ME 571 Reliability in Engineering Design

Module 2.1: Probability Rules

Ganesh Subbarayan
Professor, School of Mechanical Engineering
Purdue University
 Overview
• Quantitative reliability relies on math of probability and
statistics
• Statistics covered will include
– Continuous functions, often functions of time
• Fraction failed F(t), Reliability R(t), Hazard rate h(t)
– Discrete functions, two-state discrete systems
• Pass/fail, working/not-working
• Statistical methods provide the means for analyzing,
understanding and controlling variation

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 2

 Probability Definition #1
• If an event can occur in 𝑁 equally likely ways, and if the
event with attribute 𝐴 can happen in 𝑛 of these ways, then
the probability of 𝐴 occurring is

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 3

 Probability Definition #2
• If, in an experiment, an event with attribute 𝐴 occurs 𝑛
times out of 𝑁 experiments, then as 𝑁 becomes large, the
probability of event 𝐴 approaches 𝑛/𝑁, i.e.,

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 4

 Influence of Sample Selection

Degree of confidence on the statistics of sample depends on the size

of sample!
G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 5
 Basic Rules and Notations
• The probability of obtaining an outcome 𝐴 is denoted by
𝑃(𝐴), and so on for other outcomes
• The joint probability that 𝐴 and 𝐵 occur is denoted by
𝑃(𝐴𝐵) or 𝑃(𝐴 ∩ 𝐵)
• The probability that 𝐴 or 𝐵 occurs is denoted by 𝑃(𝐴 + 𝐵)
or 𝑃(𝐴 ∪ 𝐵)
• The conditional probability of obtaining outcome 𝐴, given
that 𝐵 has occurred, is denoted by 𝑃(𝐴/𝐵)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 6

 Definition of Complement
• If the probability of obtaining an outcome 𝐴 is denoted by
𝑃(𝐴) then, its complement, i.e., of 𝐴 not occurring is

𝑃 𝐴ҧ = 1 − 𝑃(𝐴)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 7

 Product Rule
• The joint probability that 𝐴 and 𝐵 occur is denoted by
𝑃(𝐴𝐵) or 𝑃(𝐴 ∩ 𝐵)
• If the events 𝐴 and 𝐵 are independent (occurrence of 𝐴
does not influence occurrence of 𝐵) then

𝑃 𝐴𝐵 = 𝑃(𝐴 ∩ 𝐵) = 𝑃 𝐴 𝑃(𝐵)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 8

 Addition Rule

B
A

𝑃 𝐴 + 𝐵 = 𝑃 𝐴 ∪ 𝐵 = 𝑃 𝐴 + 𝑃 𝐵 − 𝑃(𝐴 ∩ 𝐵)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 9

• If A and B are independent events

𝑃 𝐴 ∪ 𝐵 = 𝑃 𝐴 + 𝑃 𝐵 − 𝑃 𝐴 𝑃(𝐵)

• If A and B are mutually exclusive events

𝑃 𝐴∩𝐵 = 0
𝑃 𝐴 ∪ 𝐵 = 𝑃 𝐴 + 𝑃(𝐵)
G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 10
 DeMorgan’s Law

B
A

𝑃 𝐴ҧ ∩ 𝐵ത = 𝑃 𝐴 ∪ 𝐵 = 1 − 𝑃(𝐴 ∪ 𝐵)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 11

 Addition Rule

B
A

𝑃 𝐴∪𝐵∪𝐶 =𝑃 𝐴 +𝑃 𝐵 +𝑃 𝐶
−𝑃 𝐴 ∩ 𝐵 − 𝑃 𝐵 ∩ 𝐶 − 𝑃 𝐶 ∩ 𝐴
+𝑃(𝐴 ∩ 𝐵 ∩ 𝐶)

G. Subbarayan, Purdue Univ. ME 571 Reliability in Engineering Design 12