Scribd
Upload
48 views

Basic Statistics

This document provides an overview of basic statistics and reliability concepts. It discusses analyzing sample data from an experiment using R software. Key concepts introduced include reliability, availability, maintainability, failure behavior of components and systems. Probability, conditional probability, independent and mutually exclusive events are also covered. The document concludes with an introduction to time to failure and time to repair analyses which are important in reliability engineering.

Uploaded by

Riccardo Viero
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
48 views

Basic Statistics

This document provides an overview of basic statistics and reliability concepts. It discusses analyzing sample data from an experiment using R software. Key concepts introduced include reliability, availability, maintainability, failure behavior of components and systems. Probability, conditional probability, independent and mutually exclusive events are also covered. The document concludes with an introduction to time to failure and time to repair analyses which are important in reliability engineering.

Uploaded by

Riccardo Viero
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 20

Lecture 7

Basic Statistics and Introduction to Reliability I

ddr. David Bogataj

Maintenance Management
Experiment from lecture 6
30 samples of 20 elements out of
population of 90 elements
➢s <- c(54.95, 45.55, 46.25, 49.15,
36.8, 52.6, 50.1, 46.95, 40.9, 46.25,
45.25, 36.75, 42.7, 38.35,
38.55, 42.7, 47.85, 48.85,
43.95, 48.25, 50.9, 55, 45.4,
40.25, 48.05, 40.6, 46.85,
46.1, 40, 49.2)
Distribution of sample averages
> mean(s) #average
• [1] 45.50167

> sd(s) #standard deviation


• [1] 4.970022
➢h<-hist(s, breaks=10, col="red", xlab="Sample
average", main="Histogram with Normal
Curve")
➢xfit<-seq(min(s),max(s),length=40)
➢yfit<-dnorm(xfit,mean=mean(s),sd=sd(s))
➢yfit <- yfit*diff(h$mids[1:2])*length(s)
➢lines(xfit, yfit, col="blue", lwd=2)
Introduction to reliability

The generic part also of very complex product can


be characterized by
- life cycle and
- failure behavior, but also by
- repair behavior in case of failure detection, and
- presence/absence of a maintenance strategy
➢replacement and/or
➢inspection or
➢preventive action.
RELIABILITY

Reliability is
➢the probability
➢that a component (or system)
➢will perform a required function
➢for a given period of time
➢when used under specific operating
conditions.
AVAILABILITY

Availability is
➢the probability
➢that a component (or system)
➢is performing its required function
➢at a given point in time
➢when used under specific operating
conditions.
MAINTAINABILITY
Maintainability is
➢the probability
➢that a failed component (or system)
➢will be restored (or repaired) to a specified
condition
➢within a period of time
➢when maintenance is carried out in
accordance with prescribed procedures.
Reliability
Reliability has become a very frequently used term during the last 15
years, not only used
➢ by engineers and practitioners but also
➢ by shop and superstore assistants who justify the price and
performance of a product by stressing quality, reliability, warranty,
and customer service if failures occur, etc.
This term is implicit in the thought processes of modern society, from
the housewife choosing a model of washing machine to the engineers
who design the product and guarantee its performance.
➢ In doing this, engineers also consider the implications of the
warranty and repair costs, a significant proportion of which is
composed of the spare parts management costs (i. e., fulfillment,
inventory management, replacement, etc.).
Reliability
The importance of measuring reliability is closely related to risk
determination and control:
➢ the generic risk event is related to the quantification of

▪ a probability to work properly (the reliability), and


simultaneously
▪ the magnitude of the consequences.

The importance of reliability also finds justification in the continuous


quality control and improvement of the products/services, process,
and production systems, and safety requirements and expectations:
the more complex the product is, the larger the number of laws and
regulations the product must comply with.
Components and Systems
in Reliability
The aim of reliability theory is to study

➢ the failure behavior of components, such as parts of a


production system, and
➢ the failure behavior of complex systems
in order to guarantee that they function correctly during a period
when they are in operation.

The production system is made of more than one part, which is in


turn composed of several components that perform various
functions. From the point of view of reliability, a component is a
generic entity (e. g., a tool, a machine, an item of equipment, a part
of the equipment) whose failure behavior is known and can be
modeled accurately by evaluating a pool of statistical parameters.
Reliability system
A reliability system is an entity whose
➢ failure and/or repair behaviors are not known and
➢ whose complexity usually requires one to adopt effective
models to support production system reliability evaluation
to be based on the basic reliability and maintainability
parameters of the components in the system.

A part of a production system is


▪ a component when its reliability parameterization is well
known, but it is
▪ a system when a reliability evaluation and prediction
analysis has to be conducted with its components’ basic
failure and repair behaviors and parameters.
Basic Statistics in Reliability
Engineering
• A failure or a repair can be described as a random
event. (theoretically: as result of experiment in
complex of conditions).
• A random event A can be characterized by the
probability of the event occurring.
• The probability p(A) is the likelihood or chance that
A is either the case or will happen in the future.
• It is represented by a real number ranging from 0
to 1.
p(A) generally refers to a period of time T as follows:
𝒏𝑨
p(A)=
𝒏

• 𝒏𝑨 is the number of occurrences (chances) of


event A in a period of time T and
• 𝒏 is the number of all occurrences (chances) in
T.
• Event A is a set of outcomes (a subset) to
which a probability p(A) is assigned.
• Two main properties of random events:
p(A)+p(𝐀 ഥ )=1 p(0)=0⤆ probability that nothing will happen
is 0

a set without elements


A random process, sometimes called a “stochastic
process,” is the counterpart in probability theory to a
deterministic process and deterministic system.

Reliability theory mainly refers to stochastic processes


and to the basic statistics briefly introduced and
discussed here to demonstrate the proposed and applied
reliability and maintenance analytical models, which are
the subject of these lectures.
From algebra of events
➢ The conditional probability is the probability of an event A
occurring given the occurrence of another event B, as follows:
𝑝(𝐴∩𝐵)
𝑝(A│B) = - probability that both events happen if B happens.
𝑝(𝐵)
where 𝐴 ∩ 𝐵 is the intersection of events A and B (both event
happen) .
➢ Consequently, 𝑝(𝐴 ∩ 𝐵) = 𝑝(B).𝑝(A│B)
➢ A and B are statistically independent in the case where:
𝑝(A│B)= 𝑝(A) ⟹ 𝑝(𝐴 ∩ 𝐵) = 𝑝(B).𝑝(A)
➢ Considering three statistically independent events
𝑝(𝐴 ∩ 𝐵 ∩ 𝐶) = 𝑝(A).𝑝(B).𝑝(C)
➢ Two events are mutually (or statistically) exclusive in
the case of 𝑝(𝐴 ∩ 𝐵) =0
➢ Another useful property in probability analysis and reliability
evaluation is the probability of the union of events:

𝒑(𝑨 ∪ 𝑩) = 𝒑(A)+𝒑(B)- 𝒑(𝑨 ∩ 𝑩)

where 𝐴 ∪ 𝐵 is the union of events A and B.

➢ Now considering three independent events A, B, and C,


𝒑(𝑨 ∪ 𝑩 ∪ 𝑪) = 𝒑(A)+𝒑(B) +𝒑(C)- 𝒑(A).𝒑(B)-𝒑(A). 𝒑(C)- 𝒑(B).𝒑(C) +
𝒑(A).𝒑(B).𝒑(C)
𝒑(𝑨 ∪ 𝑩 ∪ 𝑪) = 𝒑(A)+𝒑(B) +𝒑(C)

➢ In the case where the events are mutually exclusive:


∪ 𝑨𝒊 ) = σ𝒊 𝒑(𝑨𝒊 )
𝒑(ณ
𝒊
Time to Failure and Time to Repair

Component (system) subject to failure and repair events

You might also like