PUC 3103

Statistical and Probabilistic Models

Lecturer: Dr. Njenga Mburu

Contacts: Mob: 0728 777 573
Email: njmburu2002@yahoo.com

23rd May 2023

Venue: CE LAB/Online

Dr.Mburu-2023-Statistics-SET-1
 Course Content
Fundamentals of probability and statistics: random space, sample space,
probability measure. Random variables and distribution. Sample statistics
and their distributions.
Statistical inference and applications: parameter estimation techniques,
hypothesis testing.
Univariate probabilistic models: elements of frequency analysis, plotting
positions and probability papers, estimation of model parameters.
Statistical linear models (bivariate and multivariate): optimization methods,
stepwise regression analysis, time series analysis, time series analysis and
synthesis, autoregressive models, moving-average models.
Hydrological models and water quality models.

Dr.Mburu-2023-Statistics-1-23.5.23
 References
1. Probability and statistics for Engineers and Scientists (Ninth edition), R. E.
Walpole, R.H. Myers, S.L. Myers, K. Ye

2. Applied statistics and probability for Engineers (Fifth Edition), Douglas C.

Montgomery, George C. Runger

3. Applied multivariate statistical analysis, Richard A. Johnson, Dean W.

Wichern

4. Stochastic water resources technology, N. T. Kottegoda

5. Applied Linear Statistical Models 5th Edition, M. H. Kutner, C. J. Nachtsheim

J. Neter. W. Li
Dr.Mburu-2023-Statistics-1-23.5.23
Civil Engineering
• Civil engineering - oldest engineering
discipline - grouped into several specialty
areas namely:-
• Structural Engineering,
• Engineering and Management,
• Transportation Engineering,
• Water Resource Engineering,
• Environmental Engineering, and
• Geotechnical Engineering.

Dr.Mburu-2023-Statistics-1-23.5.23
Civil Engineering
• It deals with planning, analysis, design, construction, and maintenance of the physical
and naturally built environment.

• The analysis is carried out based on the collected and observed data
• Generally the statistics and probability approach is required whenever it is not possible
to directly measure all of the values (population) required.
• Example : If the traffic engineer needs to know the average speed of all vehicles on a
particular section of roadway, not all vehicles could be observed (others are unknown).
• Statistics and probability mathematics are used to estimate characteristics that cannot be
established with absolute certainty and to assess the degree of certainty that exists.

Dr.Mburu-2023-Statistics-1-23.5.23
What is the role of Statistical & Probability methods in Engineering?
 Uncertainty are unavoidable in any engineering analysis
and design, construction etc.

 Uncertainty: Lack knowledge or Information, have doubt:

about material parameters, initial conditions, boundary
conditions, outcomes [present or future] etc., during
design or construction etc.

 Uncertainty arise due to

 Incompleteness of available information/data [Sample Vs
Population]
 Natural processes and phenomena are inherently random
[Earthquakes, flooding etc.]

 However decisions are required even with the incomplete

information/ data and for the natural processes.

• Statistical and Probability theory provides a formal basis

for quantifying risk or uncertainty in engineering
problems[Engineering Judgement] for the Implementation
of Engineering Project
Dr.Mburu-2023-Statistics-1-23.5.23
 Role of probability methods in Engineering
The modelling of Engineering problems and evaluation of systems
performance under conditions of Uncertainty

Systematic development of design criteria, explicitly taking into account

the significance of uncertainty [Factor of Safety……standard codes/design
codes]

The logical framework for risk assessment and risk-benefit tread off
analysis relative to decision making [Very strong structure, Initial cost,
Maintenance cost , Benefits etc.]

Dr.Mburu-2023-Statistics-1-23.5.23
 Application in Civil Engineering

• Geotechnical Engineering
• Hydrology and Water Resources Engineering
• Structural Engineering
• Construction Planning and Management
• Transportation Engineering
• Environmental Engineering

Dr.Mburu-2023-Statistics-1-23.5.23
 Geotechnical Engineering
Uncertainties

Variable nature in rock-affect the load bearing capacity

 Heterogeneous soil properties and other in-situ conditions

Reliability of design and construction methods

Cost and benefits of proposed design strategies

Dr.Mburu-2023-Statistics-1-23.5.23
 Hydrology and Water Resources Engineering
Sources of Uncertainty
Incompleteness of historical data
Variability of hydrological data etc.

Estimation of Hydrological Extreme Events

Probable Maximum Precipitation (PMP)
Probable Maximum Flood (PMF)

Hydrological Prediction
Rainfall Prediction
Stream Flow Prediction

Design of Hydraulic Structures

Design Life
Risk Analysis
Dr.Mburu-2023-Statistics-1-23.5.23
 Structural Engineering

Sources of Uncertainty

Magnitude of load
Determination of maximum wind effect
Consideration of earthquake force

Strength of structural materials

Number of load cycles until fatigue

Failure can cause monetary loss, injury or death

Dr.Mburu-2023-Statistics-1-23.5.23
 Construction Planning and Management

Uncertainty
Duration of various activities in a construction project
Time of supply of material
Availability of required manpower
Weather Conditions

Estimation of Total Duration of the Project

Estimation of Cost Involved

Dr.Mburu-2023-Statistics-1-23.5.23
 Transportation Engineering
Events- Low Probability -High-risk
Accidents in air traffic
Accidents on highways

Design factors
Width of pavement
Thickness-subgrade and other layers
Slope

Pavement life depends on

Drainage and moisture conditions
Temperature range
Density and degree of compaction of subgrade
Dr.Mburu-2023-Statistics-1-23.5.23
 Definitions
Statistics and Probability are the branches of mathematics concerned
with the laws governing random events, including the collection,
analysis, interpretation, and display of numerical data from samples and
populations.

The quantities measured in a study (scientific experiments) are called

random variable; and a particular outcome is called an observation.

Experiment: A set of conditions under which behavior of some variable

is observed.

Dr.Mburu-2023-Statistics-1-23.5.23
 Definitions
Several observations are known as data. The collection of all
possible outcomes is called a population.

In practice, we cannot usually observe the whole population. Instead

we observe a sub-set of the population, known as a sample.

In order to ensure that the sample we take is a representative of the

whole population, we usually take a random sample.

By Random it means all members of the population are equally likely
to be selected for inclusion in the sample

Dr.Mburu-2023-Statistics-1-23.5.23
 Definitions
Model : Mathematical equation that translate a conceptual
understanding of a system or processes into quantitative terms

Probabilistic Models: Statistical Analysis tool that estimates, on the basis of past
(historical) data, the probability of an event occurring again (return period).

Probability Space: is a pair (Ω, p) consisting of a set Ω and a function p which

assigns to each subset A of Ω a real number p(A) in the interval [0,1].

The set Ω is called a sample space. The elements of Ω are called sample points
and the subsets A of Ω are called events.

The function P is called a probability function. For an event A, the real number
p(A) is called the probability of A
Dr.Mburu-2023-Statistics-1-23.5.23
 Random Variables
A random variable [x], is a variable whose possible values are numerical
outcomes of a random phenomenon. There are two types of random
variables, discrete and continuous

A discrete random variable is one whose possible values either constitute

a finite set or else can be listed in an infinite sequence (a list in which there
is a first element , Second element and so on); x1, x2, x3,x4……….
Examples: number of children in a family, the Monday afternoon class attendance,
the number of patients in a doctor's consultancy, the number of defective light bulbs
in a box of ten etc.
A random variable whose set of possible value is an entire interval of numbers is not
discrete e.g. x1≤ x2

Dr.Mburu-2023-Statistics-1-23.5.23
 Random Variables
A random variable X is said to be continuous if its set of possible
values is an entire interval of numbers i.e. for some A<B, any
number x between A and B is possible e.g. depth of a lake lying
between a minimum and a maximum

Example:
If a chemical sample is taken and pH determined, then X is a continuous random
variable because pH is always between 0 and 14 (0 ≤ pH ≤ 14).

If more is known about the compound selected for analysis, then the set of possible
value might be a subinterval of [0,14] such that we can have 5.5 ≤X ≤6.5, but X is still
continuous

Dr.Mburu-2023-Statistics-1-23.5.23
 Probability Distribution
• Suppose the variable x, is the depth of a lake at a randomly chosen point on the
surface and M is the maximum depth in meters such that any number in the interval
[0,M] is a possible value of x, if we “discretize” x, by measuring x to the nearest meter
then possible value are non-negative integers ≤ M.

• The resulting probability distribution of the depth can be represented by the histogram
below.
No of Times (Frequency)

x to the nearest
meter

Dr.Mburu-2023-Statistics-1-23.5.23 Depth of Lake

 Probability Distribution
• However, if the depth is measured to the nearest centimeter on the same
measurement axis, each rectangle in the resulting probability histogram
will be narrowing.
No of Times (Frequency)

No of Times (Frequency)
x to the
x to the nearest
nearest
meter
centimeter

Depth of Lake Depth of Lake

Dr.Mburu-2023-Statistics-1-23.5.23
 Probability Distribution
• If you continue to measure depth more and more finely, the resulting
histograms will approach a smooth curve.

• The probability that the depth at a randomly chosen point is between “a”
and “b” is simply equal to the area of the curve under the points “a” and
“b”.
Dr.Mburu-2023-Statistics-1-23.5.23
 P.d.f (Probability distribution Function or Density Function)

• P.d.f. of x is defined as a function f(x) such that for any two numbers
a and b with a ≤ b,
b
p(a  x  b)   f ( x)dx
a

• In order that f(x) be a legitimate p.d.f, it must satisfy two conditions

(1) f ( x)  0 for all x


(2)
 f ( x)dx  1  the area under the entire graph of f ( x)


Dr.Mburu-2023-Statistics-1-23.5.23
Example
• Suppose that I take a bus to work and the bus arrives at the bus stop every
five minutes. Because the time of leaving the house varies, the waiting
time also varies.

• Let the waiting time for the next bus be x (continuous random variable).
Set of possible values of x is [0,5]

1
• A possible p.d.f. for x ( ) = 5 for 0 ≤ ≤ 5
0 otherwise

• Determine (i) p(1≤ x ≤ 3) (ii) p (4 ≤ x)

Dr.Mburu-2023-Statistics-1-23.5.23