Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
113 views

Probability and Random Variables

This document provides an introduction to probability and random variables. It begins with a list of references on probability and statistics. It then defines what statistics are, both in general and scientific terms. It discusses why statistics are needed, providing examples of how statistics are used in everyday life and scientific inquiry. It also introduces some key concepts in probability, distinguishing it from possibility. It provides examples of random variables that may be encountered in structural engineering applications. Finally, it discusses applications of probability and statistics to areas like hazard analysis and earthquake prediction.

Uploaded by

dinal031
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
113 views

Probability and Random Variables

This document provides an introduction to probability and random variables. It begins with a list of references on probability and statistics. It then defines what statistics are, both in general and scientific terms. It discusses why statistics are needed, providing examples of how statistics are used in everyday life and scientific inquiry. It also introduces some key concepts in probability, distinguishing it from possibility. It provides examples of random variables that may be encountered in structural engineering applications. Finally, it discusses applications of probability and statistics to areas like hazard analysis and earthquake prediction.

Uploaded by

dinal031
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 68

PROBABILITY AND

RANDOM VARIABLES
Ashar Saputra, PhD

List of Refferences
Johnson, R.A., 2005, Probability and Statistics for

Engineers, Pearson Prenstice Hall, New Jersey, USA


Walpole, R.E., Myers, R.H., Myers, S.L., Ye, K., 2007,
Probablisty and Statistics for Engineers and Scientists,
Pearson Prenstice Hall, New Jersey, USA
Madsen, H.O., Krenk, S., Lind, N.C., 1986, Methods of
Structural Safety, Prenstice Hall, New Jersey, USA
Nowak A.S., Collins K.R., 2000, Reliability of Structures,
McGraw-Hill Int, Singapore

Introduction
What is statistics?

General meaning

Collection of facts or data numerical results


A branch of knowledge

Scientific meaning the science of collection,


condensation, analysis, interpretation
Objectives estimation, prediction, and decision
making
Risks? Errors might occur (should be accepted)

statistical methods enable to measure or control such


error

Refreshments

Why we need
statistics

Why use statistics?

Universality of variation in all areas


Estimation, prediction, decision making, etc.
(statistics is the science of making decisions in the
face of uncertainty)

Statistics, as a subject, provides a body of principles and


methodology for designing the process of data
collection, summarizing and interpreting the data, and
drawing conclusions or generalities.

Statistics in our everyday life?


In our attempt to understand issues of many aspects, numerical
facts and figures need to be reviewed and interpreted. In our day
to day life, learning takes place through an often implicit analysis
of factual information.

Statistics in aid of scientific inquiry?


Sciencetific inquiry refers to a systematic process of learning. A
researcher sets the goal of an investigation, collects relevant
factual information (or data), analysis the data, draws
conclusions, and decides further courses of action.

Introduction
Uncertainties in the building process
Natural cuses of uncertainty result from the upredictability of
loads such as wind, earthquake, snow, water pressure, or
live load. Other is material used.
Human causes included intended or unintended departures
from design. Examples of uncertainty; approximation,
calculation errors, communicatin problems, omission, etc.
Also inadaquate material, methods of construction, bad
connectins, or change without analysis.

Refreshments
Statistical cases in the field of structural engineering;
Earthquake occurance
Live load of classrooms per square meter
Measurements of dimension of beams or columns
Concrete compressive strength taken from 200
concrete mixers
Nominal diameter or steel reinforcements
Weight of truck passing a bridge
Wind speed on certain area
Properties of organic material
etc

Refreshments (LRFD)
Because of these uncertainties, load and resistance (i.e.

Load-carrying capacities of structural elements) are


random variables.
It is convinient to consider a random parameter (load or
resistance) as a function of three factors;
Physical variation factor. Example include a natural variation

of wind pressure, earthquake, live load, and material


properties.
Statistical variation factor. This is due to limited sample size.
In most situation the natural variation is unknown and it is
quantified by examining limited sample data. Therefore, the
larger the sample size, the samller uncertainty described by
statistical variation factor.
Model variation factor. This is uncertainty due to simplifying
assumption, unknown boundary conditions, and unknown
effects of other variables

PROBABILITY

Probability v.s. Possibility


(Oxford Dictionary)

probability
n. (pl. probabilities) the extent to which something is probable. >
a probable or most probable event.
PHRASES in all probability most probably.
probabilistic
adj. based on or adapted to a theory of probability; involving
chance variation.
possibility
n. (pl. possibilities)
1 a thing that is possible.
2 the state or fact of being possible.
3 (possibilities) unspecified qualities of a promising nature.

example

1/16
b) 3/52
c) 3/51
d) 1/4
a)

answer

Example

a)
b)

c)
d)
e)

1/6
1/3

2/3
5/6

answer

Bayes Theorem

APPLICATION

Hazard Analysis
An important application area of probability and statistics

is the assessment of natural and man-made risks


Evaluation of safety of an engineering facility against
extreme environmental actions, such as earthquakes,
strong winds, extreme floods, ocean waves, etc.
Since environmental loads vary in time, one usually
expresses reliability through the probability that some
undesirable failure event (severe structural damage or
collapse, levee breach, dam overtopping, ship hull
buckling, etc.) occurs at least once during a reference
period of time T, for example 50 or 100 years.

Hazard Analysis
In order to quantify risk, one needs to combine two elements:
1. A description of the severity of the environment, in terms of the
probability P[LT > l] with which the maximum environmental load in T
years, LT, exceeds various levels l. Evaluating P[LT > l] is often
referred to as hazard assessment;
2. A description of the resistance of the facility in terms of the
dependence of the probability of system failure Pf on the magnitude
l of the environmental load. This function, Pf(l), is often referred to
as the fragility function.
Once quantified, the hazard and fragility functions are combined to
produce the probability of (at least one) failure in T. This is done by
using the Total Probability Theorem, which says that, if {B1,..., Bn} is
a set of mutually exclusive and collectively exhaustive events and A is
any other event, then the probability of A can be calculated as

Hazard Analysis
To use this result for risk assessment, the environmental
load LT is discretized into n distinct levels, say l1, ..., ln and the

events A and Bi in Eq. 1 are taken as

It is often reasonable to assume that the facility of interest

survives in T if it does not fail under the most intense load LT


experienced during that period. Under such simplifying
assumption and with the notation in Eq. 2, the probability of
failure in T is obtained from Eq. 1 as

Eq. 3 shows how the hazard (probabilities P[LT= li]) and the
fragility (the probabilities P[failure|LT= li]) are combined in the

assessment of risk.

Hazard Analysis
Example. In a recent study, the seismic hazard in Boston has been assessed as

follows in terms of Modified Mercalli Intensity or MMI (MMI is a discrete scale of


ground motion intensity, with integer values from 1 to 12). Over a period of 100
years, the probability that the maximum MMI value equals I is

Hazard Analysis
In a separate study, the seismic fragility of various types of structures was

assessed by a group of engineers. Some of their results are reproduced


below in the form of values of the probability of failure for different MMI.

EARTHQUAKE PREDICTION FROM IMPERFECT


PREMONITORY SIGNS
The main issue that determines the practical usefulness

of these premonitory events is the accuracy with which


predictions can be made. Accuracy can be quantified in
terms of the following probabilities

EARTHQUAKE PREDICTION
The probabilities in Eq. 1 depend on the strength of the association

between the premonitory event and the occurrence of earthquakes.


Unfortunately, this association is often weak. To make a quantitative
analysis, define the following events:

EARTHQUAKE PREDICTION
A typical daily probability for a major earthquake might be P[E] = 10-5 (hence

P[EC] = 1-10-5), meaning that at a given location large earthquakes might


occur on average every about 300 years. Also, a typical association between
a premonitory event A and large earthquakes might be

Notice that, for prediction purposes, one looks at the probability of A in the 24

hours that preceed a major earthquake. Applying Bayes theorem to this case
gives

EARTHQUAKE PREDICTION
Notice that the probability of a major earthquake, which for a generic day is 10-5,

increases 100-fold, to 10-3, after observation of the premonitory event A. After all,
only once every 1000 such warnings, an earthquake would actually occur. Also
consider that issuing false earthquake warnings is very costly and that, following
two or three such erroneous calls, people would lose confidence in the predictions.
Advocates of earthquake warning have observed that, although single premonitory
signs are seldom useful as a basis for issuing warnings, the use of several such
signs in combination may lead to more accurate predictions. Suppose for example
that two diagnostic events, A and B, are being monitored. For simplicity, assume
that, when taken in isolation, A and B have the same association with E, i.e.

EARTHQUAKE PREDICTION

END OF PROBABILITY
SESSION

RANDOM VARIABLES

example

You might also like