𝑟𝑒

So. General expression of hazard rate at time t will be:

𝟏 𝐝𝐍𝐟 (𝐭)
λ(t) = .
𝐍𝐬 (𝐭) 𝐝𝐭

𝑵𝒐 𝟏 𝐝𝐍𝐟 (𝐭)
= . .
𝑵𝒐 𝐍𝐬 (𝐭) 𝐝𝐭

𝑵𝒐 𝟏 𝐝𝐍𝐟 (𝐭)
= .
𝐍𝐬 (𝐭) 𝑵𝒐 𝐝𝐭

𝟏 𝐟(𝐭)
= . f(t) = 𝑹(𝒕) = λ (4)
𝑹(𝒕)
We know,
𝒅𝑸(𝒕) 𝐝𝐑(𝐭)
fd(t) = =−
𝒅𝒕 𝐝𝐭
𝟏 𝐝𝐑(𝐭)
λ(t) = − 𝑹(𝒕) . (5)
𝐝𝐭
Equation 4 is an interesting equation for several reasons. It confirms the comment made above that
λ (0) = f(0) since at t = 0, R(0) = 1.
In addition, it shows that the hazard rate is a conditional function of the failure density function,
the conditional relationship being the survivor function. In physical terms this relationship means
that the failure density function permits the probability of failure to be evaluated in any period into
the future whereas the hazard rate permits the probability of failure to be evaluated in the next
period of time given that it has survived up to time t. Mathematically the relationship between λ
(t) and f(t) can be described as follows.
The hazard rate is equivalent to the failure density function but covers only those times up to the
time point of interest. Since the area under this equivalent density function is less than unity, the
portion of the failure density function being considered must be normalized back to unity. This can
be achieved by dividing the failure density function by the area under the failure density function
for times greater than that of interest, that is,
𝐟(𝐭)
λ(t) = ∞ For times up to t
∫𝒕 𝐟(𝐭) 𝐝𝐭
𝐟(𝐭)
= 𝑹(𝒕) From Eq (4)

Now from equation 5

𝑹(𝒕) 𝟏 𝒕
∫𝟏 𝐝𝐑(𝐭) = ∫𝟎 −𝛌(𝐭)𝐝𝐭
𝑹(𝒕)
𝒕
ln 𝑹(𝒕) = ∫𝟎 −𝛌(𝐭)𝐝𝐭
𝒕
𝑹(𝒕) = 𝒆 − ∫𝟎 𝛌(𝐭)𝐝𝐭
(6)
For the special case in which A is a constant and independent of time, Equation (6) simplifies to

𝑹(𝒕) = 𝒆−𝛌(𝐭)
This special case is known as the exponential distribution.
Now the failure density function is,
𝐝𝐑(𝐭)
f(t) = − 𝐝𝐭

= 𝛌 𝒆−𝛌(𝐭)
And cumulative failure distribution,

Q(t) = 1- 𝑹(𝒕) = 𝟏 − 𝒆−𝛌(𝐭)

Exponential reliability functions, the failure density function f(t), cumulative failure distribution Q(t) and
hazard rate λ(t) are shown in Figures a, b, c and d, respectively.

Fig: Exponential reliability functions. (a) Areas showing Q(t) and R(t). (b)
Failure density function. (c) Cumulative failure distribution. (d) Hazard rate
𝛌(𝐭)
 Numerical 1
It is observed that the failure pattern of an electronic system follows an exponential distribution
with mean time to failure of 1000 hours. What is the probability that the system failure occurs
within 750 hours?
Solution

Since MTTF =1000 hours

1/ 𝛌 = 1000
We know,

Probability of failure Q(t) = 1- 𝑅(𝑡) = 1 − 𝑒 −λ(t)

So, the probability of failure within 750 hours is

Q(1000) = 1- 𝑒 − 750/1000
= 1- 𝑒 − 0.75
= 0.528 (ANS)

Numerical 2
A computer has a constant error rate of one error every 17 days of continuous operation.
What is the reliability associated with the computer to correctly solve a problem that
requires 5 hours’ time? Find the hazard rate after 5 hours of operation.
Solution

Given, MTTF = θ = 408 hours

So, λ = 1/ θ = 0.0024 failure/hour
Advantages
This is probably the most important distribution in reliability work and is used almost
exclusively for reliability prediction of electronic equipment. It describes the situation wherein
the hazard rate is constant which can be shown to be generated by a Poisson process. This
distribution is valuable if properly used. It has the advantages of:
1. single, easily estimated parameter (λ)
2. mathematically very tractable
3. fairly wide applicability
4. is additive that is, the sum of a number of independent exponentially distributed
variables is exponentially distributed.

Applications
Some particular applications of this model include:
1. items whose failure rate does not change significantly with age.
2. complex and repairable equipment without excessive amounts of redundancy.
3. equipment for which the early failures or “infant mortalities” have been eliminated by
“burning in” the equipment for some reasonable time period.

Conclusion
In conclusion, the exponential distribution reliability function is a valuable tool in reliability
analysis, especially for systems with a constant failure rate. However, it's important to note that
very few real-life items actually exhibit a constant hazard rate. Therefore, while the exponential
distribution is a convenient choice, it's important to validate the assumption of a constant failure
rate. The function can be used to calculate the probability of success (reliability) over a specific
period, but other models might provide more accurate representations depending on the system or
component being analyzed. Therefore, the exponential distribution should be used judiciously,
taking into consideration the nature of the system or component