Operations Research

July 2019
University of Hargeisa

1
 Introduction
• Since the advent of the industrial revolution,
the world has seen a remarkable growth in
the size and complexity of organizations.

• Increase in division of labor

• Increasing specialization has created new
problems.

2
 Cont..
• It becomes more and more difficult to allocate
the available resources to the various
activities.

• In a way that is most effective for the

organization as a whole.

3
 Cont…
• Operations research is applied to problems that
concern how to conduct and coordinate the
operations (i.e., the activities).
• The research part of the name means that operations
research uses an approach that resembles the way
research is conducted in established scientific
fields.

• To a considerable extent, the scientific method is

used to investigate the problem of concern.
4
 Cont…
• Operations Research is an Art and Science

• It had its early roots in World War II and is

flourishing in business and industry.

5
 Application Areas
• Strategic planning
• Supply chain management
• Pricing and revenue management
• Logistics
• Optimization
• Marketing

6
 Applications Areas (cont.)
• Scheduling
• Portfolio management
• Inventory analysis
• Forecasting
• Sales analysis
• Auctioning
• Risk analysis

7
 Cont…
• Operations Research is the scientific
approach to execute decision making, which
consists of:
– The art of mathematical modeling of
complex situations.
– The science of the development of solution
techniques used to solve these models.
– The ability to effectively communicate the
results to the decision maker.
8
 What Do We do
1. OR professionals aim to provide rational bases for decision
making.

2. Much of this work is done using analytical and numerical

techniques to develop and manipulate mathematical models.

9
 Operations Research Models

Deterministic Models Stochastic Models

• Linear Programming • Discrete-Time
• Network Optimization • Continuous-Time
• Integer Programming
• Nonlinear Programming • Decision Analysis
Game Theory
Inventory models

10
 Deterministic vs. Stochastic Models
Deterministic models
Assume all data are known with certainty

Stochastic models
Explicitly represent uncertain data via
random variables or stochastic processes.

Stochastic models
characterize / estimate .

11
 Model Development
• Models are representations of real objects or
situations.
• Forms of models mathematical ONE.

– Mathematical models represent real world problems

through a system of mathematical formulas and
expressions based on key assumptions, estimates, or
statistical analyses.

12
 Mathematical Models
• Frequently seek to maximize or minimize some
objective function subject to constraints.

• Are said to be stochastic if any of the

uncontrollable inputs (parameters) is subject to
variation (random), otherwise are said to be
deterministic.
• Generally, stochastic models are more difficult to
analyze.

13
 Advantages of Models
• Generally, experimenting with models
(compared to experimenting with the real
situation):
– requires less time
– is less expensive
– involves less risk

14
 Cont…
• The advantages of mathematical modelling are
many:
 Models exactly represent the real problem
situations.
 Models help managers to take decisions faster
and more accurately.
 Cont…
 Models save valuable resources like money
and time.
 Large and complex problems can be solved
with ease.
 Models act as communicators to others by
providing information.
 Cont…
• Mathematical model describes a problem
much more concisely .
• It helps to reveal important cause-and-effect
relationships.
• It facilitate in dealing with the problem with
entirety and considering all its
interrelationships simultaneously .
 Example: Project Scheduling
• Consider a construction company building a 250-unit
apartment complex.
• The project consists of hundreds of activities involving
excavating, framing, wiring, plastering, painting,
landscaping, and more.
• Some of the activities must be done sequentially and
others can be done simultaneously.
• Also, some of the activities can be completed faster
than normal by purchasing additional resources
(workers, equipment, etc.).
• What is the best schedule for the activities and for which
activities should additional resources be purchased?
18
 Example: Project Scheduling
• Question:
– How could OR/management science be used to
solve this problem?
• Answer:
– OR can provide a structured, quantitative
approach for determining the minimum project
completion time based on the activities'
normal times and then based on the activities'
expedited (reduced) times.

19
 Example: Project Scheduling
• Question:
– What would be the decision variables of the
mathematical model?
– The objective function?
– The constraints?
• Answer:
– Decision variables: which activities to expedite and by
how much, and when to start each activity.
– Objective function: minimize project completion time
– Constraints: do not violate any activity precedence
relationships and do not expedite in excess of the funds
available. 20
 Limitation of OR
• Magnitude of computation
• Non quantifiable factors
• Distance between User and Analyst
• Time and Money Costs
Foundation of Operation
Research
CH-II
 Mathematics and Statistics in OR
• OR use different existing classes of methods
• Define three (methodological) categories:

– mathematically based procedures

– statistically based procedures

• A method can be used for multiple objectives

 Mathematically Based
Algorithms
• Mathematical programming
– linear,
– non-linear,
– integer programming

• Network analyses/affinity analysis

– data flow is represented within a network
 Cont…
• Mathematical programming

– defining objective function and constraints

• Network analyses/affinity analysis

– complex relations are difficult to represent
 Statistically Based Algorithms
• Regression
– variables that better explain the relationship

• Time series
– black-box type of approach, no regressors
 Cont…
• Technical Problems
– Regression
– data base, variable selection/reduction methods
 Formulating A Mathematical Model
• First define the problem
• The next phase is to reformulate this problem
in a form that is convenient for analysis.
• The conventional OR approach for doing
this is:
– to construct a mathematical model (Quantitative)
that represents the essence of the problem.
 Cont…
• Quantitative Techniques adopt a scientific
approach to decision-making.

• Past data is used in determining decisions

that would prove most valuable in the future.
Mathematical Model components
• Mathematical models typically contain types
of quantities:
– decision variables ,
– objective functions and
– Constraints
– parameters (constants)
 Cont…
• Linear programming is a widely used mathematical
modeling technique.
• To determine the optimum allocation of scarce
resources among competing demands.
• Resources typically include:
– raw materials,
– manpower,
– machinery,
– time,
– money and
– space
 Cont…
• The technique is very powerful and found useful.
• Its application to many different types of real
business problems in areas like:
– finance,
– production,
– sales and distribution,
– personnel,
– marketing and
– many more areas of management
FORMULATION OF LINEAR PROGRAMMING

• The representation of problem situation in a

mathematical form.
• It involves well defined:-

1. decision variables,
2. objective function and
3. set of constraints
 Objective function
• The objective of the problem is identified and
converted into a suitable objective function.

• The objective function represents the aim or

goal of the system.
 Cont…
• Generally, the objective in most cases will
be either:-

 to maximize resources or profits or,

to minimize the cost or time
 Cont…
• For example, assume that a furniture
manufacturer produces tables and chairs.

• Objective of the Manufacturer wants to

maximize his profits

• It has to determine the optimal quantity(how

much to produce each line of product) of tables
and chairs to be produced.
 Cont…
Let
x1 = Optimal production of tables
p1 = Profit from each table sold
x2 = Optimal production of chairs
p2 = Profit from each chair sold.
 Cont…
Total profit from tables = p1 x1
Total profit from chairs = p2 x2
The objective function is formulated as below,

Maximize Z = p1 x1  p2 x2
 Constraints
• Let w be the amount of wood available to
produce tables and chairs.
– Each unit of table consumes w1 unit of wood.
– Each unit of chair consumes w2 units of wood.
• For the constraint of raw material
availability, the mathematical expression is,

w1 x1  w2 x2  w
 Non-negativity constraint
• Negative values of physical quantities are
impossible:
– like finishing the project with negative time,
producing negative number of chairs, tables, etc.

– so it is necessary to include the element of non-

negativity as a constraint i.e.,
x1 , x2  0
 Linear Programming II

Cont….
Typical Objectives (Goals):
General: Maximize profit
Minimize Project duration
Maximize Worker Productivity
Maximize Economic Efficiency
Minimize Maintenance and Operational Costs
Minimize all Costs incurred over Cash Flow Period

41
Linear Programming

CH-III
 Assumptions in LP

• Linear Programming Assumptions:

– Proportionality
– Additivity
– Divisibility
– Certainty
 Cont…
•Proportionality

This assumption implies that all the x terms of

the linear equations cannot have exponents
greater than 1.

Note: if there is a term that is a product of

different variables, even though the
proportionality assumption is satisfied, the
additivity assumption is violated.
 But there are many situations, in which
proportionality is not a realistic assumption.
 Cont…
• Additivity
– The contribution of all variables to the objective
function and to the left-hand side of the functional
constraints has to be additive,
– i.e., it has to be the sum of the individual
contributions of the respective activitie.s

– Therefore cross products of variables are ruled out.

 Cont…
• Certainty
– The parameters of the model, (coefficients of the
objective function and of the functional
constraints, and the righ-hand sides of the
functional constraints) are assumed to be known
constants.

– Sometimes we use approximations  important

to perform sensitivity analysis to identify
sensitive parameters.
• Divisibility
– Decision variables in an LP model are
allowed to have any values, including
non integer values, that satisfy the
functional and non negativity constraints.

– i.e., activities can be run at fractional

levels.
 Solution to LP Model
• Linear programming problems with two
variables can be represented and solved
graphically with ease.

• Though in real-life, the two variable problems

are practiced very little.

• The interpretation of this method will help to

understand the simplex method.
 Linear Programming II

Some Standard Definitions

• Feasible solution space

– Any combination of decision variables that satisfies
the set of constraint conditions is called feasible
solution space
• Optimum solution
– It is a feasible solution that satisfies the goal of the
objective function as well.
– The optimal solution is a statement of how
resources or inputs should be used to achieve the
goal in the most efficient and effective manner.
49
 Solution Graphical Method
• Linear programming problems with two
variables can be represented and solved
graphically with ease.
• Example 8: A company manufactures two new
types of product. X1 and X2.
• The product undergo two major processes:
cutting and pinning operations.
• The profits per unit are Rs. 6 and Rs. 4
respectively.
 Cont…
The objective function is,
Zmax= 6x1 + 4x2
• Constraints
2x1 + 3x2 <= 120
2x1+ x2 <= 60

where x1, x2 >= 0

 Graphical Solution
• Step 1: Convert the inequality constraint as
equations and find co-ordinates of the line.
• Step2: Plot the lines on the graph. (Note: If
the constraint is >= type, then the solution
zone lies away from the centre. If the
constraint is <= type, then solution zone is
towards the centre.)
 Cont…
• Step 3: Obtain the feasible zone.
• Step 4: Find the co-ordinates of the objectives
function (profit line) and plot it on the graph
representing it with a dotted line.
• Step 5: Locate the solution point.
– (Note: If the given problem is maximization, zmax
then locate the solution point at the far most point of
the feasible zone from the origin and if minimization,
Zmin then locate the solution at the shortest point of
the solution zone from the origin).
 Solution of LP Graphically
• As a first step, the inequality constraints are removed
by replacing ‘equal to’ sign to give the following
equations:
2x1 + 3x2 = 120
2x1 + x2 = 60
• Find the co-ordinates of the lines by substituting x1 = 0
and x2 = 0 in each equation.
• In equation (1), put x1 = 0 to get x2 and vice versa
2x1 + 3x2 = 120
2(0) + 3x2 = 120, x2 = 40
 Cont…
• Similarly, put x2 = 0,
2x1 + 3x2 = 120
2x1 + 3(0) = 120, x1 = 60
• The line 2x1 + 3x2 = 120 passes through co-
ordinates (0, 40) (60, 0).
• The line 2x1 + x2 = 60 passes through co-
ordinates (0,60)(30,0).
Cont…
 Cont…
• Therefore, we conclude that to maximize profit,
15 numbers of corrugated boxes and 30
numbers of carton boxes should be produced to
get a maximum profit.
• Substituting x1 = 15 and x2= 30 in objective
function, we get
Zmax= 6x1 + 4x2
= 6(15) + 4(30)
• Maximum profit : 210.00
 Linear Programming II

Cont…
Note:
• Linear Programming- is only one of many tools
that can be used for solving continuous-variable
resource allocation problems in project
management.
• Other tools include
- Calculus
- Trial and Error
- Simulation (typically with aid of computer)
58
 Exercise-1
• Solve the following LP graphically:

• Maximize Z = 8x1 + 10x2

Subject to constraints,
2x1 + 3x2 <= 20
4x 1 + 2x2 <= 25
where x1 , x2 >= 0
 Simplex Methods
• In practice, most problems contain more than
two variables.
• An algebraic technique is used to
solve large problems using Simplex Method.
• This method is carried out through iterative
process systematically step by step.
• Finally the maximum or minimum values of
the objective function are attained.
 Cont…
• This method is carried out through iterative
process systematically step by step.

• Finally the maximum or minimum values of

the objective function are attained.
 ADDITIONAL VARIABLES USED IN SOLVING
LPP
• Three types of additional variables are used in simplex
method such as,
(a) Slack variables (S1, S2, S3..…Sn): Slack variables refer to
the amount of unused resources like raw materials,
labour and money.
(b) Surplus variables (-S1, -S2, -S3..…-Sn): Surplus variable is
the amount of resources by which the left hand side of
the equation exceeds the minimum limit.
(c) Artificial Variables (a1, a2, a3.. …an): Artificial variables
are temporary slack variables which are used for
purposes of calculation.
 Cont…
• Maximize Z = 6x1 + 4x2

Subject to constraints

2x1+3x2 <= 120 (Cutting machine)

2x1+ x2 <= 60 (Pinning machine)
where x1, x2 >= 0
 Cont…
• Introducing the slack variable, we have the
equation
2x1+ 3x2 + S3 = 120
• Similarly for pinning machine, the equation is
2x1+ x2 + S4 = 60
• The variables S3 and S4 are known as slack
variables corresponding to the three
constraints.
 Cont…
• If variables x1 and x2 are equated to zero,
i.e., x1 = 0 and x2 = 0, then
S3 = 120
S4 = 60
• This is the basic solution of the system, and
variables S3 and S4 are known as Basic Variables,
SB.
• While x1 and x2 known as Non-Basic Variables.
 Cont…
• Rewriting the constraints with slack variables
gives us,
Z max= 6x1 + 4x2 + 0S3 + 0S4
Subject to constraints,
2x1 + 3x2 + S3 = 120
2x1 + x2 + S4 = 60
where x1, x2>= 0
Cont…
 Cont…
• If the objective of the given problem is a maximization
one, enter the co-efficient of the objective function Zj
with opposite sign.
• Take the most negative coefficient of the objective
function and that is the key column Kc. In this case, it is -
6.
• Find the ratio between the solution value and the key
column coefficient and enter it in the minimum ratio
column.
• The intersecting coefficients of the key column and key
row are called the pivotal element i.e. 2.
 Cont…
• Using these equations, enter the values of
basic variables SB and objective function Z.
• If all the values in the objective function are
non-negative, the solution is optimal.
• Repeat the steps to find the key row and
pivotal equation values for the iteration 2 and
check for optimality.
 Crashing the Activities
• Projects have three dimensions: time,
resources, and scope.
• How long will the project take to complete if
everything goes according to schedule.
• Which tasks have some slack, in the sense that
they can be delayed to some extent without
delaying the project?
 Cont….
• The tasks in a project almost always compete
for resources.

• Whether dollars or nonfinancial resources,

and no real project management application
can afford to ignore these resources.
 Cont…
• Project crashing is ultimately a trade-off
between time and cost.

• We have minimized the crashing cost for a

given time deadline.
 Cont…
• For example, one version of the problem we
analyze is the crashing problem.
• In this problem, we decide how to spend money
optimally to speed up (crash) the completion of the
project.
• For example, if we find that the project will not be
completed until 16 weeks from now, but we have a
deadline of 14 weeks from now, we need to find a
way to crash critical tasks to save a couple of
weeks.
 Cont…

• The most popular optimization models for

project scheduling, called crashing.
• In the crashing model, we are allowed to shorten
the activity durations by spending extra money
on them—that is, we can crash the activities.
• The problem is to spend as little extra money as
possible to complete the project within a given
deadline.
 Cont…
• The objective in many project-scheduling
analyses is to find a minimum-cost method of
reducing activity times to meet a deadline.

• The term crashing the activities is often used

to mean reducing the activity times.
 Cont…
• Of course, it typically costs money to crash
activities hiring extra workers, using extra
equipment, using overtime, and so on.

• So the problem becomes one of crashing just

the right activities in just the right amounts to
meet a deadline at minimum cost.
 Example

Activities Recedence Normal Time days Cost Crashing Time Crashing cost

A 8 10000 7 11500

B A 4 22000 2 24000

C B 6 5000 3 10000

D C 3 14000 2 18000
 Cont…
• Question:
-Find the maximum optimal crashing project
completion time?
What is the crashing cost the project incur?
- If the project deadline is 15 days which
activity you crash as the project manager?
- What is the minimum cost that the project
incur to crash these activities?
 Solution
• Find the slop of each activities

Crash Cos t  Normal Cos t

Slope 
Normaltime  Crashtime
• Find also crash limit by the difference between
Normal Time and Crash Time.
•
 Crashing Example
Crashing Activity Example Column1 Column2 Column3 Column4 Column5 Column6 Column7 Column8 Column9 Column10

Normal Crashing Crashing

Activities Recedence Time Cost Time cost time d cost d cost/time
A 8 10000 7 11500 1 1500 1500
B A 4 22000 2 24000 2 2000 1000
C B 6 5000 3 10000 3 5000 1666.66667
D C 3 14000 2 18000 1 4000 4000
To make comparative analysis
51000 14

How many days we can shoerten this project?

we can go for the cheepest activity per day.

1 day B 1000
1* B 1000
1* A 1500
1* C 1667
1* C 1667
1* C 1667
1* D 4000

Maximum
days of
Total Crashing 7 daya 12501
 Exercise
Exercise of Crashing the Project

Activities Normal time Normal cost crash time crash cost Crash limit cost diff slope

A 2 22000 1 23000 1 1000 1000

B 3 30000 1 34000 2 4000 2000

C 2 26000 1 27000 1 1000 1000

D 4 48000 3 49000 1 1000 1000

E 4 56000 2 58000 2 2000 1000

F 3 30000 2 30500 1 500 500

G 5 80000 2 86000 3 6000 2000

H 2 16000 1 19000 1 3000 3000

 Cont…
• The deadline for the project is only 20 weeks.

• Question
– Which project activities the project manager has
to crash?
– Determine the minimum cost to incur to crash
these activities?
 Example for Formulation LP Portfolio
Selection
Somtel has $5 million to invest among 6 Projects.

Decision: How much to invest in each of 6

projects options?

Objective: Maximize benefit earned

 Data
Benefits
Investment Rate Risk Score
Trade credits 7% 1.7
Corp. bonds 10% 1.2
Gold stocks 19% 3.7
Platinum stocks 12% 2.4
Mortgage securities 8% 2.0
Construction loans 14% 2.9
 Constraints
• Invest up to $ 5 million
• No more than 25% into any one investment
• At least 30% into precious metals
• At least 45% into trade credits and corporate
bonds
 Decision Variables
T = $ invested in trade credit
B = $ invested in corporate bonds
G = $ invested gold stocks
P = $ invested in platinum stocks
M = $ invested in mortgage securities
C = $ invested in construction loans
Objective Function (in $ of benefit earned)
Max 0.07T + 0.10B + 0.19G + 0.12P
+ 0.08M + 0.14C
Subject to the constraints:

Invest Up To $5 Million
T + B + G + P + M + C < 5,000,000
No More Than 25% Into Any One Investment
T < 0.25 (T + B + G + P + M + C)
B < 0.25 (T + B + G + P + M + C)
G < 0.25 (T + B + G + P + M + C)
P < 0.25 (T + B + G + P + M + C)
M < 0.25 (T + B + G + P + M + C)
C < 0.25 (T + B + G + P + M + C)
 At Least 30% Into Precious Metals
G + P > 0.30 (T + B + G + P + M + C)

At Least 45% Into

Trade Credits And Corporate Bonds
T + B > 0.45 (T + B + G + P + M + C)
Finally non negativity: T, B, G, P, M, C > 0

Formulate the programming?

Network Modeling

Ch-VI
 Introduction
• Any project involves planning, scheduling and
controlling a number of interrelated activities.

• It use limited resources, namely:-

– men, machines, materials, money and time

 Cont…
• It is required that managers must have a
dynamic planning and scheduling system to
produce the best possible results.
• A convenient analytical and visual technique
of PERT and CPM used.

• These are extremely valuable in assisting the

managers in managing the projects.
 Cont…
• The techniques use similar terminology and
have the same purpose.
• PERT stands for Project Evaluation and
Review Technique developed during 1950’s.
• CPM stands for Critical Path Method which
was developed by DuPont Company.
– Applied first to the construction projects in the
chemical industry.
 Cont…
• The cost and time for all the activities are
estimated.
• A network diagram is developed showing
sequential interrelationships.

• Predecessor and successor between various

activities during the planning stage.
 NETWORK COMPONENTS
• Networks contain two major components
1. Activities
2. Events
• Activity:
– An activity represents an action
– Consumption of resources (time, money, energy)
– Activity is represented by an arrow
 Cont…
Event:
– An event (or node) will always occur at the
beginning and end of an activity.
– The event has no resources
– Represented by a circle
• The ith event and jth event are the tail event
and head event, respectively .
Cont…
Cont….
 Dummy Activity
• An imaginary activity which does not
consume any resource and time.

• Dummy activities are simply used to represent

a connection between events in order to
maintain a logic in the network.

• It is represented by a dotted line in a network.

Cont…
Cont…
 PROCEDURE FOR NUMBERING THE EVENTS
USING FULKERSON'S RULE

• Step1: Number the start or initial event as 1.

Step2: From event 1, strike off all outgoing
activities.
• This would have made one or more events as
initial events (event which do not have
incoming activities). Number that event as 2.
• Step3: Repeat step 2 for event 2, event 3 and
till the end event. The end event must have
the highest number.
 Cont…
Example 1:
• Draw a network for a house construction
project.
• The sequence of activities with their
predecessors.
Cont…
Cont…
 Network Models
• A project is viewed as a sequence of activities that form
a “network” representing a project
• The path taking longest time through this network of
activities is called the “critical path”
• The critical path provides a wide range of scheduling
information useful in managing a project
• Critical Path Method (CPM) helps to identify the critical
path(s) in the project network
 Prerequisites for Critical Path Method

A project must have:

• well-defined jobs or tasks whose completion
marks the end of the project;
• independent jobs or tasks;
• and tasks that follow a given sequence.
 Project Time Management Processes
• Project time management involves the
processes required to ensure timely completion
of a project.
• Processes include:
– Activity definition
– Activity sequencing
– Activity duration estimating
– Schedule development
– Schedule control
109
 Activity Definition
• Project schedules grow out of the basic
document that initiate a project.

– Scope statement and WBS help define what will be

done.

• Activity definition involves developing a more

detailed WBS.

110
 Activity Sequencing
• Involves reviewing activities and determining
dependencies
– Mandatory dependencies: inherent in the nature
of the work; logic
– Discretionary dependencies: defined by the
project team; soft logic
• You must determine dependencies in order to
use critical path analysis

111
 Project Network Diagrams
• Project network diagrams are the preferred
technique for showing activity sequencing.

• A project network diagram is a schematic

display of the logical relationships among, or
sequencing of, project activities.

112
 Types of Critical Path Methods
• CPM with a Single Time Estimate
– Used when activity times are known with certainty
– Used to determine timing estimates for the project
• Time-Cost Models
– Used when cost trade-off information is a major
consideration in planning
– Used to determine the least cost in reducing total project
time
• CPM with Three Activity Time Estimates
– Used when activity times are uncertain
– Used to obtain the same information as the Single Time
Estimate model and probability information
 Exercise
Crashing Project Network

Normal time Crashing time crashing

Activities (week) Normal cost (week) cost
1-2 7 700 4 850
1-3 5 500 3 700
1-4 8 600 5 1200
2-5 9 800 7 1250
3-5 5 700 3 1000
3-6 6 1100 5 1300
4-6 7 1200 5 1450
5-7 2 400 1 500
6-7 3 500 2 850
 Cont…
• If indirect cost per week is 200$ find the
optimal crashing project completion time?
Steeps
• First find the slope of the activities.
• Draw the network diagram.
• Find critical Path.
Non Linear programming

V
Types Of Nonlinear Programming Problems

• Nonlinear programming problems come in

many different shapes and forms.

• Unlike the simplex method for linear

programming, no single algorithm can solve
all these different types of problems.
 Unconstrained Optimization
• Unconstrained optimization problems have no
constraints, so the objective is simply to

– Maximize f(x)
• The necessary condition that a particular
solution x = x* be optimal when f(x) is a
differentiable function.
 Cont…
• A key assumption of linear programming is
that all its functions (objective function and
constraint functions) are linear.

• Although this assumption essentially holds for

numerous practical problems, it frequently
does not hold.
 Cont…
• Nonlinear optimization or nonlinear
programming.

– Either objective or a constraint (or both) are

nonlinear functions of the decision variables.
Overview

■ Problems that fit the general linear programming format but

contain nonlinear functions are termed nonlinear programming
(NLP) problems.
■ Solution methods are more complex than linear programming
methods.
■ Determining an optimal solution is often difficult, if not
impossible.
■ Solution techniques generally involve searching a solution surface
for high or low points requiring the use of advanced mathematics.

Copyright © 2010 Pearson Education, Inc.

Publishing as Prentice Hall
Optimal Value of a Single Nonlinear Function

Profit, Z = 1,696.8p - 24.6p2 - 22,000

Figure
Copyright ©10.2 The Education,
2010 Pearson NonlinearInc. Profit Function
Publishing as Prentice Hall
Optimal Value of a Single Nonlinear Function
Maximum Point on a Curve
■ The slope of a curve at any point is equal to the derivative of the
curve’s function.
■ The slope of a curve at its highest point equals zero.

Figure 10.3 Maximum

Copyright © 2010profit for theInc.profit function
Pearson Education,
Publishing as Prentice Hall
Optimal Value of a Single Nonlinear Function
Solution Using Calculus
Z = 1,696.8p - 24.6p2 - 2,000
dZ/dp = 1,696.8 - 49.2p
=0
p = 1696.8/49.2
= $34.49
v = 1,500 - 24.6p
v = 651.6 pairs of jeans
Z = $7,259.45

Copyright © 2010 Pearson Education, Inc.

Publishing as Prentice Hall
Figure 10.4
Constrained Optimization in Nonlinear Problems
Definition

■ A nonlinear problem containing one or more constraints becomes

a constrained optimization model or a nonlinear programming
(NLP) model.
■ A nonlinear programming model has the same general form as the
linear programming model except that the objective function
and/or the constraint(s) are nonlinear.
■ Solution procedures are much more complex and no guaranteed
procedure exists for all NLP models.

Copyright © 2010 Pearson Education, Inc.

Publishing as Prentice Hall
Constrained Optimization in Nonlinear Problems
Graphical Interpretation (1 of 3)
Effect of adding constraints to nonlinear problem:

Figure 10.5 Nonlinear Profit Curve for the Profit Analysis Model
Copyright © 2010 Pearson Education, Inc.
Publishing as Prentice Hall
Constrained Optimization in Nonlinear Problems
Graphical Interpretation (2 of 3)

Figure 10.6 A Constrained Optimization Model

Copyright © 2010 Pearson Education, Inc.
Publishing as Prentice Hall
Constrained Optimization in Nonlinear Problems
Graphical Interpretation (3 of 3)

Copyright © 2010 Pearson Education, Inc.

Figure 10.7
Publishing as Prentice Hall
Constrained Optimization in Nonlinear Problems
Characteristics

■ Unlike linear programming, solution is often not on the

boundary of the feasible solution space.
■ Cannot simply look at points on the solution space boundary but
must consider other points on the surface of the objective
function.
■ This greatly complicates solution approaches.
■ Solution techniques can be very complex.

Copyright © 2010 Pearson Education, Inc.

Publishing as Prentice Hall
 Example
• The firm’s profit from producing and selling x
units of the product then would be the sales
revenue, xp(x), minus the production and
distribution costs .

• Therefore, if the unit cost for producing and

distributing the product is fixed at c.
 Cont….
P(x) = xp(x) - cx

• If each of the firm’s n products has a similar

profit function, say, Pj(xj) for producing and
selling xj units of product j ( j = 1, 2, . . . , n),
then the overall objective function is
Cont…
 Cont…
• Nonlinearities also may arise in the gi(x)
constraint functions in a similar fashion.

• For example, if there is a budget constraint on

total production cost, the cost function will be
nonlinear if the marginal cost of production
varies as just described.
Statistical Methods
Regression
CH-VII
 Cont…
What is regression analysis?
• The statistical method that helps to formulate an
algebraic relationship or equation between two
or more variables.

• To estimate the values of associated parameters

with independent variables, given the value of
another variable (dependent variable) is known
as regression analysis.
1 апреля, 2024 Prepared by Kedir Jemal 135
 What are the differences between correlation
and regression analysis?

• Correlation analysis shows the existence of a

relationship between two variables.
– But not they have cause-and-effect relationship

• In correlation analysis both y and x are

independent variables.

1 апреля, 2024 Prepared by Kedir Jemal 136

 Cont…
• In regression analysis y is dependent variable
and x is dependent variable.
• Regression analysis implies causality between
dependent variable y and independent variable
x.
• Correlation analysis implies no causality or
dependency but refers simply to the type and
degree of association between two variables.

1 апреля, 2024 Prepared by Kedir Jemal 137

 Cont…
• Generally, the estimated equation can be used
to predict the value of dependent variable.

• In correlation analysis the correlation

coefficient show the:
– existence, direction and magnitude of a
relationship between the variables; cannot be used
to predict one for the other

1 апреля, 2024 Prepared by Kedir Jemal 138

 Cont…
• When do we apply regression analysis?
• To characterize the relationship between
dependent and independent variables by
determining the:
– extent,
– direction and
– strength of the relationship

1 апреля, 2024 Prepared by Kedir Jemal 139

 Cont…
• To obtain a quantitative formula or equation to
describe the dependent variable as a function
of the independent variables.

• To determine which of several independent

variables are important.

1 апреля, 2024 Prepared by Kedir Jemal 140

 Cont…

• To describe quantitatively or qualitatively the

relationship between different observed
variables.

1 апреля, 2024 Prepared by Kedir Jemal 141

 TYPES OF REGRESSION MODELS

• List types of regression models

Most frequently used regression models

includes:
Simple regression model
Multiple regression model
Multivariate regression model
Logit, Probit, Multinomial regressions etc.

1 апреля, 2024 Prepared by Kedir Jemal 142

 Cont…
• Econometric research or inquiry generally
proceeds along the following lines/stages.
 1. Specification the model
 2. Estimation of the model
 3. Evaluation of the estimates
 4. Evaluation of he forecasting power of the
estimated model

1 апреля, 2024 Prepared by Kedir Jemal 143

 Specification the model
Starting with the postulated theoretical relationships
among economic variables:

Let Y = α + α X + u
i 0 1 i i

Where Y dependent variable

X independent variable
U disturbance term
1 апреля, 2024 Prepared by Kedir Jemal 144
Simple Linear Regression Model
Random
Error term

Slope Independent
Coefficient Variable
intercept

Yi  β 0  β1X i  ε i
Linear component Random Error
component
Dependent
Variable
 Simple Linear Regression Model
Introduction to
Regression Analysis

For each observed value

Xi, an observed value of
Yi is generated by the
population model.
 Cont…
• Explain the variables involved in a regression model.
These variables are observable, unobservable and unknown
parameters.
1. Observable Variables –
These are the variables in which their values are collected from
the field through questionnaires, interviews and other means of
data collection mechanisms.

 yi = the ith value of the dependent variable.

 x = the ith value of the independent variable.

1 апреля, 2024 Prepared by Kedir Jemal 147

 Cont…
Unobservable variables –
• These are the values that will be determined
from the observations and estimated values of
the data set.
• The ε is the random error term for the ith
member of the population is also called –
– The disturbance term
– The stochastic term

1 апреля, 2024 Prepared by Kedir Jemal 148

 Cont…
• The stochastic error term measures the residual
variance in Y not covered by X.
• This is akin to saying there is measurement
error and our predictions/models will not be
perfect.

• The more X variables we add to a model, the

lower the error of estimation.
 Cont….
• Unknown Parameters (or regression
coefficients)

• The regression coefficients are the values that

will be estimated from the sample data of
dependent and independent variables.

1 апреля, 2024 Prepared by Kedir Jemal 150

Regression Application
in Project
Cost Estimation
 Cont…
• Parametric cost estimates:
 A cost estimating methodology using
statistical relationships between historical
costs and other project variables.

• Generally, an estimator selects parametric cost

estimating when only a few key pieces of data
are known.
 Fundamental Assumption

 The implicit assumption in parametric cost

estimating is that the same forces that affected
cost in the past will affect cost in the future.
 Advantages of Parametric Estimating
• Benefits are:
 Better estimates are provided, often in a matter
of minutes;
 There exists a high-quality, statistically
significant and verifiable link between the
technical and cost proposals;
 Early costing cannot be done effectively any
other way.
 Cont…
• Parametric estimating is a process:

– Collecting data from historical projects.

– Identifying cost drivers (determinants)
– Developing Cost Estimating Relationships (CERs)
 Cont…
• The basic idea in CER development is to
– Identify one or more parameters of a project that
best explain its cost,
– Find some historical data that are representative of
the desired cost, and appropriately normalize it
Identify one or more mathematical functions that
“fit” the data
Use to estimate future costs based on similar past
data
 Cost estimation using regression

• Y = The dependent variable (total cost)

• X = The explanatory variables

Y = *X +

where X = machine hours and  = random error.

TC = FC + VC*X + .
 Simple linear regression
• One explanatory variable
• Cost estimation equation
• Coefficient of determination (R2)
– Goodness of fit
– Measure of importance
• F-statistic (hypothesis testing)
• P-value
 Cont…
• The dependent Y variable typically represents
cost, while the independent X factor that affect
cost.
• The independent variables are known as cost
drivers, and typically may be:-
– physical,
– operational characteristics associated with the
project to be estimated
 Cont…
• Generally proceeds along the following
lines/stages.
 1. Specification the model
 2. Estimation of the model
 3. Evaluation of the estimates
 4. Evaluation of he forecasting power of the
estimated model

1 апреля, 2024 Prepared by Kedir Jemal 160

 Specification the model
Starting with the postulated theoretical relationships
among economic variables:

Let Cost = α + β X + u
i 0 1 i i

Where Cost dependent variable

X independent variable like hourly labor, machine
etc
U disturbance term
1 апреля, 2024 Prepared by Kedir Jemal 161
 Cont…
• Derivation of the normal equation

Let Cost = α + α X + u
i 0 1 i i

• Estimation done wit statistical package

1 апреля, 2024 Prepared by Kedir Jemal 162

 Historic data from ten projects
Project Cost Material labor hr
1 6 1 2
2 11 1.5 2.5
3 14 2 3
4 16 1.8 3.2
5 20 1.9 4
6 18 2.5 3
7 22 3 4.2
8 30 3.4 6
9 26 3.7 5
10 31 3.9 6.3
. reg cost material labourhr

Source SS df MS Number of obs = 10

F( 2, 7) = 123.78
Model 574.165341 2 287.082671 Prob > F = 0.0000
Residual 16.2346589 7 2.31923698 R-squared = 0.9725
Adj R-squared = 0.9646
Total 590.4 9 65.6 Root MSE = 1.5229

cost Coef. Std. Err. t P>|t| [95% Conf. Interval]

material 2.666936 1.259027 2.12 0.072 -.310189 5.644061

labourhr 3.774316 .853392 4.42 0.003 1.756365 5.792268
_cons -1.982652 1.447548 -1.37 0.213 -5.40556 1.440256
 Estimated Equation

 Total Project Cost = 2 + 2.7 Material +

3. 8 labor
 Cont…
• The model is useful in estimating project, cost
at early stages of the project.

• Since the information needed is only the labor

and material amount are needed.
• So the estimation could be achieved within
minutes.
Historic data from five Similar Projects

Project Cost ClassRoom labour hr

1 20000 4 10
2 29000 6 12
3 42000 12 20
4 100000 25 50
5 70000 15 35
. reg cost classroom labourhr Cont…
Source SS df MS Number of obs = 5
F( 2, 2) = 364.15
Model 4.2691e+09 2 2.1345e+09 Prob > F = 0.0027
Residual 11723480 2 5861739.99 R-squared = 0.9973
Adj R-squared = 0.9945
Total 4.2808e+09 4 1.0702e+09 Root MSE = 2421.1

cost Coef. Std. Err. t P>|t| [95% Conf. Interval]

classroom 456.4836 748.2103 0.61 0.604 -2762.805 3675.773

labourhr 1711.45 368.4336 4.65 0.043 126.2084 3296.692
_cons 3068.771 2118.964 1.45 0.285 -6048.395 12185.94

.
 Cont…

• Total Project Cost = 3068+ 456 Classroom +

1711labor
 Cont…
 The model is useful in estimating project, cost
at early stages of the project.

• The information needed is only the Quantity

of Class room for the new proposed project.
 Cont…
. reg cost classroom

Source SS df MS Number of obs = 5

F( 1, 3) = 89.92
Model 4.1426e+09 1 4.1426e+09 Prob > F = 0.0025
Residual 138207792 3 46069264.1 R-squared = 0.9677
Adj R-squared = 0.9570
Total 4.2808e+09 4 1.0702e+09 Root MSE = 6787.4

cost Coef. Std. Err. t P>|t| [95% Conf. Interval]

classroom 3865.801 407.6703 9.48 0.002 2568.412 5163.19

_cons 4264.069 5896.44 0.72 0.522 -14501.03 23029.17
 Cont…

 Total Project Cost = 4664 + 3865 Class Room

 Interpreted the result of the regression

 Application of regression formulas yields the

forecasting model.
 Cont…
 It may be used as a primary estimating
method to forecast costs.

 To cross check an estimate developed using

another estimating technique.
 Construction Sector Problems
• Construction projects usually faced with
inaccurate early cost estimates.

– This are burdened with excessive financial and

contractual risks.
• Construction investors and professionals
demand accurate early cost estimates.
 Cont…
• The overall success of construction projects
depends heavily on accurate early cost
estimates.

• Because these estimates provide the basis for

project funding and are essential for feasibility
studies.
 Cont…
• Generally, estimating inaccuracies increase the
potential for:
– Cost overruns,
– Losses,
– Less than expected profits,
– Delays failures,
– Litigation,
– Client dissatisfaction and
– Other undesirable results
 Cont…
• In order to minimize the inaccuracies of early
cost estimates practitioners have developed
several early cost estimating models and
methodologies.
– square foot costs,
– cost indices,
– cost modeling,
– expert analysis
– statistical analysis (Regressions)
 Cont…

• Regression techniques:

– It helps managers determine how well the

estimated regression equation describes the
relations between costs and activities.
Forecasting Models

CH-VII
 Introduction
• Forecast:
– A prediction,
– projection, or
– estimate of some future activity, event, or
occurrence
Alternative Forecasting Techniques

Source: ICAO Manual on Air Traffic Forecasting

 Selecting a Forecasting Technique
• What is the purpose of the forecast?
• How is it to be used?
• How important is the past in estimating the
forecast?
 Qualitative Methods
• These types of forecasting methods are based on:
– Judgments, opinions, intuition, emotions, or personal
experiences and are subjective in nature.
– They do not rely on any rigorous mathematical
computations.
• Delphi Method
– Formation of a group judgment

– The objective of most Delphi applications is the

reliable and creative exploration of ideas.
 Cont…
• Group of experts used to make long-range
projections.
• Issues like
– future direction of business conditions,
– business activities,
– technology,
– new product development, and
– market conditions
 Cont…
• These experts are kept apart from each other so that
their opinions are established independently.

• The Delphi Method is a group decision process about

the likelihood that certain events will occur.
 Forecasting General Economic Conditions

• It is measuring the GDP, which is the value of goods and

services produced within a country during a given year.
• Some of the others factors are;
– stock market fluctuation,
– personal income,
– level of employment,
etc
• These data are usually available from government or trade
associations.

186
 Cont…
• The three principles of all forecasting techniques:

– Forecasting is always wrong

– Every forecast should include an estimate of error
– The longer the forecast horizon the worst is the
forecast
– what can you say?
THANK YOU !!!