14 Apr 2012

School

Course

Professor

Decision-making and Quantitative Modeling

Chapter1:

What do all of these have in common?

They are models. They can be physical models such as a miniature model of an airplane, or of a

building, a globe, etc. Physical models are built to look like the real thing and to resemble the

larger object. Two dimensional models such as maps also show the shape of the object.

The second type of models are Analog Models. Analog Models try to represent something but

they do not look like the object they are modeling. For example, a thermometer is a model that

measures the heat in the room but it does not look like the heat. Other examples are

hourglasses, speedometers etc.

Finally, there are Mathematical Models. These are models that represent a certain object using

a mathematical relationship or equation. For example, Profit = Revenues – Expenses.

Mathematical

Relationships

Independent or

decision variables

(controllable

factors)

Parameters

(uncontrollable

factors)

Dependent

Variables

Area

Dependent Variables

Inde

Para

Finance

Profit

Rate of Return

Earnings per Share

Investment amount

Period of investment

Timing of investment

Inflation rate

Interest rate

Marketing

Market share

Customer satisfaction

Advertising budget

Zonal sales reps

price

Disposable income

Competitors actions

Manufacturing

Total cost

Quality measures

spoilage

Production amounts

Inventory levels

Machine capacity

Technology

Materials prices

Services

Customer satisfaction

Number of servers

Demand for service

What is a model?

A simplified representation or an abstraction of reality.

Physical models are simplified because they are smaller and have less detail.

Mathematical models look at the more detailed aspects and have an abstraction

Important qualities:

o Validity (it should look like what it tries to represent)

o Usability (should be usable)

o Value (there is a cost associated in building it that we should get back)

Benefits of Modeling

Models enable the compression of time

Allow for easier manipulation than the real system

Lower cost that experimenting with the real system

Allows us to consider risk

Cost of mistakes is smaller

Models help us learn and understand the real system

The modeling process forces the use of rigorous thinking

Mathematical models let us analyze a large number of possible solutions

Drawbacks of Modeling

Can be time-consuming and expensive

Data collection can be expensive or impossible

It may be difficult to assess uncertainties

Oversimplification may produce poor results

Managers must agree to accept results

Common perception that if it is done on a computer, it must be correct

Uses of Models

Prescriptive

o Used to find the best or optimal solution to a problem by

Enumeration or

Algorithm

Descriptive

o Characterize things as they are

o Investigate outcomes and consequences

o Predict the behaviour of systems in certain situations

o Solutions are not necessarily optimal

Effective Modelers

Internal skills

o Creativity

o Sensitivity to the client

o Persistence

Interpersonal skills

o Communication

o Teamwork

Expertise in quantitative business modeling

The Modeling Process

1. Opportunity / problem recognition

2. Model formulation

3. Data collection

4. Analysis of