Probabilistic Risk Assessment
- Bayesian Estimation -
Centre for Risk, Integrity, and Safety Engineering (C-RISE)
Course instructor:
Dr. Mohammed Taleb-Berrouane | PhD in Process Safety Engineering
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
Weibull model
• A flexible distribution that is used widely in reliability is
the Weibull distribution.
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
2
Effect of the shape parameter
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
3
Effect of the shape parameter
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
4
Effect of the shape parameter
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
5
Effects of scale parameter
(characteristic life parameter)
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
6
Effects of scale parameter
(characteristic life parameter)
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
7
Event frequency: Poisson model
• Events occur over a continuum (time, space) at an
average constant rate, λ.
• The occurrence of an event in a given interval is assumed
to be independent of that in any other non-overlapping
interval.
• This distribution is used to describe the occurrence of
initiating events (IEs) in risk assessment.
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
8
Event frequency: Poisson model
• The Poisson distribution gives the probability of exactly
k events occurring in (0,t):
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
9
Bayesian Estimation
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
10
Bayesian estimation
• Bayesian parameter estimation is comprised of two
main steps:
1. The first step involves using available information to
fit a prior distribution to a parameter, such as a
failure rate.
2. The second step of Bayesian estimation involves
incorporating additional or new data to update the
prior distribution. This gives a posterior distribution,
which better represents the parameter of interest. This
step is often referred to as “Bayesian Updating.”
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
11
BE for a continuous variable
• The Bayes’ Theorem for continuous PDF, estimates the
parameter for posterior PDF f(θ/t) using the following
relationships:
• Here θ be a parameter of interest, h(θ) be a continuous
prior PDF and l(t/θ) be the likelihood function based on
sample data t.
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
12
BE for a discrete variable
• For a discrete PMF, the Bayes’ Theorem can be written
as:
• Pr(θi/ε): the conditional probability of θi given ε, or the
posterior probability for θi ;
• Pr(θi) = prior probability (can be expressed using a
PMF): and
• Pr(ε/θi) = the probability of obtaining the new information
(ε) given a certain value (θi) for the parameter.
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
13
BE for a discrete variable
• The following notation for the posterior distribution is
also common:
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
14
Prior distributions
• Prior distributions can be specified in different forms.
Possible forms include:
A. Parametric (gamma, lognormal, beta):
• Gamma or lognormal for rates of events
• Beta, truncated lognormal for event probabilities per
demand
B. Numerical (histogram, CDF values/percentiles)
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
15
Selection of Likelihood function
• Poisson Process: The Poisson distribution is the proper
likelihood function
• Bernoulli Process: The Binomial distribution is the
proper likelihood function:
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
16
Conjugate prior distribution
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
17
Any questions?
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
18
Useful software!
• GeNie Software: https://www.bayesfusion.com/download
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
19
Announcements
• PRA needs practice!
• Feedback about the midterm exam…
• Assignment 3.
CENTRE FOR RISK, INTEGRITY AND SAFETY ENGINEERING
WWW.MUN.CA/ENGINEERING/RESEARCH/CRISE
20
