Operations Research, MTH 1007 Sections 1.1 and 1.2 Prof. F.

Rispoli

Introductory Comments Try to obtain the book, Quantitative Models for Management, Third Edition, by Abramson, Cosares and Rispoli. The book is available in the bookstore and can be ordered online. If you are having any trouble getting the book let know. The class notes are meant to be studied along with the text. Course Materials will be posted on Blackboard. Be sure to email me today to confirm your registration. For the notes in this modules, please read Section 1.1 and Section 1.2. Before we start with material in the book I would like to begin with some useful quantitative models that help when making decisions in the business and financial world.

The Sharpe Ratio (Created by Professor Sharpe who won a Nobel Prize in Economics in 1990) The Sharpe ratio is a measure of the excess return per unit of risk in an investment asset. This can be used to measure the efficiency of an investment. The model is given by S(x) = x is the investment (stock) where rx is the expected rate of return of x

Rf is the risk free rate of return (T-bills) StdDev(x) is the standard deviation of x

Example In February of 2009 the ETF GLD had a average return of 18.82% and a standard deviation of 21.23 (obtained from Yahoo finance). If the risk free rate of return is 3.75% what is the Sharpe Ratio for GLD?

S(x) = (rx Rf)/StdDev(x) = (18.82 3.75)/21.23 = .71

In May 2009 the ETFs GLD and IYR were compared using statistics from Yahoo Finance.

Based on prior 3 years we compared Gold to the Real Estate ETF denoted IYR

Statistic Beta (against S&P 500 index) Return Standard Deviation Sharpe Ratio

GLD 2.76 14% 20.23 0.44

IYR 1.66 18% 36.86 0.45

Asset Allocation Models These models suggest how to divide up a portfolio among asset classes based on the percentages of the portfolio. The allocations must be updated at least every 6 months.

Asset Class Domestic Stock Foreign Stock Bonds Real Estate Cash, CD's

Balanced 40% 20% 20% 15% 5%

Conservative 20% 15% 45% 10% 10%

Very Conservative 10% 5% 75% 0% 10%

How does one find the best asset allocation? It depends on the investors tolerance for short term risk. We will look at problems like this when we discuss simulation.

Control Charts Control charts are used in manufacturing to control product quality. Samples are taken at regular intervals and the sample means are plotted. Control limits are determined using mean 3*standard deviation

For more information on control charts, try performing a Google search

Now some material from the book In Section 1.1 there is a discussion of Process maps which are used to model a process or operation. The symbols used in a process map are given below

A rectangle indicates an activity. Statements within the rectangle usually begin with a verb. A diamond signifies a decision point. Only two paths should emerge from a decision point: No and Yes. A parallelogram shows that there is data.

An ellipse shows the start and end of a process.

An arrow shows the connection and direction of flow.

A circle with a number symbolizes a connection to a flow chart on another page.

The special types of vertices and symbols used in a process map

Yes Start Step A Step B Defects No Finish

A simple process map

A process map of the management viewpoint of making pizza

Pizza Dough

No
Take Order from Cashier Add Ingredients Place in Oven Observe Frequently Check if Done

Yes

Remove from Oven

Start New Pizza

Scrap

No 1
Pizza Correct Place in Box Tape Order on Box Put on Delivery Rack

Yes

A process map of the worker viewpoint of making pizza

Cost-Volume Profit Model, Section 1.2

Atlas Bookbinders It costs Atlas Bookbinders about $80,000 to prepare for production the new edition of their popular Business Math textbook. This includes the payment to the authors and editors, the costs of setting the presses, adding the book to their catalog, and designing the books cover. These are one-time "fixed" costs that are incurred, even if only one copy of the textbook is printed. Materials, marketing, pressing and binding cost a total of about $44.00 for each book. These are called "variable" costs because the total depends on the number of units that will be produced. Any number of books can be produced at that price. Copies of the book are sold to campus bookstores for $60 each. The total demand for this book is not known exactly, but some estimates are available. Would it be a good idea for Atlas to spend the $80,000 to initiate production of the textbook? If so, how many books should Atlas make in its first production run?

We use the letter x for the decision variable representing the number of units of the book that Atlas will make. Then some useful expressions are:

Revenue from selling x units = R(x) = r x Cost of producing x units = C(x) = F + v x

r = selling price F = fixed cost v = variable cost

We can put these together to make an equation Profit from selling x units = P(x) = R(x) - C(x) = rx (F + vx) = (r - v) x - F The break-even point is the number of units for which total costs equals total revenue; i.e., the value of x satisfying P(x) = 0.

Break-even point = xBEP =

Example How many books must be sold for the book company to: a) Breakeven? b) Make a profit of $40,000?

Solution

BEP =

= 5,000 units

F = 80,000, r = 60, v = 44

The amount that must be sold to earn a profit of $40,000 is x = 7,500

What if we want to earn a profit of $P? Number of units to sell to obtain P dollars of profit = xP = (F + P) / (r v)

Revenue = rx
Profit

Cost = F + vx

F
Loss

xBEP

Quantity (x)

Suppose we have a forecast for sales that tells us sales follow a normal distribution.

Prob. of Profit

Prob. Of loss Loss

xBEP

Quantity (x)

To find probabilities we need to calculate z-scores: z-score of a value x =

For example, suppose that the demand for textbook is normally distributed with a mean of 9,000 units and a standard deviation of 3,000 units. The probability of at least breaking even is equal to the probability that the sales will exceed 5,000 units. We calculate the z-score of 5,000, which is z = .

Use the table at the back of the book (page 231). Look up the probability corresponding to z = -1.33. This should be 0.0918. This process should be somewhat familiar from the prerequisite course in Statistics.

The probability that sales will exceed the break-even point is equal to the probability of exceeding the value 1.33 in the Standard Normal distribution, which is 1 (.0918) = .9082, or roughly 91%. Notice that the probability of a loss is .0918 or roughly 9%

What is the probability of earning a profit of $60,000 or more?

Example 1 Suppose that the selling price of a lamp is $7.50, its variable cost is $4.50, and the fixed cost for production is $1,500,000. Suppose it is known that sales are normally distributed with mean 600,000 units and standard deviation of 300,000 units. What is the probability of at least breaking even? What is the probability of a loss? What is the probability of earning a profit of $360,000 or more? Suppose a manager has decided that she will go into production only if the chances of breaking even are at least 90%. Would she decide to produce lamps?

Example 2 Alpha Cosmetics company has developed a new perfume that they want to market. If Alpha goes into small-scale production its annual fixed costs will be $400,000 and it can produce for a variable cost of $5 per unit. If it goes into large-scale production, its annual fixed costs are $1,050,000 and it can produce for a variable cost of $3 per unit. The selling price will be $10 per unit regardless of the scale of production.

a) Determine the break-even points for small-scale and large-scale production. b) Suppose the expected demand is for 150,000 units. Which production method would be more profitable? Compute the profit for both scales of production. c) Suppose the expected demand is for 500,000 units. Which production method would be more profitable? Compute the profit for both scales of production.

d) Find the number of units for which one would be indifferent between smallscale and large-scale production, i.e., the number of units for which the costs are equal.