1

Introduction to Operations Research

N Onur Bakır
1 Operations Research
Operations research is simply a scientific approach to decision making that seeks
to best design and operate a system, usually under conditions requiring the
allocation of scarce resources.
By a system, we mean an organization of interdependent components that work
together towards a goal.
The term operations research was coined during World War II when British
military leaders asked scientists and engineers to analyze several military
problems such as deployment of radar and the management of convoy, bombing,
antisubmarine, and mining operations.
The scientific approach to decision making usually involves the use of one or
more mathematical models. A mathematical model is a mathematical
representation of an actual situation that may be used to make better decisions
or simply to understand the actual situation better.
1 Prescriptive or Optimization
Models
Most of the models discussed in this course will be prescriptive or optimization
models. A prescriptive model “prescribes” behavior for an organization that will
enable it to best meet its goals. The components of a prescriptive model include

objective function(s)
decision variables
contraints

In short, an optimization model seeks to find values of the decision variables that
optimize an objective function among the set of all values for the decision
variables that satisfy the given constraints.
1 Prescriptive or Optimization
Models
Optimization methods are used to solve problems such as:
1) What blend of raw materials are needed to meet production requirements?
2) How should a process be designed and operated to maximize expected
profit?
3) How should I schedule the production of goods from several plants to meet
contractual requirements while minimizing cost?
4) From a given level of business risk, what mix of investments will maximize
expected return?
5) How do I plan and operate a process with uncertainties in process behavior,
prices, and product demand?
1 Static and Dynamic Models
A static model is one in which the decision variables do not involve sequences of
decisions over multiple periods. A dynamic model is a model in which the
decision variables do involve sequences of decisions over multiple periods. Hence,
the static model solves a one-shot problem in which optimal values of all decision
variables in the model are determined at one time. In dynamic problems, on the
other hand, decisions at various points in time are solved in a dynamic fashion.

In this course, we will discuss both types of models. However, the solution
methods that will be taught can be used to solve static problems. We should also
note that the solution methods developed for static models form the basis of
solution methods for dynamic problems.
1 The Seven-Step Model-Building
Process
1) Formulate the problem: Define the organization’s problem.
2) Observe the system: Collect data to estimate the value of parameters of the
problem.
3) Formulate a mathematical model of the problem: Develop a mathematical
model.
4) Verify the model and use the model for prediction: Make sure that the
mathematical model developed in step 3 is an accurate representation of
reality.
5) Select a suitable alternative: Find the alternative that best meets the
objective.
6) Present the results and conclusion of the study to the organization.
7) Implement and evaluate recommendations.
1 The Seven-Step Model-Building
Process
Let’s now see how we may use the seven-step list to solve a queuing problem.
Suppose that a bank manager wants to reduce expenditures on tellers’ salaries
while still maintaining an adequate level of customer service.

Step 1: The OR analyst describes bank’s objectives. The objective might be, e.g.,
• The bank wants to minimize the weekly salary cost needed to ensure that the
average waiting a customer waits in line is at most 3 minutes.
• The bank wants to minimize the weekly salary cost required to ensure that
only 5% of all customers wait in line more than 3 minutes.

The analyst must also identify the aspects of the bank’s operations that affect the
achievement of the bank’s objectives, e.g.,
• On the average, how many customers arrive at the bank each hour?
• On the average, how many customers can a teller serve per hour?
1 The Seven-Step Model-Building
Process
Step 2: The OR analyst observes the bank and estimates, among others, the
following parameters:
• On the average, how many customers arrive each hour? Does the arrival rate
depend on the time of day?
• On the average, how many customers can a teller serve each hour? Does the
service speed depend on the number of customers waiting in line?

Step 3: The OR analyst develops a mathematical model. In this example a

queuing model is appropriate. Let
1 The Seven-Step Model-Building
Process
A certain mathematical queuing model yields a connection between these
parameters:

Step 4: The analyst tries to verify that the model represents reality well enough.
This means that the OR analyst will estimate the parameter Wq , λ, and µ
statistically, and then she will check whether the above equation is valid, or close
enough.

Step 5: The analyst will optimize the model.

Steps 6 and 7? Quite routine.