0% found this document useful (0 votes)

1K views

Introduction To Spreadsheet Modeling - Winston Albright

introduction to spreadsheet modeling

Uploaded by

Jotham Hensen

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

1K views

Introduction To Spreadsheet Modeling - Winston Albright

introduction to spreadsheet modeling

Uploaded by

Jotham Hensen

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 46

',%48)6

Introduction to Spreadsheet Modeling

Lise Gagne/istockphoto

%2%0=7-73*,-:%-(7

ER] SJ QEREKIQIRX WGMIRGIW QSWX WYGGIWWJYP ETTPMGEXMSRW EVI XVEHMXMSREP

JYRGXMSREP EVIEW SJ FYWMRIWW MRGPYHMRK STIVEXMSRW QEREKIQIRX PSKMWXMGW
JMRERGI ERH QEVOIXMRK -RHIIH QER] WYGL ETTPMGEXMSRW EVI EREP]^IH MR XLMW
FSSO ,S[IZIV ERSXLIV EVIE [LIVI QEREKIQIRX WGMIRGI LEW LEH E WXVSRK
MRJPYIRGI SZIV XLI TEWX X[SHIGEHIW LEW FIIR XLI EREP]WMW SJ XLI [SVPH[MHI
,-:%-(7 ITMHIQMG 2SX SRP] LEZI XLISVIXMGEP QSHIPW FIIR HIZIPSTIH FYX
IZIR QSVI MQTSVXERX XLI] LEZI EPWS FIIR ETTPMIH XS LIPT YRHIVWXERH XLI
ITMHIQMG ERH VIHYGI MXW WTVIEH 8S LMKLPMKLX XLI MQTSVXERGI SJ QEREKIQIRX
WGMIRGI QSHIPMRK MR XLMW EVIE ER IRXMVI WTIGMEP MWWYI 1E].YRI SJ
-RXIVJEGIW XLI NSYVREP XLEX VITSVXW WYGGIWWJYP QEREKIQIRX WGMIRGI ETTPMGEXMSRW
[EW HIZSXIH XS ,-:%-(7 QSHIPW 7SQI SJ XLI LMKLPMKLXW EVI HMWGYWWIH LIVI XS
KMZI ]SY ER MHIE SJ [LEX QEREKIQIRX WGMIRGI LEW XS SJJIV MR XLMW MQTSVXERX EVIE
/ELR IX EP TVSZMHIW ER SZIVZMI[ SJ XLI TVSFPIQ 8LI] HMWGYWW LS[
KSZIVRQIRXW TYFPMGLIEPXL EKIRGMIW ERH LIEPXLGEVI TVSZMHIVW QYWXHIXIV
QMRI LS[ FIWX XS EPPSGEXI WGEVGI VIWSYVGIW JSV ,-: XVIEXQIRX ERH TVIZIRXMSR
EQSRK HMJJIVIRX TVSKVEQW ERH TSTYPEXMSRW 8LI] HMWGYWW MR WSQI HITXL LS[
QEREKIQIRX WGMIRGI QSHIPW LEZI MRJPYIRGIH ERH [MPP GSRXMRYI XS MRJPYIRGI
%-(7 TSPMG] HIGMWMSRW 3XLIV EVXMGPIW MR XLI MWWYI HMWGYWW QSVI WTIGMJMG TVSF
PIQW 'EYPOMRW IX EP EREP]^I [LIXLIV XLI HMWXVMFYXMSR SJ HMJJMGYPXXS
VIYWI W]VMRKIW [SYPH VIHYGI XLI WTVIEH SJ ,-: EQSRK MRNIGXMSR HVYK YWIVW
&EWIH SR XLIMV QSHIP XLI] GSRGPYHI XLEX XLI I\XVE I\TIRWI SJ XLIWI X]TIW SJ
W]VMRKIW [SYPH RSX FI [SVXL XLI QEVKMREP FIRIJMX XLI] QMKLX TVSZMHI

Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.

4EPXMIPERH*VIIHFIVK MRZIWXMKEXIXLIGSWXWERHFIRIJMXWSJHIZIPSTMRKERH

EHQMRMWXIVMRKXVIEXQIRXWJSVG]XSQIKEPSZMVYW'1: ERMRJIGXMSRXS[LMGL,-:GEVVMIVW
EVIMRGVIEWMRKP]I\TSWIH 6IXMRMXMW '1:WQSWXGSQQSRQERMJIWXEXMSR MWEWWSGMEXIH
[MXLFPMRHRIWWERHWSQIXMQIWHIEXL 8LIMVQSHIPWYKKIWXWXLEXXLIGSWXWGSQTEVI
YRJEZSVEFP][MXLEPXIVREXMZIYWIWSJWGEVGIVIWSYVGIW 3[IRWIXEP EREP]^IXLI
IJJIGXSJ[SQIRWVIPETWIXSLMKLVMWOWI\YEPERHRIIHPIWLEVMRKFILEZMSVSRXLIGSWXW
ERHFIRIJMXWSJEZSPYRXEV]TVSKVEQXSWGVIIR[SQIRSJGLMPHFIEVMRKEKIJSV,-: 8LI]
JMRH JSVI\EQTPI XLEXXLIIJJIGXSJVIPETWIXSLMKLVMWOFILEZMSVWSRWGVIIRMRKTVSKVEQ
GSWXWERHFIRIJMXWGERFIWYFWXERXMEP WYKKIWXMRKXLEXFILEZMSVEPMRXIVZIRXMSRWXLEXTVS
HYGIWYWXEMRIHVIHYGXMSRWMRVMWOFILEZMSV IZIRMJI\TIRWMZI GSYPHFIGSWXWEZMRK
8LIMQTSVXERXTSMRXMWXLEXXLIWIEVXMGPIWERHSXLIVWRSXQIRXMSRIHLIVI FEWIXLIMV
VIWYPXWSRVMKSVSYWQEREKIQIRXWGMIRGIQSHIPWSJXLI,-:%-(7TLIRSQIRSR -REHHM
XMSR XLI]EVIFEGOIHYT[MXLVIEPHEXE 8LI]EVIRSXWMQTP]STMRMSRWSJXLIEYXLSVW Q

2.1 INTRODUCTION
This book is all about spreadsheet modeling. By the time you are finished, you will have
seen some reasonably complexand realisticmodels. Many of you will also be transformed into Excel power users. However, we dont want to start too quickly or assume too
much background on your part. For practice in getting up to speed with basic Excel features,
we have included an Excel tutorial on this textbooks essential resource Web site. (See the
([FHO7XWRULDO[OV[ file.) You can work through this tutorial at your own speed and cover
the topics you need help with. Even if you have used Excel extensively, give this tutorial a
look. You might be surprised how some of the tips can improve your productivity.
In addition, this chapter provides an introduction to Excel modeling and illustrates
some interesting and relatively simple models. The chapter also covers the modeling
process and includes some of the less well known, but particularly helpful, Excel tools
that are available. These tools include data tables, Goal Seek, lookup tables, and auditing
commands. Keep in mind, however, that our objective is not the same as that of the many
how-to Excel books on the market. We are not teaching Excel just for its many interesting features. Rather, we plan to XVHthese features to provide insights into real business problems. In short, Excel is a problem-solving tool, not an end in itself, in this
book.

2.2 BASIC SPREADSHEET MODELING:

CONCEPTS AND BEST PRACTICES
Most mathematical models, including spreadsheet models, involve LQSXWV, GHFLVLRQ YDUL
DEOHV, and RXWSXWV. The LQSXWV have given fixed values, at least for the purposes of the
model. The GHFLVLRQYDULDEOHV are those a decision maker controls. The RXWSXWV are the
ultimate values of interest; they are determined by the inputs and the decision variables.
For example, suppose a manager must place an order for a certain seasonal product. This
product will go out of date fairly soon, so this is the only order that will be made for the
product. The inputs are the fixed cost of the order; the unit variable cost of each item
ordered; the price charged for each item sold; the salvage value for each item, if any, left in
inventory after the product has gone out of date; and the demand for the product. The decision variable is the number of items to order. Finally, the key output is the profit (or loss)
from the product. This output can also be broken down into the outputs that contribute to

'LETXIV -RXVSHYGXMSRXS7TVIEHWLIIX1SHIPMRK
Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.

7SQIMRTYXW WYGLEW
HIQERHMRXLMW
I\EQTPI GSRXEMRE
GSRWMHIVEFPIHIKVII
SJ YRGIVXEMRX] -R
WSQI GEWIW EWMR
)\EQTPIPEXIVMR
XLMWGLETXIV XLMW
YRGIVXEMRX]MWQSHIPIH
I\TPMGMXP]

it: the total ordering cost, the revenue from sales, and the salvage value from leftover items.
These outputs must be calculated to obtain profit.
Spreadsheet modeling is the process of entering the inputs and decision variables into
a spreadsheet and then relating them appropriately, by means of formulas, to obtain the
outputs. After you have done this, you can then proceed in several directions. You might
want to perform a sensitivity analysis to see how one or more outputs change as selected
inputs or decision variables change. You might want to find the values of the decision variable(s) that minimize or maximize a particular output, possibly subject to certain constraints. You might also want to create charts that show graphically how certain parameters
of the model are related.
These operations are illustrated with several examples in this chapter. Getting all the
spreadsheet logic correct and producing useful results is a big part of the battle; however,
we go farther by stressing good spreadsheet modeling SUDFWLFHV You probaby wont be
developing spreadsheet models for your sole use; instead, you will be sharing them with
colleagues or even a boss (or an instructor). The point is that other people will be reading
and trying to make sense out of your spreadsheet models. Therefore, you should construct
your spreadsheet models with UHDGDELOLW\in mind. Features that can improve readability
include the following:
Q
Q
Q
Q
Q
Q
Q

A clear, logical layout to the overall model

Separation of different parts of a model, possibly across multiple worksheets
Clear headings for different sections of the model and for all inputs, decision variables, and outputs
Use of range names
Use of boldface, italics, larger font size, coloring, indentation, and other formatting
features
Use of cell comments
Use of text boxes for assumptions and explanations

Obviously, the formulas and logic in any spreadsheet model must be correct; however,
correctness will not take you very far if no one can understand what you have done. Much
of the power of spreadsheets derives from their flexibility. A blank spreadsheet is like a big
blank canvas waiting for you to insert useful data and formulas. Almost anything is
allowed. However, you can abuse this power if you dont have an overall plan for what
should go where. Plan ahead before diving in, and if your plan doesnt look good after you
start filling in the spreadsheet, revise your plan.
The following example illustrates the process of building a spreadsheet model according
to these guidelines. We build this model in stages. In the first stage, we build a model that is
correct, but not very readable. At each subsequent stage, we modify the model to make it
more readable. You do not need to go through each of these stages explicitly when you build
your own models. You can often strive for the final stage right away, at least after you get
accustomed to the modeling process. The various stages are shown here simply for contrast.

)<%140)

2.1 O RDERING NCAA T-S HIRTS

t is March, and the annual NCAA Basketball Tournament is down to the final four teams.
Randy Kitchell is a T-shirt vendor who plans to order T-shirts with the names of the final
four teams from a manufacturer and then sell them to the fans. The fixed cost of any order
is $750, the variable cost per T-shirt to Randy is $8, and Randys selling price is $18.
However, this price will be charged only until a week after the tournament. After that time,
Randy figures that interest in the T-shirts will be low, so he plans to sell all remaining
&EWMG7TVIEHWLIIX1SHIPMRK 'SRGITXWERH&IWX4VEGXMGIW

T-shirts, if any, at $6 each. His best guess is that demand for the T-shirts during the fullprice period will be 1500. He is thinking about ordering 1450 T-shirts, but he wants to
build a spreadsheet model that will let him experiment with the uncertain demand and his
order quantity. How should he proceed?
2EMHFWLYH To build a spreadsheet model in a series of stages, all stages being correct but
each stage being more readable and flexible than the previous stages.

7SPYXMSR
The logic behind the model is fairly simple, but the model is built for generality.
Specifically, the formulas used allow for the order quantity to be less than, equal to, or
greater than demand. If demand is greater than the order quantity, Randy will sell all the Tshirts ordered for $18 each. However, if demand is less than the order quantity, Randy will
sell as many T-shirts as are demanded at the $18 price and all leftovers at the $6 price. You
can implement this logic in Excel with an IF function.
A first attempt at a spreadsheet model appears in Figure 2.1. (See the file 76KLUW6DOHV
)LQLVKHG[OV[where each stage appears on a separate worksheet.) You enter a possible
demand in cell B3, a possible order quantity in cell B4, and then calculate the profit in
cell B5 with the formula
%,)%!% % % %%
This formula subtracts the fixed and variable costs and then adds the revenue according to the logic just described.
*MKYVI
Base Model

1
2
3
4
5

A
B
NCAA t-shirt sales
Demand
Order
Prot

1500
1450
13750

([FHO )XQFWLRQ ,)
([FHOV ,)IXQFWLRQLVSU REDEO\DOUHDG\IDPLOLDUWR\RXE XWLWLVWRRLPSRUWDQWQRWWRGLV
FXVV,WKDVWKHV\QWD[ =IFFRQGLWLRQUHVXOW,I 7UXHUHVXOW,I)DOVH 7KH condition LVDQ\
H[SUHVVLRQWKDWLVHLWKHUWUXHRUIDOVH 7KHWZRH [SUHVVLRQVresultIf True DQGresultIf False
FDQEHDQ\H [SUHVVLRQV\RXZRXOGHQWHULQDFHOOQXPEHU VWH[WRURWKHU([FHO IXQFWLRQV
LQFOXGLQJRWKHU,)IXQFWLRQV1RWHWKDWLIHLWKHUH [SUHVVLRQLVWH[WLWPXVWEHHQFORVHGLQ
GRXEOHTXRWHVVXFKDV
,)6FRUH! $%
)LQDOO\condition FDQEHFRPSOH[FRPELQDWLRQVRIFRQGLWLRQVXVLQJWKHNH\ZRUGV$1'RU
257KHQWKHV\QWD[LVIRUH[DPSOH

,)$1'6FRUH6FRUH)DLO3DVV

2IZIVLEVHGSHI
RYQFIVWMRXS)\GIP
JSVQYPEW 9WIGIPP
VIJIVIRGIWMRWXIEH

This model in Figure 2.1 is entirely correct, but it isnt very readable or flexible because
it breaks a rule that you should strive never to break: It KDUGFRGHV input values into the
profit formula. A spreadsheet model should QHYHU include input numbers in formulas.
Instead, the spreadsheet model should store input values in separate cells and then include
FHOOUHIHUHQFHVto these inputs in its formulas. A remedy appears in Figure 2.2. Here, the
inputs have been entered in the range B3:B6, and the profit formula in cell B10 has been
changed to
%% %,)%!%% % %% %%

*MKYVI
Model with Input
Cells

1
2
3
4
5
6
7
8
9
10

A
NCAA t-shirt sales
Fixed order cost
Variable cost
Selling price
Discount price
Demand
Order
Prot

$750
$8
$18
$6
1500
1450
$13,750

This is exactly the same formula as before, but it is now more flexible. If an input changes,
the profit recalculates automatically. Most important, the inputs are no longer buried in the
formula.1
Still, the profit formula is not very readable as it stands. You can make it more readable by using range names. The mechanics of range names are covered in detail later in this
chapter. For now, the results of using range names for cells B3 through B6, B8, and B9 are
shown in Figure 2.3. This model looks exactly like the previous model, but the formula in
cell B10 is now
)L[HGBRUGHUBFRVW9DULDEOHBFRVW 2UGHU,)'HPDQG!2UGHU
6HOOLQJBSULFH 2UGHU6HOOLQJBSULFH 'HPDQG'LVFRXQWB3ULFH 2UGHU'HPDQG
This formula is admittedly more long-winded, but it is certainly easier to read.
*MKYVI
Model with Range
Names in Profit
Formula

1
2
3
4
5
6
7
8
9
10

A
NCAA t-shirt sales
Fixed order cost
Variable cost
Selling price
Discount price
Demand
Order
Prot

$750
$8
$18
$6

Range names used

Demand
Discount_price
Fixed_order_cost
Order
Selling_price
Variable_cost

1500
1450
$13,750

='Model 3'!$B$8
='Model 3'!$B$6
='Model 3'!$B$3
='Model 3'!$B$9
='Model 3'!$B$5
='Model 3'!$B$4

Randy might like to have profit broken down into various costs and revenues
(Figure 2.4), rather than one single profit cell. The formulas in cells B12, B13, B15, and
B16 are straightforward, so they are not repeated here. You can then accumulate these to
get profit in cell B17 with the formula
%%%%
*MKYVI
Model with
Intermediate
Outputs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

A
NCAA t-shirt sales

Fixed order cost

Variable cost
Selling price
Discount price

$750
$8
$18
$6

Demand
Order

1500
1450

Costs
Fixed cost
Variable costs
Revenues
Full-price shirts
Discount-price shirts
Profit

Range names used

Demand
Discount_price
Fixed_order_cost
Order
Selling_price
Variable_cost

='Model 4'!$B$8
='Model 4'!$B$6
='Model 4'!$B$3
='Model 4'!$B$9
='Model 4'!$B$5
='Model 4'!$B$4

$750
$11,600
$26,100
$0
$13,750

1Some people refer to such numbers buried in formulas as PDJLFQXPEHUV because they just seem to appear out of
nowhere. Avoid magic numbers!

&EWMG7TVIEHWLIIX1SHIPMRK 'SRGITXWERH&IWX4VEGXMGIW

*MKYVI
Model with
Category Labels and
Color Coding

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

A
NCAA t-shirt sales

Input variables
Fixed order cost
Variable cost
Selling price
Discount price

$750
$8
$18
$6

Uncertain variable
Demand

1500

Decision variable
Order

1450

Output variables
Costs
Fixed cost
Variable costs
Revenues
Full-price shirts
Discount-price shirts
Profit

Range names used

Demand
Discount_price
Fixed_order_cost
Order
Selling_price
Variable_cost

='Model 5'!$B$10
='Model 5'!$B$7
='Model 5'!$B$4
='Model 5'!$B$13
='Model 5'!$B$6
='Model 5'!$B$5

$750
$11,600
$26,100
$0
$13,750

Of course, range names could be used for these intermediate output cells, but this is probably more work than its worth. You should always use some judgment when deciding how
many range names to use.
If Randys assistant is presented with this model, how does she know at a glance which
cells contain inputs or decision variables or outputs? Labels and/or color coding can help to
distinguish these types. A blue/red/gray color-coding style has been applied in Figure 2.5,
along with descriptive labels in boldface. The blue cells at the top are input cells, the red cell
in the middle is a decision variable, and the gray cell at the bottom is the key output.2 There
is nothing sacred about this particular convention.
Feel free to adopt your own convention and style,
) 8 1 '$ 0 ( 1 7$ / , 1 6 , * + 7
but be sure to use it consistently.
The model in Figure 2.5 is still not the last
7TVIEHWLIIX0E
]SYXERH(SGYQIRXEXMSR
word on this example. As shown in later examples,
you can create data tables to see how sensitive
-J]SY[ERX] SYVWTVIEHWLIIXWXSFIYWIHERH] SY
profit is to the inputs, the demand, and the order
[ERX]SYVZEPYIMR]SYVGSQTER]XSVMWI KMZIEPSXSJ
quantity. You can also create charts to show any
XLSYKLXXS] SYVWTVIEHWLIIXPE]SYXERHXLIRHSGY
numerical results graphically. But this is enough for
QIRX] SYV[ SVOGEV IJYPP] *SVPE ]SYX XLMROEFSYX
now. You can see that the model in Figure 2.5 is
[LIXLIVGIV XEMRHEXEEV IFIWXSVMIRXIHMRV S[WSV
now much more readable and flexible than the origGSPYQRW [LIXLIV] SYV[ SVOMWFIXXIVTPEGIHMRE
inal model in Figure 2.1. Q
WMRKPIWLIIXSVMRQYPXMTPIWLIIXW ERHWSSR *SVHSG
YQIRXEXMSR YWIHIWGVMTXMZIPEFIPWERHLIEHMRKW GSPSV
Because good spreadsheet style is so important, the
GSHMRK GIPPGSQQIRXW ERHXI\XFS\IWXSQEOI]SYV
appendix to this chapter discusses a few tools for
WTVIEHWLIIXWQSVIVIEHEFPI -XXEOIWXMQIERHGEVIJYP
editing and documenting your spreadsheet models.
TPERRMRKXSHIWMKRERHXLIRHSGYQIRX] SYVWTVIEH
Use these tools right away and as you progress
WLIIXQSHIPW FYXXLIXMQIMW[ IPPWTIRX %RHMJ] SY
through the book.
GSQIFEGOMREJI[HE]WXSEWTVIEHWLIIXQSHIP]SY
In the rest of this chapter, we discuss a number
HIZIPSTIHERH] SYGER XQEO ILIEHWSVXEMPWSJMX
of interesting examples and introduce important
HSRXFIEJVEMHXSV IHIWMKR]SYV[SVOGSQTPIXIP]
modeling concepts (such as sensitivity analysis),
JVSQXLIKVSYRHYT
important Excel features (such as data tables), and
even some important business concepts (such as

2This

color convention shows up clearly in the Excel files that accompany the book. However, in this two-color
book (shades of gray and blue), it is difficult to see the color-coding scheme. We recommend that you look not only
at the figures in the book, but at the actual Excel files.

net present value). To get the most from these examples, follow along at your own PC,
starting with a blank spreadsheet. It is one thing to read about spreadsheet modeling; it is
quite another to actually GRit!

2.3 COST PROJECTIONS

In this next example, a company wants to project its costs of producing products, given that
material and labor costs are likely to increase through time. We build a simple model and
then use Excels charting capabilities to obtain a graphical image of projected costs.

)<%140)

2.2 P ROJECTING

THE

C OSTS

B OOKSHELVES

WOODWORKS

he Woodworks Company produces a variety of custom-designed wood furniture for its

customers. One favorite item is a bookshelf, made from either cherry or oak. The company knows that wood prices and labor costs are likely to increase in the future. Table 2.1
shows the number of board-feet and labor hours required for a bookshelf, the current costs
per board-foot and labor hour, and the anticipated annual increases in these costs. (The top
row indicates that either type of bookshelf requires 30 board-feet of wood and 16 hours of
labor.) Build a spreadsheet model that enables the company to experiment with the growth
rates in wood and labor costs so that a manager can see, both numerically and graphically,
how the costs of the bookshelves vary in the next few years.

8EFPI Input Data for Manufacturing a Bookshelf

5HVRXUFH
Required per bookshelf
Current unit cost
Anticipated annual cost increase

&KHUU\

2DN

/DERU

30
$5.50
2.4%

30
$4.30
1.7%

16
$18.50
1.5%

%XVLQHVV2EMHFWLYHV To build a model that allows Woodworks to see, numerically and

graphically, how its costs of manufacturing bookshelves increase in the future and to allow
the company to answer what-if questions with this model.
([FHO 2EMHFWLYHV To learn good spreadsheet practices, to enable copying formulas with
the careful use of relative and absolute addresses, and to create line charts from multiple
series of data.

7SPYXMSR
Listing the key variables in a table before developing the actual spreadsheet model is useful, so we will continue to do this in many later examples (see Table 2.2.) This practice
forces you to examine the roles of the variableswhich are inputs, which are decision
variables, and which are outputs. Although the variables and their roles are fairly clear for
this example, later examples will require more thought.

3In later chapters, we simply list the Objective of each example as we did in Example 2.1. However, because
this chapter has been written to enhance basic spreadsheet skills, we separate the business objectives from the
Excel objectives.

'SWX4VSNIGXMSRW

8EFPI Key Variables for the Bookshelf Manufacturing Example

,QSXWYDULDEOHV
2XWSXWYDULDEOHV

Wood and labor requirements per bookshelf, current unit costs of wood and
labor, anticipated annual increases in unit costs
Projected unit costs of wood and labor, projected total bookshelf costs

The reasoning behind the model is straightforward. You first project the unit costs for
wood and labor into the future. Then for any year you multiply the unit costs by the
required numbers of board-feet and labor hours per bookshelf. Finally, you add the wood
and labor costs to obtain the total cost of a bookshelf.

():)034-2+8,) 746)%(7,))813()0
The completed spreadsheet model appears in Figure 2.6 and in the file %RRNVKHOI
&RVWV[OV[.4 You can develop it with the following steps.

*MKYVI
Bookshelf Cost
Model

%P[E]WIRXIVMRTYX
ZEPYIWMRMRTYXGIPPW
ERHXLIRVIJIVXSXLIQ
MR)\GIPJSVQYPEW (S
RSXFYV]MRTYXZEPYIW
MRJSVQYPEW

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

A
B
Projecng bookshelf costs at Woodworks

Projected Bookshelf Costs

Inputs
Requirements per bookshelf
Board-feet required
Labor hours required

Cherry
30
16

Oak
30
16

Costs of wood
Current cost per board-foot
Projected annual increase

Cherry
$5.50
2.4%

Oak
$4.30
1.7%

Cost of labor
Current cost per labor hour
Projected annual increase

$18.50
1.5%

$550.00
$500.00
$450.00
$400.00

Cherry

$350.00

Oak

$300.00
0

Years from Now

Projected costs
Years from now
0
1
2
3
4
5
6

Cost per board-foot

Cherry
Oak
$5.50
$4.30
$5.63
$4.37
$5.77
$4.45
$5.91
$4.52
$6.05
$4.60
$6.19
$4.68
$6.34
$4.76

Cost per hour

Labor
$18.50
$18.78
$19.06
$19.35
$19.64
$19.93
$20.23

Cost per bookshelf

Cherry
Oak
$461.00
$425.00
$469.40
$431.63
$477.96
$438.37
$486.69
$445.21
$495.58
$452.16
$504.65
$459.22
$513.89
$466.39

,QSXWV You should usually enter the inputs for a model in the upper-left corner of a
worksheet as you can see in the shaded ranges in Figure 2.6, using the data from Table 2.1.
We have used our standard convention of coloring inputsthe numbers from the statement
of the problemblue. You can develop your own convention, but the input cells should be
distinguished in some way. Note that the inputs are grouped logically and are explained
with appropriate labels. You should always document your spreadsheet model with informational labels. Also, note that by entering inputs explicitly in input cells, you can UHIHUto
them later in Excel formulas.
'HVLJQRXWSXWWDEOH Plan ahead for how you want to structure your outputs. We created a table where there is a row for every year in the future (year 0 corresponds to the current year), there are three columns for projected unit costs (columns BD), and there are
two columns for projected total bookshelf costs (columns EF). The headings reflect this
design. Of course, this isnt the only possible design, but it works well. The important
point is that you should have VRPHlogical design in mind before diving in.
4 This textbooks essential resource Web site includes templates and completed files for all examples in the book,
where all of the latter have Finished appended to their file names. However, especially in this chapter, we suggest that you start with a blank spreadsheet and follow the step-by-step instructions on your own.

%P[E]WXV]XSSVKERM^I
]SYVWTVIEHWLIIX
QSHIPWSXLEX]SYGER
GST]JSVQYPEWEGVSWW
QYPXMTPIGIPPW

3URMHFWHGXQLWFRVWVRIZRRG The dollar values in the range B19:F25 are all calculated from Excel formulas. Although the logic in this example is straightforward, it is still
important to have a strategy in mind before you enter formulas. In particular, you should
always try to design your spreadsheet so that you can enter a sLQJOHformula and then copy
it. This saves work and avoids errors. For the costs per board-foot in columns B and C,
enter the formula
%
in cell B19 and copy it to cell C19. Then enter the general formula
% %
in cell B20 and copy it to the range B20:C25. We assume you know the rules for absolute
and relative addresses (dollar sign for absolute, no dollar sign for relative), but it takes
some planning to use these so that copying is possible. Make sure you understand why we
made row 10 absolute but column B relative.

8]TMRKHSPPEVWMKRWMR
JSVQYPEWJSVEFWSPYXI
VIJIVIRGIWMWMRIJJMGMIRX
4VIWWXLI *OI]
MRWXIEH

([FHO7LS5HODWLYHDQG$EVROXWH$GGUHVVHVLQ)RUPXODV
5HODWLYHDQGDEVROXWHDGGUHVVHVDUHXVHGLQ([FHO IRUPXODVWRIDFLOLWDWHFRS\LQJ$GROODU
VLJQQH[WWRDFROXPQRUU RZDGGUHVVLQGLFDWHVWKDWWKHDGGU HVVLVDEVROXWHDQGZLOOQRW
FKDQJHZKHQFRS\LQJ7KHODFNRIDGROODUVLJQLQGLFDWHVWKDWWKHDGGU HVVLVUHODWLYHDQG
ZLOOFKDQJHZKHQFRS\LQJ$IWHU\RXVHOHFWDFHOOLQDIRUPXOD\RXFDQSU HVVWKH)N H\
UHSHDWHGO\WRF\FOHWKURXJKWKHUHODWLYHDEVROXWHSRVVLELOLWLHVIRUH[DPSOH %ERWKFRO
XPQDQGU RZUHODWLYH %ERWKFROXPQDQGU RZDEVROXWH %FROXPQU HODWLYH
URZDEVROXWHDQG %FROXPQDEVROXWHURZUHODWLYH
3URMHFWHGXQLWODERUFRVWV
formula
%

To calculate projected hourly labor costs, enter the

in cell D19. Then enter the formula

' %
in cell D20 and copy it down column D.
3URMHFWHGERRNVKHOIFRVWV Each bookshelf cost is the sum of its wood and labor
costs. By a careful use of absolute and relative addresses, you can enter a single formula
for these costsfor all years and for both types of wood. To do this, enter the formula
% %% '
in cell E19 and copy it to the range E19:F25. The idea here is that the units of wood and
labor per bookshelf are always in rows 5 and 6, and the projected unit labor cost is always
in column D, but all other references are relative to allow copying.
&KDUW A chart is a valuable addition to any table of data, especially in the business
world, so charting in Excel is a skill worth mastering. Although not everyone agrees, the
many changes Microsoft made regarding charts in Excel 2007 and 2010 help you create
charts more efficiently and effectively. We illustrate some of the possibilities here, but
we urge you to experiment with other possibilities on your own. Start by selecting the
range E18:F25yes, including the labels in row 18. Next, click on the Line dropdown list
on the Insert ribbon and select the Line with Markers type. You instantly get the basic line
chart you want, with one series for Cherry and another for Oak. Also, when the chart is
selected (that is, it has a border around it), you see three Chart Tools ribbons: Design,
Layout, and Format. The most important button on any of these ribbons is the Select Data
button on the Design ribbon. It lets you choose the ranges of the data for charting in case
'SWX4VSNIGXMSRW

*MKYVI
Select Data
Dialog Box

*MKYVI
Dialog Box for
Changing
Horizontal
Axis Labels

8LIQER]GLEVX
STXMSRWEVIIEWMP]
EGGIWWMFPIJVSQXLI
XLVII'LEVX8SSPW
VMFFSRWMR)\GIP
ERH (SRXFI
EJVEMHXSI\TIVMQIRX
[MXLXLIQ XSTVSHYGI
TVSJIWWMSREPPSSOMRK
GLEVXW

Excels default choices (which are based on the selected range when you create the chart)
are wrong. Click on Select Data now to obtain the dialog box in Figure 2.7. On the left,
you control the series (one series or multiple series) being charted; on the right, you control the data used for the horizontal axis. By selecting E18:F25, you have the series on the
left correct, including the names of these series (Cherry and Oak), but if you didnt, you
could select one of the series and click on Edit to change it. The data on the horizontal
axis is currently the default 1, 2, and so on. To make it the data in column A, click on the
Edit button on the right and select the range A19:A25. (See Figure 2.8.) Your chart is now
correctly labeled and charts the correct data. Beyond this, you can experiment with various formatting options to make the chart even better. For example, we rescaled the vertical axis to start at $300 rather than $0 (right-click on the numbers on the vertical axis and
select Format Axis, or look at the many options on the Axes dropdown list on the Layout
ribbon), and we added a chart title at the top and a title for the horizontal axis at the bottom (see buttons on the Labels group on the Layout ribbon). You can spend a lot of time
fine-tuning chartsmaybe even WRR much timebut professional-looking charts are definitely appreciated.

Using the Model for What-If Questions

) 8 1 '$ 0 ( 1 7$ / , 1 6 , * + 7

The model in Figure 2.6 can now be used to answer

many what-if questions. In fact, many models are
built for the purpose of permitting experimentation
%GLEVXMWX]TMGEPP]QYGLQSVIMRJSVQEXMZIXSEFYWM
with various scenarios. The important point is that
RIWWQEREKIVXLERXLIXEFPISJR YQFIVWMXMWFEWIH
the model has been built in such a way that a manSR (SRXYRHIV IWXMQEXIXLITS [IVSJ)\GIP GLEVXW
ager can enter any desired values in the input cells,
JSVKIXXMRK] SYV TSMRXWEGV SWW ERHMRGPYHIXLIQ
and all of the outputs, including the chart, will
MR] SYVWTV IEHWLIIXQSHIPW[LIRI ZIVTSWWMFPI
update automatically. As a simple example, if the
,S[IZIV FITV ITEVIHXSHSWSQIMR ZIWXMKEXMRKSR
annual percentage increases for wood costs are
]SYVS[R )\GIP SJJIVWEREFYRHERGISJGLEV XX]TIW
twice as high as Woodworks anticipated, you can
ERHGLEVXSTXMSRWXSGLSSWIJV SQ ERHXLI]EVIRSX
enter these higher values in row 10 and immediEPPIUYEPP]WYMXIHXSXIPPMRK]SYVWXSV]
ately see the effect, as shown in Figure 2.9. By
comparing bookshelf costs in this scenario to those
in the original scenario, the projected cost in year 6 for cherry bookshelves, for example,
increases by about 5.5%, from $513.89 to $542.26.

8LI4S[IVSJ'LEVXW

*MKYVI Effect of Higher Increases in Wood Costs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

%GEVIJYPP]GSRWXVYGXIH
QSHIP[MXLRSMRTYX
RYQFIVWFYVMIHMR
JSVQYPEWEPPS[WE
QEREKIVXSERW[IV
QER][LEXMJUYIWXMSRW
[MXLEJI[OI]WXVSOIW

A
B
Projecng bookshelf costs at Woodworks

Projected Bookshelf Costs

Inputs
Requirements per bookshelf
Board-feet required
Labor hours required

Cherry
30
16

Oak
30
16

Costs of wood
Current cost per board-foot
Projected annual increase

Cherry
$5.50
4.8%

Oak
$4.30
3.4%

$600.00
$550.00
$500.00
$450.00
Cherry

$400.00

Oak

$350.00

Cost of labor
Current cost per labor hour
Projected annual increase

$300.00

$18.50
1.5%

Years from Now

Projected costs
Years from now
0
1
2
3
4
5
6

Cost per board-foot

Cherry
Oak
$5.50
$4.30
$5.76
$4.45
$6.04
$4.60
$6.33
$4.75
$6.63
$4.92
$6.95
$5.08
$7.29
$5.26

Cost per hour

Labor
$18.50
$18.78
$19.06
$19.35
$19.64
$19.93
$20.23

Cost per bookshelf

Cherry
Oak
$461.00
$425.00
$473.36
$433.83
$486.17
$442.87
$499.44
$452.13
$513.20
$461.62
$527.46
$471.35
$542.26
$481.32

You should appreciate by now why burying input numbers inside Excel formulas is
such a bad practice. For example, if you had buried the annual increases of wood costs
from row 10 in the formulas in columns B and C, imagine how difficult it would be to
answer the what-if question in the previous paragraph. You would first have to find and
then change all the numbers in the formulas, which is a lot of work. Even worse, it is
likely to lead to errors. Q

2.4 BREAKEVEN ANALYSIS

Many business problems require you to find the appropriate level of some activity. This
might be the level that maximizes profit (or minimizes cost), or it might be the level that
allows a company to break evenno profit, no loss. We discuss a typical breakeven analysis in the following example.

&VIEOIZIR%REP]WMW

)<%140)

2.3 B REAKEVEN A NALYSIS

Q UALITY S WEATERS

he Quality Sweaters Company sells hand-knitted sweaters. The company is planning to

print a catalog of its products and undertake a direct mail campaign. The cost of printing
the catalog is $20,000 plus $0.10 per catalog. The cost of mailing each catalog (including
postage, order forms, and buying names from a mail-order database) is $0.15. In addition, the
company plans to include direct reply envelopes in its mailings and incurs $0.20 in extra
costs for each direct mail envelope used by a respondent. The average size of a customer
order is $40, and the companys variable cost per order (due primarily to labor and material
costs) averages about 80% of the orders valuethat is, $32. The company plans to mail
100,000 catalogs. It wants to develop a spreadsheet model to answer the following questions:
How does a change in the response rate affect profit?
For what response rate does the company break even?
If the company estimates a response rate of 3%, should it proceed with the mailing?
How does the presence of uncertainty affect the usefulness of the model?

%XVLQHVV2EMHFWLYHV To create a model to determine the companys profit and to see how
sensitive the profit is to the response rate from the mailing.
([FHO 2EMHFWLYHV To learn how to work with range names, to learn how to answer whatif questions with one-way data tables, to introduce Excels Goal Seek tool, and to learn
how to document and audit Excel models with cell comments and the auditing toolbar.

7SPYXMSR
The key variables appear in Table 2.3. Note that we have designated all variables as input
variables, decision variables, or output variables. Furthermore, there is typically a key output variable, in this case, profit, that is of most concern. (In the next few chapters, we refer
to it as the REMHFWLYH variable.) Therefore, we distinguish this key output variable from the
other output variables that we calculate along the way.

8EFPI Key Variables in Quality Sweaters Problem

,QSXWYDULDEOHV
'HFLVLRQYDULDEOH
.H\RXWSXWYDULDEOH
2WKHURXWSXWYDULDEOHV

%HSTXWSQIPE]SYX
ERH JSVQEXXMRK
GSRZIRXMSRW IZIRMJ
XLI]HMJJIVJVSQSYVW
XSQEOI]SYV
WTVIEHWLIIXWVIEHEFPI
ERHIEW]XSJSPPS[

Various unit costs, average order size, response rate

Number mailed
Profit
Number of responses, revenue, and cost totals

The logic for converting inputs and decision variable into outputs is straightforward.
After you do this, you can investigate how the response rate affects the profit with a sensitivity analysis.
The completed spreadsheet model appears in Figure 2.10. (See the file %UHDNHYHQ
$QDO\VLV[OV[.) First, note the clear layout of the model. The input cells are colored blue,
they are separated from the outputs, headings are boldfaced, several headings are indented,
numbers are formatted appropriately, and a list to the right spells out all range names we
have used. (See the next Excel Tip on how to create this list.) Also, following the convention we use throughout the book, the decision variable (number mailed) is colored red, and
the bottom-line output (profit) is colored gray.

*MKYVI Quality Sweaters Model

1
2
3
4
5
6
7
8
9
10
11
12
13

A
Quality Sweaters direct mail model

Catalog inputs
Fixed cost of prinUng
Variable cost of prinUng mailing

$20,000
$0.25

Decision variable
Number mailed

100000

Order inputs
Average order
Variable cost per order

$40
$32.20

;IVIJIVXSXLMWEWXLI
'VIEXIJVSQ7IPIGXMSR
WLSVXGYX -J]SYPMOIMX
]SYGERKIXXLIHMEPSK
FS\MR*MKYVI
IZIRQSVIUYMGOP]
TVIWW'XVP7LMJX*

Model of responses
Response rate
Number of responses
Model of revenue, costs, and profit
Total Revenue
Fixed cost of prinUng
Total variable cost of prinUng mailing
Total variable cost of orders
Total cost
Profit

8%
8000

$320,000
$20,000
$25,000
$257,600
$302,600
$17,400

G
Range names used
Average_order
Fixed_cost_of_prinUng
Number_mailed
Number_of_responses
Profit
Response_rate
Total_cost
Total_Revenue
Variable_cost_of_prinUng_mailing
Variable_cost_per_order

=Model!$B$11
=Model!$B$4
=Model!$B$8
=Model!$E$5
=Model!$E$13
=Model!$E$4
=Model!$E$12
=Model!$E$8
=Model!$B$5
=Model!$B$12

([FHO 7LS &UHDWLQJ5DQJH1DPHV

7RFUHDWHDU DQJHQDPHIRUDU DQJHRIFHOOVZKLF KFRXOGEHDVLQJOHFHOOKLJKOLJKWWKH

FHOOVFOLFNLQWKH1DPH%R[MXVWWRWKHOHIWRIWKH) RUPXOD%DUDQGW\SHDU DQJHQDPH
$OWHUQDWLYHO\LIDFROXPQRUU RZRIODEHOVDSSHDU VQH[WWRWKHFHOOVWREHU DQJHQDPHG
\RXFDQXVHWKHVHODEHOVDVWKHU DQJHQDPHV7RGRWKLVKLJKOLJKWWKHODEHOVDQGWKHFHOOV
WREHQDPHGIRUH [DPSOH$%LQ) LJXUHVHOHFW&U HDWHIU RP6HOHFWLRQRQWKH
)RUPXODVULEERQDQGPDN HVXUHWKHDSSURSULDWHER[LQWKHU HVXOWLQJGLDORJER[VHH)LJ
XUHLVFKHFNHG7KHODEHOVLQRXUH [DPSOHDUHWRWKHOHIWRIWKHFHOOVWREHQDPHGVR
WKH/HIWFROXPQER[VKRXOGEHFKHFNHG7KLVLVDYHU\TXLFNZD\WRFUHDWHUDQJHQDPHVDQG
ZHGLGLWIRUDOOUDQJHQDPHVLQWKHH[DPSOH,QIDFWE\NHHSLQJ\RXUILQJHURQWKH&WUONH\
\RXFDQVHOHFWPXOWLSOHUDQJHV5 $IWHUDOO\RXUUDQJHVDUHVHOHFWHG\RXFDQVRPHWLPHVFUH
DWHDOO\RXUU DQJHQDPHVLQRQHVWHS1RWHWKDWLIDODEHOFRQWDLQVDQ\LOOH JDOUDQJH
QDPHFKDUDFWHUVVXFKDVDVSDFHWKHLOOHJDOFKDUDFWHUVDUHFRQYHUWHGWRXQGHUVFRUHV

*MKYVI
Range Name Create
Dialog Box

-J]SYPMOIXLMWXMT ]SY
GERTIVJSVQMXIZIR
JEWXIV TVIWWXLI*
OI] XSFVMRKYTXLI
4EWXI2EQIHMEPSK
FS\ 8LMW[SVOWSRP]
MJXLIVIMWEXPIEWXSRI
VERKIREQIMRXLI
[SVOFSSO

([FHO 7LS 3DVWLQJ5DQJH1DPHV

,QFOXGLQJDOLVWRIWKHU DQJHQDPHVLQ\RXUVSU HDGVKHHWLVXVHIXOIRUGRFXPHQWDWLRQSXU

SRVHV7RGRWKLVVHOHFWDFHOOVXF KDVFHOO*LQ) LJXUHVHOHFWWKH8VHLQ)RUPXOD
GURSGRZQOLVWIURPWKH)RUPXODVULEERQDQGWKHQFOLFNRQWKH3DVWH/LVWRSWLRQ<RXJHWD
OLVWRIDOOUDQJHQDPHVDQGWKHLUFHOODGGUHVVHV+RZHYHULI\RXFKDQJHDQ\RIWKHVHUDQJH
QDPHVGHOHWHRQHIRUH[DPSOHWKHSDVWHOLVWGRHVQRWXSGDWHDXWRPDWLFDOO\\RXKDYHWR
FUHDWHLWDJDLQ

5Many users apparently believe range names are more work than they are worth. This shortcut for creating range
names helps to remedy this problem.

&VIEOIZIR%REP]WMW

DEVELOPING THE SPREADSHEET MODEL

To create this model, you can proceed through the following steps.
+HDGLQJVDQGUDQJHQDPHV We have named a lot of cells, more than you might
want to name, but you will see their value when you create formulas. In general, we
strongly support range names, but it is possible to go overboard. You can waste time naming ranges that do not UHDOO\need to be named. Of course, you can use the Create from
Selection shortcut described previously to speed up the process.6
9DOXHVRILQSXWYDULDEOHVDQGWKHGHFLVLRQYDULDEOH Enter these values and format
them appropriately. As usual, we have used our blue/red/gray color-coding scheme. Note
that the number mailed has been designated as a GHFLVLRQvariable, not as an input variable (and it is colored red, not blue). This is because the company gets to choose the value
of this variable. Finally, note that some of the values have been combined in the statement
of the problem. For example, the $32.20 in cell B12 is really 80% of the $40 average order
size, plus the $0.20 per return envelope. To document this process, comments appear in a
few cells, as shown in Figure 2.12.
*MKYVI Cell Comments in Model

1
2
3
4
5
6
7
8
9
10
11
12
13

A
Great Threads direct mail model

Catalog inputs
Fixed cost of prinUng
Variable cost of prinUng mailing

$20,000
$0.25

Decision variable
Number mailed

100000

Order inputs
Average order
Variable cost per order

Model of responses
Includes $0.10
for
Response
rate
printing and $0.15 for
mailing eachNumber
catalog of responses

Model of revenue, costs, and profit

Total Revenue
Fixed cost of prinUng
Includes 80%Total
of thevariable
average cost of prinUng mailing
$0.20 percost of orders
$40 $40 order size,Toplus
tal variable
return envelope
$32.20
Total cost
Profit

8%
8000

$320,000
$20,000
$25,000
$257,600
$302,600
$17,400

G
Range names used
verdo
age_order
Trial value,Awill
sensitivity analysis
on
Fixed_cost_of_prinUng
Number_mailed
Number_of_responses
Profit
Response_rate
Total_cost
Total_Revenue
Variable_cost_of_prinUng_mailing
Variable_cost_per_order

=Model!$B$11
=Model!$B$4
=Model!$B$8
=Model!$E$5
=Model!$E$13
=Model!$E$4
=Model!$E$12
=Model!$E$8
=Model!$B$5
=Model!$B$12

([FHO 7LS ,QVHUWLQJ&HOO&RPPHQWV

,QVHUWLQJFRPPHQWVLQFHOOVLVDJU HDWZD\WRGRFXPHQW\RXUVSU HDGVKHHWPRGHOVZLWKRXW

LQWURGXFLQJH[FHVVLYHFOXWWHU7RHQWHUDFRPPHQWLQDFHOOULJKWFOLFNRQWKHFHOOVHOHFWWKH
,QVHUW&RPPHQWLWHPDQGW\SH\RXUFRPPHQW7KLVFU HDWHVDOLWWOHU HGPDUNLQWKHFHOO
LQGLFDWLQJDFRPPHQWDQG\RXFDQVHHWKHFRPPHQWE\U HVWLQJWKHFXU VRURYHUWKHFHOO
:KHQDFHOOFRQWDLQVDFRPPHQW\RXFDQHGLWRUGHOHWHWKHFRPPHQWE\ULJKWFOLF NLQJRQ
WKHFHOODQGVHOHFWLQJWKHDSSURSULDWHLWHP,I\RXZDQWDOOWKHFHOOFRPPHQWVWREHYLVLEOH
IRUH [DPSOHLQDSULQWRXWDVLQ) LJXUHFOLF NRQWKH) LOHWDERU2I ILFHE XWWRQLQ
([FHOWKHQRQ2SWLRQV([FHO2SWLRQVLQ([FHOWKHQRQWKH$GYDQFHGOLQN
DQGVHOHFWWKH&RPPHQW ,QGLFDWRURSWLRQIURPWKH'LVSOD\JURXS1RWHWKDWWKH,QGLFDWRU
2QO\RSWLRQLVWKHGHIDXOW

',%2+)7 -2 )<')0

%JXIV1MGVSWSJXKSXEPPSJYWYWIHXSXLI3JJMGIFYXXSRMRXLIYTTIVPIJXGSVRIVSJEPP3JJMGI
ETTPMGEXMSRW MXW[MXGLIHXSE*MPIXEFMR3JJMGI8LIQIRYWXVYGXYVIYRHIVXLMW*MPIXEFMW
WPMKLXP]HMJJIVIRXJVSQXLIWXVYGXYVIYRHIVXLI3JJMGIFYXXSR FYXXLIJYRGXMSREPMX]MWFEWMGEPP]XLI
WEQI -RTEVXMGYPEV XLMWMW[LIVI]SYKSXSGLERKIQSWXSJXLI)\GIPSTXMSRW

6 We have heard of one company that does not allow any formulas in its corporate spreadsheets to include cell references; they must all reference range names. This is probably too extreme, but that companys formulas are certainly easy to read.

0RGHOWKHUHVSRQVHV You have not yet specified the response rate to the mailing, so
enter DQ\reasonable value, such as 8%, in the Response_rate cell. You will perform sensitivity on this value later on. Then enter the formula
1XPEHUBPDLOHG 5HVSRQVHBUDWH
in cell E5. (Are you starting to see the advantage of range names?)
0RGHOWKHUHYHQXHFRVWVDQGSURILWV Enter the formula
1XPEHUBRIBUHVSRQVHV $YHUDJHBRUGHU
in cell E8, enter the formulas
)L[HGBFRVWBRIBSULQWLQJ
9DULDEOHBFRVWBRIBSULQWLQJBPDLOLQJ 1XPEHUBPDLOHG
and
1XPEHUBRIBUHVSRQVHV 9DULDEOHBFRVWBSHUBRUGHU
in cells E9, E10, and E11, enter the formula
680((
in cell E12, and enter the formula
7RWDOBUHYHQXH7RWDOBFRVW
in cell E13. These formulas should all be self-explanatory, especially because of the range
names used.

([FHO7LS (QWHULQJ)RUPXODVZLWK5DQJH1DPHV

7RHQWHUDIRUPXODWKDWFRQWDLQVUDQJHQDPHV\RXGRnot KDYHWRW\SHWKHIXOOUDQJHQDPHV
<RXDFWXDOO\KDYHWZRFRQYHQLHQWRSWLRQV2QH\RXFDQSRLQWWRWKHFHOOVDQGUDQJHQDPHV
ZLOODSSHDULQ\RXUIRUPXODV2UWZR\RXFDQVWDUWW\SLQJWKHU DQJHQDPHLQWKHIRUPXOD
DQGDIWHUDIHZOHWWHUV([FHOZLOOVKRZ\RXDOLVW\RXFDQFKRRVHIURP

Forming a One-Way Data Table

*MKYVI
Data Table for Profit

Now that a basic model has been created, the questions posed by the company can be
answered. For question 1, you can form a one-way data table to show how profit varies with
the response rate as shown in Figure 2.13. Data tables are used often in this book, so make
sure you understand how to create them. We will walk you through the procedure once or
twice, but from then on, you are on your own. First, enter a sequence of trial values of the
response rate in column A, and enter a link to profit in cell B17 with the formula 3URILW
This cell is shaded for emphasis, but this isnt necessary. (In general, other outputs could be
part of the table, and they would be placed in columns C, D, and so on. There would be a
link to each output in row 17.) Finally, highlight the entire table range, A17:B27, and select
Data Table from the What-If Analysis dropdown list on the Data ribbon to bring up the
A
B
15 Queson 1 - sensivity of prot to response rate
16
Response rate
Profit
17
$17,400
18
1%
-$37,200
19
2%
-$29,400
20
3%
-$21,600
21
4%
-$13,800
22
5%
-$6,000
23
6%
$1,800
24
7%
$9,600
25
8%
$17,400
26
9%
$25,200
27
10%
$33,000

Prot versus Response Rate

$40,000
$20,000
Prot

(EXEXEFPIWEVIEPWS
GEPPIH[LEXMJXEFPIW
8LI]PIX]SYWII[LEX
LETTIRWXSWIPIGXIH
SYXTYXWEWWIPIGXIH
MRTYXWGLERKI

$0
-$20,000

1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

-$40,000
-$60,000

Response Rate

&VIEOIZIR%REP]WMW

*MKYVI
Data Table Dialog
Box

dialog box in Figure 2.14. Fill it in as shown to indicate that the only input, Response_rate,
is listed along a column. (You can enter either a range name or a cell address in this dialog
box. The easiest way is to point to the cell.)
When you click on OK, Excel substitutes each response rate value in the table into the
Response_rate cell, recalculates profit, and reports it in the table. For a final touch, you can
create a chart of the values in the data table. (To do this, highlight the A18:B27 range and
select the type of chart you want from the Insert ribbon. Then you can fix it up by adding
titles, removing the legend, and making other modifications to suit your taste.)

([FHO 7RRO 2QH:D\'DWD7DEOH

$ RQHZD\ GDWD WDEOH DOORZV \RX WR VHH KRZ RQH RU PRUH RXWSXW YDULDEOHV YDU\ DV D VLQJOH LQSXW
YDULDEOH YDULHV RYHU D VHOHFWHG UDQJH RI YDOXHV 7KHVH LQSXW YDOXHV FDQ EH DUUDQJHG YHUWLFDOO\ LQ
D FROXPQ RU KRUL]RQWDOO\ LQ D URZ :H ZLOO H[SODLQ RQO\
WKH YHUWLFDO DUUDQJHPHQW EHFDXVH LW LV WKH PRVW FRP
) 8 1 '$ 0 ( 1 7$ / , 1 6 , * + 7
PRQ 7R FUHDWH WKH WDEOH HQWHU WKH LQSXW YDOXHV LQ D
FROXPQ UDQJH VXFK DV $$ RI )LJXUH DQG
8LI4S[IVSJ(EXE
8EFPIW
HQWHU OLQNV WR RQH RU PRUH RXWSXW FHOOV LQ FROXPQV WR
9RJSVXYREXIP] QER])\GIP YWIVWQSWXSJXLIQ# EV I
WKH ULJKW DQG RQH URZ DERYH WKH LQSXWV DV LQ FHOO %
YRE[EVISJHEXEXEFPIW [LMGLEV IEQSRKXLIQSWX
RI )LJXUH 7KHQ KLJKOLJKW WKH HQWLUH WDEOH EHJLQ
TS[IVJYPERHYWIJYPXSSPW)\GIP LEWXSSJJIV %JXIV]SY
QLQJ ZLWK WKH XSSHUOHIW EODQN FHOO $ LQ WKH ILJXUH
LEZIHI ZIPSTIHEQSHIPXLEXV IPEXIWMRTYXWXSSYX
VHOHFW 'DWD 7DEOH IURP WKH :KDW,I $QDO\VLV GURSGRZQ
TYXW ]SYGERXLIRFYMPHHEXEXEFPIWMREQEXXIVSJWIG
OLVW RQ WKH 'DWD ULEERQ DQG ILOO LQ WKH UHVXOWLQJ GLDORJ
SRHWXSWIILS[XLISYXTYXWZEV ]EWOI]MRTYXWZEV]
ER[ DV LQ )LJXUH /HDYH WKH 5RZ ,QSXW FHOO EODQN
SZIVWSQIVERKI (EXEXEFPIWEV IXLITIVJ IGXQIERW
DQG XVH WKH FHOO ZKHUH WKH RULJLQDO YDOXH RI WKH LQSXW
JSVERW [IVMRKEPEVKIR YQFIVSJ[LEXMJUYIWXMSRW
YDULDEOH OLYHV DV WKH &ROXPQ ,QSXW FHOO :KHQ \RX
UYMGOP]ERHIEWMP]
FOLFN RQ 2. HDFK YDOXH LQ WKH OHIW FROXPQ RI WKH WDEOH
LV VXEVWLWXWHG LQWR WKH FROXPQ LQSXW FHOO WKH VSUHDG
VKHHW UHFDOFXODWHV DQG WKH UHVXOWLQJ YDOXH RI WKH RXWSXW LV SODFHG LQ WKH WDEOH $OVR LI \RX FOLFN
DQ\ZKHUH LQ WKH ERG\ RI WKH WDEOH %% LQ WKH ILJXUH \RX ZLOO VHH WKDW ([FHO KDV HQWHUHG
WKH 7$%/( IXQFWLRQ WR UHPLQG \RX WKDW D GDWD WDEOH OLYHV KHUH 1RWH WKDW WKH FROXPQ LQSXW FHOO
PXVW EH RQ WKH VDPH ZRUNVKHHW DV WKH WDEOH LWVHOI RWKHUZLVH ([FHO LVVXHV DQ HUURU
As the chart indicates, profit increases in a linear manner as the response rate varies. More
specifically, each percentage point increase in the response rate increases profit by $7800. Here
is the reasoning. Each percentage point increase in response rate results in 100,000(0.01) 5
1000 more orders. Each order yields a revenue of $40, on average, but incurs a variable cost of
$32.20. The net gain in profit is $7.80 per order, or $7800 for 1000 orders.
8LITYVTSWISJXLI
+SEP7IIOXSSPMWXS
WSPZISRIIUYEXMSRMR
SRIYRORS[R -XMW
YWIHLIVIXS JMRHXLI
VIWTSRWIVEXIXLEX
QEOIWTVSJMXIUYEP
XS

97-2+ +3%0 7))/

From the data table, you can see that profit changes from negative to positive when the
response rate is somewhere between 5% and 6%. Question 2 asks for the exact breakeven
point. You could find this by trial and error, but it is easier to use Excels Goal Seek tool.
Essentially, Goal Seek is used to solve a VLQJOH equation in a VLQJOH unknown. Here, the
equation is Profit=0, and the unknown is the response rate. In Excel terminology, the
unknown is called the FKDQJLQJ FHOO because you can change it to make the equation true.

*MKYVI
Goal Seek Dialog
Box

To implement Goal Seek, select Goal Seek from the What-If Analysis dropdown list on the
Data ribbon and fill in the resulting dialog box as shown in Figure 2.15. (Range names or
cell addresses can be used in the top and bottom boxes, but a QXPEHU must be entered in the
middle box.) After you click on OK, the Response_rate and Profit cells have values 5.77%
and $0. In words, if the response rate is 5.77%, Great Threads breaks even. If the response
rate is greater than 5.77%, the company makes money; if the rate is less than 5.77%, the
company loses money. Of course, this assumes that the company mails 100,000 catalogs. If
it sends more or fewer catalogs, the breakeven response rate will change.

([FHO 7RRO *RDO6HHN

7KH SXUSRVH RI WKH *RDO 6HHN WRRO LV WR VROYH RQH HTXDWLRQ LQ RQH XQNQRZQ 6SHFLILFDOO\ *RDO
6HHN DOORZV \RX WR YDU\ D VLQJOH input FHOO WR IRUFH D VLQJOH output FHOO WR D VHOHFWHG YDOXH 7R XVH
LW VHOHFW *RDO 6HHN IURP WKH :KDW,I $QDO\VLV GURSGRZQOLVW RQ WKH 'DWD ULEERQ DQG ILOO LQ WKH
UHVXOWLQJ GLDORJ ER[ LQ )LJXUH (QWHU D UHIHUHQFH WR WKH RXWSXW FHOO LQ WKH 6HW FHOO ER[ HQWHU
WKH QXPHULF YDOXH \RX ZDQW WKH RXWSXW FHOO WR HTXDO LQ WKH 7R YDOXH ER[ DQG HQWHU D UHIHUHQFH WR
WKH LQSXW FHOO LQ WKH %\ FKDQJLQJ FHOO ER[ 1RWH WKDW *RDO 6HHN VRPHWLPHV VWRSV ZKHQ WKH 6HW FHOO
LV FORVH EXW QRW exactly HTXDO WR WKH GHVLUHG YDOXH 7R LPSURYH *RDO 6HHNV DFFXUDF\ FOLFN RQ
WKH )LOHWDEWKH2IILFH EXWWRQLQ([FHO WKHQ 2SWLRQV([FHO2SWLRQVLQ([FHO DQG
WKHQ WKH )RUPXODV OLQN 7KHQ FKHFN WKH (QDEOH LWHUDWLYH FDOFXODWLRQ ER[ DQG UHGXFH 0D[LPXP
&KDQJH WR DQ\ GHVLUHG OHYHO RI SUHFLVLRQ :H FKRVH D SUHFLVLRQ OHYHO RI )RU WKLV OHYHO
RI SUHFLVLRQ *RDO 6HHN VHDUFKHV XQWLO SURILW LV ZLWKLQ RI WKH GHVLUHG YDOXH

Limitations of the Model

0EXIVGLETXIVW
IWTIGMEPP]'LETXIVW
XLVSYKL HIEP
I\TPMGMXP][MXL
YRGIVXEMRX]

Question 3 asks whether the company should proceed with the mailing if the response rate
is only 3%. From the data table (see Figure 2.13), the apparent answer is no, because profit
is negative. However, like many companies, we are taking a short-term view with this reasoning. The model does not include the fact that many customers who respond to direct
mail will UHRUGHUin the future. The company nets $7.80 per order. If each of the respondents ordered two more times, say, the company would earn 3000($7.80)(2) = $46,800
more than appears in the model, and profit would then be positive. The moral is that managers must look at the long-term impact of their decisions. However, if you want to incorporate the long term explicitly into the model, you must build a more complex model.
Finally, question 4 asks about the impact of uncertainty in the model. Obviously, not all
model inputs are known with certainty. For example, the size of an order is not always
$40it might range, say, from $10 to $100. When there is a high degree of uncertainty
about model inputs, it makes little sense to talk about WKH profit level or WKH breakeven
response rate. It makes more sense to talk about the SUREDELOLW\ that profit will have a certain
value or the SUREDELOLW\ that the company will break even. You will see how this can be done
in the following example and in many more such examples in Chapters 10 through 12.

Using the Formula Auditing Tool

The model in this example is fairly small and simple. Even so, you can use a handy Excel
feature to see how all the parts fit together. This is the Formula Auditing tool, which is
available on the Formulas ribbon. See Figure 2.16.
&VIEOIZIR%REP]WMW

*MKYVI
Formula Auditing
Toolbar
8LI*SVQYPE%YHMXMRK
XSSPMWMRHMWTIRWEFPI
JSVYRXERKPMRKXLIPSKMG
MREWTVIEHWLIIX
IWTIGMEPP]MJWSQISRI
IPWIHIZIPSTIHMX

*MKYVI
Dependents of
Number_of_
responses Cell

*MKYVI
Precedents of
Total_revenue Cell

The Trace Precedents and Trace Dependents buttons are probably the most useful
buttons in this group. To see which formulas have direct links to the Number_of_
responses cell, select this cell and click on the Trace Dependents button. Arrows are
drawn to each cell that directly depends on the number of responses, as shown in
Figure 2.17. Alternatively, to see which cells are used to create the formula in the Total_
revenue cell, select this cell and click on the Trace Precedents button. Now you see that
the Average_order and Number_of_responses cells are used directly to calculate revenue,
as shown in Figure 2.18. Using these two buttons, you can trace your logic (or someone
elses logic) as far backward or forward as you like. When you are finished, just click on
the Remove Arrows button.

1
2
3
4
5
6
7
8
9
10
11
12
13

A
Great Threads direct mail model

Catalog inputs
Fixed cost of prinUng
Variable cost of prinUng mailing

$20,000
$0.25

Decision variable
Number mailed

100000

Order inputs
Average order
Variable cost per order

$40
$32.20

A
Great Threads direct mail model

Catalog inputs
Fixed cost of prinUng
Variable cost of prinUng mailing

$20,000
$0.25

Decision variable
Number mailed

100000

Order inputs
Average order
Variable cost per order

$40
$32.20

Model of responses
Response rate
Number of responses

8%
8000

Model of revenue, costs, and profit

Total Revenue
Fixed cost of prinUng
Total variable cost of prinUng mailing
Total variable cost of orders
Total cost
Profit

$320,000
$20,000
$25,000
$257,600
$302,600
$17,400

8%
8000

$320,000
$20,000
$25,000
$257,600
$302,600
$17,400

([FHO 7RRO )RUPXOD$XGLWLQJ7RROEDU

7KH IRUPXOD DXGLWLQJ WRROEDU DOORZV \RX WR VHH GHSHQGHQWV RI D VHOHFWHG FHOO ZKLFK
FHOOV KDYH IRUPXODV WKDW UHIHUHQFH WKLV FHOO RU SUHFHGHQWV RI D JLYHQ FHOO ZKLFK FHOOV
DUH UHIHUHQFHG LQ WKLV FHOOV IRUPXOD ,Q IDFW \RX FDQ HYHQ VHH GHSHQGHQWV RU SUHFHGHQWV
WKDW UHVLGH RQ D GLIIHUHQW ZRUNVKHHW ,Q WKLV FDVH WKH DXGLWLQJ DUURZV DSSHDU DV GDVKHG
OLQHV DQG SRLQW WR D VPDOO VSUHDGVKHHW LFRQ %\ GRXEOHFOLFNLQJ RQ WKH GDVKHG OLQH \RX
FDQ VHH D OLVW RI GHSHQGHQWV RU SUHFHGHQWV RQ RWKHU ZRUNVKHHWV 7KHVH WRROV DUHHVSHFLDOO\

XVHIXO IRU XQGHUVWDQGLQJ KRZ VRPHRQH HOVHV VSUHDGVKHHW ZRUNV 8QOLNH LQ SUH
YHUVLRQV RI ([FHO WKH )RUPXOD $XGLWLQJ WRROV LQ([FHODQGDUH FOHDUO\ YLVLEOH
RQ WKH )RUPXODV ULEERQ Q

13()0-2+ -779)7
=SYGERTPEGIGLEVXW
SRXLIWEQI[SVOWLIIX
EWXLIYRHIVP]MRKHEXE
SV SRWITEVEXIGLEVX
WLIIXW8LIGLSMGIMW
E QEXXIVSJTIVWSREP
TVIJIVIRGI

Is the spreadsheet layout in Figure 2.12 the best possible layout? This question is not too
crucial because this model is so small. However, we have put all the inputs together (usually a good practice), and we have put all the outputs together in a logical order. You might
want to put the answers to questions 1 and 2 on separate worksheets, but with such a small
model, it is arguably better to keep everything on a single worksheet. We generally avoid
separate worksheets unless things start getting bigger and more complex.
One other issue is the placement of the chart. From the Chart Tools Design ribbon,
you can click on the Move Chart button to select whether you want to place the chart on the
worksheet (floating above the cells) or on a separate chart sheet that has no rows or
columns. This choice depends on your personal preferenceneither choice is necessarily
better than the otherbut for this small model, we favor keeping everything on a single
worksheet.
Finally, we could have chosen the number mailed, rather than the response rate, as the
basis for a sensitivity analysis. A sensitivity analysis is typically based on an uncertain
input variable, such as the response rate, or a decision variable that the decision maker controls. Fortunately, there is no limit to the number of data tables you can create for a particular model. Q

PROBLEMS
6ROXWLRQVIRUSUREOHPVZKRVHQXPEHUVDSSHDUZLWKLQD
FRORUHGER[FDQEHIRXQGLQWKH6WXGHQW6ROXWLRQV)LOHV
5HIHUWRWKLVERRNVSUHIDFHIRUSXUFKDVHLQIRUPDWLRQ

Skill-Building Problems

The sensitivity analysis in the Quality Sweaters example was on the response rate. Suppose now that the
response rate is NQRZQto be 8%, and the company
wants to perform a sensitivity analysis on the number
mailed. After all, this is a variable under direct control
of the company. Create a one-way data table and a corresponding line chart of profit versus the number
mailed, where the number mailed varies from 80,000
to 150,000 in increments of 10,000. Does it appear,
from the results you see here, that there is an optimal
number to mail, from all possible values, that maximizes profit? Write a concise memo to management
about your results.
Continuing the previous problem, use Goal Seek for
HDFKvalue of number mailed (once for 80,000, once
for 90,000, and so on). For each, find the response rate

that allows the company to break even. Then chart

these values, where the number mailed is on the horizontal axis, and the breakeven response rate is on the
vertical axis. Explain the behavior in this chart in a
brief memo to management.

In the Quality Sweaters model, the range E9:E11 does

not have a range name. Open your completed Excel
file and name this range &RVWV Then look at the formula in cell E12. It does QRWautomatically use the new
range name. Modify the formula so that it does. Then
click on cell G4 and paste the new list of range names
over the previous list.

Skill-Extending Problem

As the Quality Sweaters problem is now modeled, if

all inputs remain fixed except for the number mailed,
profit will increase indefinitely as the number mailed
increases. This hardly seems realisticthe company
could become infinitely rich. Discuss realistic ways to
modify the model so that this unrealistic behavior is
eliminated.

&VIEOIZIR%REP]WMW

2.5 ORDERING WITH QUANTITY DISCOUNTS

AND DEMAND UNCERTAINTY
In the following example, we again attempt to find the appropriate level of some activity:
how much of a product to order when customer demand for the product is uncertain. Two
important features of this example are the presence of quantity discounts and the explicit
use of probabilities to model uncertain demand. Except for these features, the problem is
very similar to the one discussed in Example 2.1.

)<%140)

2.4 O RDERING

WITH

Q UANTITY D ISCOUNTS

S AM S B OOKSTORE

ams Bookstore, with many locations across the United States, places orders for all of
the latest books and then distributes them to its individual bookstores. Sams needs a
model to help it order the appropriate number of any title. For example, Sams plans to
order a popular new hardback novel, which it will sell for $30. It can purchase any number
of this book from the publisher, but due to quantity discounts, the unit cost for all books it
orders depends on the number ordered. Specifically, if the number ordered is less than
1000, the unit cost is $24. After each 1000, the unit cost drops: to $23 for at least 1000
copies, to $22.25 for at least 2000, to $21.75 for at least 3000, and to $21.30 (the lowest
possible unit cost) for at least 4000. For example, if Sams orders 2500 books, its total cost
is $22.25(2500) 5 $55,625. Sams is very uncertain about the demand for this bookit
estimates that demand could be anywhere from 500 to 4500. Also, as with most hardback
novels, this one will eventually come out in paperback. Therefore, if Sams has any hardbacks left when the paperback comes out, it will put them on sale for $10, at which price,
it believes all leftovers will be sold. How many copies of this hardback novel should Sams
order from the publisher?
%XVLQHVV2EMHFWLYHV To create a model to determine the companys profit, given fixed
values of demand and the order quantity, and then to model the demand uncertainty explicitly and to choose the expected profit-maximizing order quantity.

([FHO2EMHFWLYHV To learn how to build in complex logic with IF formulas, to get online
help about Excel functions with the I[ button, to learn how to use lookup functions, to see
how two-way data tables allow you to answer more extensive what-if questions, and to
learn about Excels SUMPRODUCT function.

7SPYXMSR
The key variables for this model appear in Table 2.4. The primary modeling tasks are (1) to
show how any combination of demand and order quantity determines the number of units
sold, both at the regular price and at the leftover sale price, and (2) to calculate the total
ordering cost for any order quantity. After you accomplish these tasks, you can model the
uncertainty of demand explicitly and then find the optimal order quantity.
8EFPI Key Variables for Sams Bookstore Problem
,QSXWYDULDEOHV
8QFHUWDLQYDULDEOH
'HFLVLRQYDULDEOH
.H\RXWSXWYDULDEOH
2WKHURXWSXWYDULDEOHV

Unit prices, table of unit costs specifying quantity discount structure

Demand
Order quantity
Profit
Units sold at each price, revenue, and cost totals

The first step is to develop a spreadsheet model to calculate Sams profit for any order
quantity and any possible demand. Then you can perform a sensitivity analysis to see how
profit depends on these two quantities. Finally, you can decide how Sams might choose
the optimal order quantity.

DEVELOPING THE SPREADSHEET MODEL

;LIRIZIVXLIXIVQ
XVMEPZEPYIMWYWIHJSV
ERMRTYXSVEHIGMWMSR
ZEVMEFPI ]SYGERFI
WYVIXLEX[I[MPPJSPPS[
YT[MXLEHEXEXEFPI
SVMRPEXIVGLETXIVW
F]VYRRMRK7SPZIVXS
STXMQM^I

The profit model appears in Figure 2.19. (See the file 4XDQWLW\'LVFRXQWV[OV[.) Note that
the order quantity and demand in the Order_quantity and Demand cells are trial values.
(Comments in these cells are a reminder of this.) You can put any values in these cells, just
to test the logic of the model. The Order_quantity cell is colored red because the company
can choose its value. In contrast, the Demand cell is colored green here and in later chapters to indicate that this input value is uncertain and is being treated explicitly as such.
Also, note that a table is used to indicate the quantity discounts cost structure. You can use
the following steps to build the model.

*MKYVI Sams Profit Model

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

A
B
Ordering decision with quanty discounts
Inputs
Unit cost - see table to right
Regular price
LeUover price

$30
$10

Decision variable
Order quantity

2500

Uncertain quanty
Demand
Demand

2000
2000

Prot model
Units sold at regular price
Units sold at leUover price
Revenue
Cost
Profit

Quantity discount structure

At least Unit cost
0
$24.00
1000
$23.00
2000
$22.25
3000
$21.75
4000
$21.30

H
I
J
Range names used:
Cost
CostLookup
Demand
Leftover_price
Order_quantity
Probabilities
Profit
Regular_price
Revenue
Units_sold_at_leUover_price
Units_sold_at_regular_price
Units_sold_at_regular_price

K
=Model!$B$18
l!$B$18
=Model!$D$5:$E$9
=Model!$B$12
=Model!$B$6
=Model!$B$9
=Model!$B$35:$J$35
=Model!$B$19
=Model!$B$5
=Model!$B$17
=Model!$B$16
=Model!$B$15
Model!$B$15

2000
500
$65,000
$55,625
$9,375

,QSXWVDQGUDQJHQDPHV Enter all inputs and name the ranges as indicated. Note
that the Create from Selection shortcut was used to name all ranges except for CostLookup
and Probabilities. For these latter two, you can highlight the ranges and enter the names
in the Name Boxthe manual method. (Why the difference? To use the Create from
Selection shortcut, you must have appropriate labels in adjacent cells. Sometimes this is
simply not convenient.)
5HYHQXHV The company can sell only what it has, and it sells any leftovers at the discounted sale price. Therefore, enter the formulas
0,12UGHUBTXDQWLW\'HPDQG
,)2UGHUBTXDQWLW\!'HPDQG2UGHUBTXDQWLW\'HPDQG
and
8QLWVBVROGBDWBUHJXODUBSULFH 5HJXODUBSULFH
8QLWVBVROGBDWBOHIWRYHUBSULFH /HIWRYHUBSULFH
in cells B15, B16, and B17. The logic in the first two of these cells is necessary to account
correctly for the cases when the order quantity is greater than demand and when it is less
than or equal to demand. Note that you could use the following equivalent alternative to the
IF function in cell B16:
0$;2UGHUBTXDQWLW\'HPDQG
3VHIVMRK[MXL5YERXMX](MWGSYRXWERH(IQERH9RGIVXEMRX]

([FHO 7RRO I[ %XWWRQDQG)XQFWLRQ/LEUDU\*URXS

,I\RXZDQWWROHDUQPRUHDERXWKRZDQ([FHO IXQFWLRQRSHUDWHVFOLFNRQWKHI[ EXWWRQQH[W

WRWKH)RUPXODEDU7KLVLV FDOOHGWKH,QVHUW)XQFWLRQEXWWRQDOWKRXJKVRPHSHRSOHFDOOLW
WKH)XQFWLRQ:L]DUG,I WKHUHLVDOUHDG\DIXQFWLRQVXFKDVDQ,)IXQFWLRQLQDFHOODQG\RX
WKHQFOLFNRQWKHI[ EXWWRQ\RXZLOOJHWKHOSRQWKLVIXQFWLRQ,I\RXVHOHFWDQHPSW\FHOODQG
WKHQFOLFNRQWKHI [ EXWWRQ\RXFDQF KRRVHDIXQFWLRQWRJ HWKHOSRQ7KHVDPHKHOSLV
DYDLODEOHIURPWKH)XQFWLRQ/LEUDU\JURXSRQWKH)RUPXODVULEERQ

7RWDORUGHULQJFRVW Depending on the order quantity, you can find the appropriate
unit cost from the unit cost table and multiply it by the order quantity to obtain the total
ordering cost. This can be accomplished with a complex nested IF formula, but a much
better way is to use the VLOOKUP function. Specifically, enter the formula
9/22.832UGHUBTXDQWLW\&RVW/RRNXS 2UGHUBTXDQWLW\
in cell B18. The VLOOKUP part of this formula says to compare the order quantity to the
first (leftmost) column of the table in the CostLookup range and return the corresponding
value in the second column (because the last argument is 2).

([FHO )XQFWLRQ9/22.83
7KH9/22.83IXQFWLRQDFWVOLN HDWD[WDEOH ZKHUH\RXORRNXSWKHWD[FRUU HVSRQGLQJWR
\RXUDGMXVWHGJURVVLQFRPHIURPDWDEOHRILQFRPHVDQGWD[HV7RXVHLWILUVWFUHDWHDYHU
WLFDOORRNXSWDEOHZLWKYDOXHVWRXVHIRUFRPSDULVRQOLVWHGLQWKHOHIWFROXPQRIWKHWDEOH
DQGFRUU HVSRQGLQJRXWSXWYDOXHVLQDVPDQ\FROXPQVWRWKHULJKWDV\RXOLN
H6HHWKH
&RVW/RRNXSU DQJHLQ) LJXUHIRUDQH [DPSOH7KHQWKH9/22.83IXQFWLRQWDN HV
WKUHHRUIRXUDUJXPHQWVWKHYDOXH\RXZDQWWRFRPSDUHWRWKHYDOXHVLQWKHOHIWFROXPQ
WKHORRNXSWDEOHUDQJHWKHLQGH[RIWKHFROXPQ\RXZDQWWKHUHWXUQHGYDOXHWRFRPH
IURPZKHUHWKHLQGH[RIWKHOHIWFROXPQLVWKHLQGH [RIWKHQH[WFROXPQLVDQGVRRQ
DQGRSWLRQDOO\75 8(IRUDQDSSU R[LPDWHPDWFKWKHGHIDXOWRU) $/6(IRUDQH [DFW
PDWFK,I\RXRPLWWKHODVWDU JXPHQWWKHYDOXHVLQWKHOHIWFROXPQRIWKHWDEOHPXVWEH
HQWHUHGLQDVFHQGLQJRUGHU6HHRQOLQHKHOSIRUPRUHGHWDLOV,IWKHODVWDUJXPHQWLV758(
RULVRPLWWHG([FHOVFDQVGRZQWKHOHIWPRVWFROXPQRIWKHWDEOHDQGILQGVWKHODVWHQWU\OHVV
WKDQRUHTXDOWRWKHILUVWDUJXPHQW,QWKLVVHQVHLWILQGVDQDSSUR[LPDWHPDWFK7KHUHLV
DOVRDQ+/22.83IXQFWLRQWKDWZRUNVH[DFWO\WKHVDPHZD\H[FHSWWKDWWKHORRNXSWDEOH
LVDUUDQJHGLQURZVQRWFROXPQV
3URILW Calculate the profit with the formula
5HYHQXH&RVW

Two-Way Data Table

%X[S[E]HEXEXEFPI
EPPS[W]SYXSWIILS[
EWMRKPISYXTYXZEVMIW
EWX[SMRTYXWZEV]
WMQYPXERISYWP]

The next step is to create a two-way data table for profit as a function of the order
quantity and demand (see Figure 2.20). To create this table, first enter a link to the
profit with the formula 3URILW in cell A22, and enter possible order quantities and
possible demands in column A and row 22, respectively. (We used the same values for
both order quantity and demand, from 500 to 4500 in increments of 500. This is not
necessarythe demand could change in increments of 100 or even 1but it is reasonable. Perhaps Sams is required by the publisher to order in multiples of 500.) Then select
Data Table from the What-If Analysis dropdown list on the Data ribbon, and enter the
Demand cell as the Row Input cell and the Order_quantity cell as the Column Input
cell (see Figure 2.21).

*MKYVI Profit as a Function of Order Quantity and Demand

A
B
C
D
E
F
21 Data table of p
prot as a funcon of order quanty
q
y (along
(
g side)) and demand ((alongg top)
p)
22
$9,375
500
1000
1500
2000
2500
23
500
$3,000
$3,000
$3,000
$3,000
$3,000
24
1000
-$3,000
$7,000
$7,000
$7,000
$7,000
25
1500
-$9,500
$500 $10,500 $10,500 $10,500
26
2000
-$14,500
-$4,500
$5,500 $15,500 $15,500
27
2500
-$20,625 -$10,625
-$625
$9,375 $19,375
28
3000
-$25,250 -$15,250
-$5,250
$4,750 $14,750
29
3500
-$31,125 -$21,125 -$11,125
-$1,125
$8,875
30
4000
-$35,200
$35 200 -$25,200
$25 200 -$15,200
$15 200
-$5,200
$5 200
$4 800
$4,800
31
4500
-$40,850 -$30,850 -$20,850 -$10,850
-$850

G
3000
$3,000
$7,000
$10,500
$15,500
$19,375
$24,750
$18,875
$14 800
$14,800
$9,150

H
3500
$3,000
$7,000
$10,500
$15,500
$19,375
$24,750
$28,875
$24 800
$24,800
$19,150

I
4000
$3,000
$7,000
$10,500
$15,500
$19,375
$24,750
$28,875
$34 800
$34,800
$29,150

J
4500
$3,000
$7,000
$10,500
$15,500
$19,375
$24,750
$28,875
$34 800
$34,800
$39,150

*MKYVI
Dialog Box for
Two-Way Data Table

([FHO 7RRO 7ZR:D\'DWD7DEOH

$WZRZD\GDWDWDEOHDOORZV\RXWRVHHKRZDVLQJOH output FHOOYDULHVDV\RXYDU\WZR

input FHOOV8QOLNHDRQHZD\GDWDWDEOHRQO\Dsingle RXWSXWFHOOFDQEHFKRVHQ7RFUH
DWHWKLVW\SHRIWDEOHHQWHUDUHIHUHQFHWRWKHRXWSXWFHOOLQWKHWRSOHIWFRUQHURIWKHWDEOH
HQWHUSRVVLEOHYDOXHVRIWKHWZRLQSXWVEHORZDQGWRWKHULJKWRIWKLVFRUQHUFHOODQGKLJK
OLJKWWKHHQWLU HWDEOH7KHQVHOHFW'DWD7 DEOHIURPWKH:KDW,I$QDO\VLVGU RSGRZQRQWKH
'DWDULEERQDQGHQWHUU HIHUHQFHVWRWKHFHOOVZKHU HWKHRULJLQDOWZRLQSXWYDULDEOHVOLYH
7KH5RZ,QSXWFHOOFRUU HVSRQGVWRWKHYDOXHVDORQJWKHWRSU RZRIWKHWDEOH DQGWKH
&ROXPQ,QSXWFHOOFRUUHVSRQGVWRWKHYDOXHVDORQJWKHOHIWPRVWFROXPQRIWKHWDEOH:KHQ
\RXFOLF NRQ2. ([FHO VXEVWLWXWHVHDF KSDLURILQSXWYDOXHVLQWRWKHVHWZRLQSXWFHOOV
UHFDOFXODWHVWKHVSU HDGVKHHWDQGHQWHU VWKHFRUU HVSRQGLQJRXWSXWYDOXHLQWKHWDEOH %\
FOLFNLQJRQDQ\FHOOLQWKHERG\RIWKHWDEOH\RXFDQVHHWKDW([FHO DOVRHQWHUVWKHIXQFWLRQ
7$%/(DVDUHPLQGHUWKDWWKHFHOOLVSDUWRIDGDWDWDEOH

8LMWMWEGXYEPP]E
TVIZMI[SJHIGMWMSR
QEOMRKYRHIV
YRGIVXEMRX]8S
GEPGYPEXIERI\TIGXIH
TVSJMX ]SYQYPXMTP]
IEGLTVSJMXF]MXW
TVSFEFMPMX]ERHEHH
XLI TVSHYGXW8LMW
XSTMGMWGSZIVIHMR
HITXLMR'LETXIV

The resulting data table shows that profit depends heavily on both order quantity and
demand and (by scanning across rows) how higher demands lead to larger profits. But
which order quantity Sams should select is still unclear. Remember that Sams has complete control over the order quantity (it can choose the URZof the data table), but it has no
direct control over demand (it cannot choose the column).
The ordering decision depends not only on which demands are SRVVLEOH but on
which demands are OLNHO\ to occur. The usual way to express this information is with a
set of probabilities that sum to 1. Suppose Sams estimates these as the values in row
35 of Figure 2.22. These estimates are probably based on other similar books it has
sold in the past. The most likely demands are 2000 and 2500, with other values on both
sides less likely. You can use these probabilities to find an H[SHFWHG profit for each order
quantity. This expected profit is a weighted average of the profits in any row in the
data table, using the probabilities as the weights. The easiest way to do this is to enter
the formula
680352'8&7%-3UREDELOLWLHV

3VHIVMRK[MXL5YERXMX](MWGSYRXWERH(IQERH9RGIVXEMRX]

*MKYVI Comparison of Expected Profits

C
1000
1000
0.05

1500
1500
0.15

2000
2000
0.25

2500
2500
0.25

3000
3000
0.15

3500
3500
4000
4000
0.07
0.04
Sum of probabili\es -->

K
4500
4500
0.015
1

Expected Prot versus Order Quanty

Order 2000 to
maximize the
expected prot.
Expected Prot

A
B
33 Model of expected demands
34 Demand
500
500
35 Probability
0.025
36
37
Order quan\ty Expected prot
38
500
$3,000
39
1000
$6,750
40
1500
$9,500
41
2000
$12,250
42
2500
$11,375
43
3000
$9,500
3500
$4,875
44
45
4000
$1,350
46
4500
-$4,150
47
48
49
50
51
52

$14,000
$12,000
$10,000
$8,000
$6,000
$4,000
$2,000
$0
-$2,000
-$4,000
-$6,000

500

1000

1500

2000

2500

3000

3500

4000

4500

Order Quanity

in cell B38 and copy it down to cell B46. You can also create a bar chart of these expected
profits, as shown in Figure 2.22. (Excel refers to these as FROXPQFKDUWV. The height of each
bar is the expected profit for that particular order quantity.)

([FHO )XQFWLRQ 680352'8&7

7KH680352'8&7IXQFWLRQWDNHVWZRUDQJHDUJXPHQWVZKLFKPXVWEHexactly the same

size and shape, DQGLWVXPVWKHSU RGXFWVRIWKHFRUUHVSRQGLQJYDOXHVLQWKHVHWZRU DQJHV
)RUH[DPSOHWKHIRUPXOD=SUMPRODUCT(A10:B11,E12:F13) LVDVKRUWFXWIRUDIRUPXOD
LQYROYLQJWKHVXPRISU RGXFWV =A10*E12+A11*E13+B10*F12+B11*F13. 7KLVLVDQ
H[WUHPHO\XVHIXOIXQFWLRQHVSHFLDOO\ZKHQWKHU DQJHVLQ YROYHGDU HODU JHDQGLWLVXVHG
UHSHDWHGO\WKURXJKRXWWKLVERRN$FWXDOO\ WKH68035 2'8&7IXQFWLRQFDQKDYHPRU H
WKDQWZRU DQJHDUJXPHQWVDOORIWKHVDPHVL]HDQGVKDSH E XWWKHPRVWFRPPRQXVHRI
680352'8&7LVZKHQRQO\WZRUDQJHVDUHLQYROYHG

The largest of the expected profits, $12,250, corresponds to an order quantity of 2000,
so we would recommend that Sams order 2000 copies of the book. This does not guarantee that Sams will make a profit of $12,250the actual profit depends on the eventual
demandbut it represents a reasonable way to proceed in the face of uncertain demand.
You will learn much more about making decisions under uncertainty and the expected
value criterion in Chapter 9. Q

PROBLEMS
Skill-Building Problems

In some ordering problems, like the one for Sams

Bookstore, whenever demand exceeds existing inventory, the excess demand is not lost but is filled by
expedited ordersat a premium cost to the company.
Change Sams model to reflect this behavior. Assume
that the unit cost of expediting is $40, well above the
highest regular unit cost.

The spreadsheet model for Sams Bookstore contains a

two-way data table for profit versus order quantity and
demand. Experiment with Excels chart types to create
a chart that shows this information graphically in an
intuitive format. (Choose the format you would choose
to give a presentation to your boss.)

there is a list of possible demands, those currently in

row 34. Then insert a new row below row 11 that lists
the probabilities of these demands. Next, in the rows
below the Profit Model label, calculate the units
sold, revenue, cost, and profit for HDFKdemand. For
example, the quantities in column C will be for the
second possible demand. Finally, use SUMPRODUCT
to calculate H[SHFWHGprofit below the Profit row.

In the Sams Bookstore problem, the quantity discount

structure is such that DOO the units ordered have the
same unit cost. For example, if the order quantity is
2500, then each unit costs $22.25. Sometimes the
quantity discount structure is such that the unit cost
for the first so many items is one value, the unit cost
for the next so many units is a slightly lower value,
and so on. Modify the model so that Sams pays
$24 for units 1 to 1500, $23 for units 1501 to 2500,
and $22 for units 2501 and above. For example,
the total cost for an order quantity of 2750 is
1500(24) 1 1000(23) 1 250(22). (+LQW: Use IF
functions, not VLOOKUP.)

Skill-Extending Problems

The current spreadsheet model essentially finds the

expected profit in several steps. It first finds the profit
in cell B19 for a IL[HGvalue of demand. Then it uses
a data table to find the profit for each of several
demands, and finally it uses SUMPRODUCT to find
the expected profit. Modify the model so that expected
profit is found directly, without a data table. To do
this, change row 11 so that instead of a single demand,

Continuing Problem 5, create a two-way data table for

expected profit with order quantity along the side and
unit expediting cost along the top. Allow the order
quantity to vary from 500 to 4500 in increments of
500, and allow the unit expediting cost to vary from
$36 to $45 in increments of $1. Each column of this
table will allow you to choose an optimal order quantity for a given unit expediting cost. How does this
best order quantity change as the unit expediting cost
increases? Write up your results in a concise memo to
management. (+LQW: You will have to modify the existing spreadsheet model so that there is a cell for
expected profit that changes automatically when you
change either the order quantity or the unit expediting
cost. See Problem 8 for guidelines.)

2.6 ESTIMATING THE RELATIONSHIP BETWEEN

PRICE AND DEMAND
The following example illustrates a very important modeling concept: estimating relationships between variables by FXUYHILWWLQJ You will study this topic in much more depth in
the discussion of regression in Chapter 14, but the ideas can be illustrated at a relatively
low level by taking advantage of some of Excels useful features.

)<%140)

2.5 E STIMATING S ENSITIVITY

AT THE L INKS C OMPANY

D EMAND

P RICE

he Links Company sells its golf clubs at golf outlet stores throughout the United States.
The company knows that demand for its clubs varies considerably with price. In fact,
the price has varied over the past 12 months, and the demand at each price level has been
observed. The data are in the data sheet of the file *ROI&OXE'HPDQG[OV[ (see Figure
2.23.) For example, during the past month, when the price was $390, 6800 sets of clubs
were sold. (The demands in column C are in hundreds of units. The cell comment in cell C3
reminds you of this.) The company wants to estimate the relationship between demand and
price and then use this estimated relationship to answer the following questions:
Assuming the unit cost of producing a set of clubs is $250 and the price must be a
multiple of $10, what price should Links charge to maximize its profit?
How does the optimal price depend on the unit cost of producing a set of clubs?
Is the model an accurate representation of reality?

)WXMQEXMRKXLI6IPEXMSRWLMT&IX[IIR4VMGIERH(IQERH

*MKYVI
Demand and Price
Data for Golf Clubs

A
B
1 Demand for golf clubs
2
Month
Price
3
1
450
4
2
300
5
3
440
6
4
360
7
5
290
8
6
450
9
7
340
10
8
370
11
9
500
12
10
490
13
11
430
14
12
390
15

Demand
45
103
49
86
125
52
87
68
45
44
58
68

%XVLQHVV2EMHFWLYHV To estimate the relationship between demand and price, and to use
this relationship to find the optimal price to charge.
([FHO2EMHFWL YHV To illustrate Excels Trendline tool, and to illustrate conditional
formatting.

7SPYXMSR
This example is divided into two parts: estimating the relationship between price and
demand, and creating the profit model.

Estimating the Relationship Between Price and Demand

A scatterplot of demand versus price appears in Figure 2.24. (This can be created in the
usual way with Excels Scatter chart.) Obviously, demand decreases as price increases, but
the goal is to quantify this relationship. Therefore, after creating this chart, right-click on
any point on the chart to bring up the dialog box in Figure 2.25. This allows you to superimpose several different curves (including a straight line) on the scatterplot. We consider

*MKYVI
Scatterplot of
Demand Versus
Price

130
120
110
100
90
80
70
60
50
40
280

320

360

400

440

480

520

three possibilities, the OLQHDU SRZHU and H[SRQHQWLDO curves, defined by the following
general equations (where \ and [, a general output and a general input, correspond to
demand and price for this example):
Q
Q
Q

Linear: \5 D1 E[

Power: \5 D[E
Exponential: \5 DHE[

Before proceeding, we describe some general properties of these three functions because
of their widespread applicability. The linear function is the easiest. Its graph is a straight
line. When [changes by 1 unit, \changes by Eunits. The constant Dis called the intercept,
and Eis called the slope.
The power function is a curve except in the special case where the exponent Eis 1.
(Then it is a straight line.) Assuming that D is positive, the shape of this curve depends primarily on the exponent E. If E. 1, \increases at an increasing rate as [increases. If 0 , E
, 1, \increases, but at a decreasing rate, as [increases. Finally, if E, 0, \decreases as [
increases. An important property of the power curve is that when [ changes by 1%, \
changes by a constant percentage, and this percentage is approximately equal to E%. For
example, if \ 5 100[22.35, then every 1% increase in [ leads to an approximate 2.35%
decrease in \.
The exponential function also represents a curve whose shape depends on the constant
E in the exponent. Again, assume that D is positive. Then if E . 0, \ increases as [
increases; if E, 0, \decreases as [increases. An important property of the exponential
function is that if [changes by 1 unit, \changes by a constant percentage, and this percentage is approximately equal to 100 3 E%. For example, if \5 100H20.014[, then whenever [increases by 1 unit, \decreases by approximately 1.4%. Here His the special number
2.7182 . . . , and H to any power can be calculated in Excel with the EXP function. For
example, you can calculate H20014 with the formula (;3.

*MKYVI
Trendline Options
Dialog Box

)WXMQEXMRKXLI6IPEXMSRWLMT&IX[IIR4VMGIERH(IQERH

Returning to the example, if you superimpose any of these curves on the scatterplot of
demand versus price, Excel chooses the best-fitting curve of that type. Better yet, if you
check the Display Equation on Chart option, you see the equation of this best-fitting curve.
Doing this for each type of curve gives the results in Figures 2.26, 2.27, and 2.28. (The
equations might not appear exactly as in the figures. However, they can be resized and
reformatted to appear as shown.)

*MKYVI
Best-Fitting
Straight Line

Linear Fit
130
120

y = -0.3546x + 211.3147

110
100
90
80
70
60
50
40
280

*MKYVI
Best-Fitting
Power Curve

320

360

400

440

480

520

480

520

Power Fit
130
120

y = 5871064.2031x-1.9082

110
100
90
80
70
60
50
40
280

320

360

400

440

*MKYVI Best-Fitting Exponential Curve

Exponenal Fit
130
120
110

y = 466.5101e-0.0049x

100
90
80
70
60
50
40
280

320

360

400

440

480

520

Each of these curves provides the best-fitting member of its family to the demand/price
data, but which of these three is best overall? You can answer this question by finding the
PHDQDEVROXWHSHUFHQWDJHHUURU(MAPE) for each of the three curves. To do so, for any
price in the data set and any of the three curves, first predict demand by substituting the
given price into the equation for the curve. The predicted demand is typically not the same
as the observed demand, so you can calculate the absolute percentage error (APE) with the
general formula:
) Observed demand 2 Predicted demand)
APE 5 }}}}}
Observed demand

Then for any curve, MAPE is the average of these APE values. The curve with the smallest MAPE is the best fit overall.
The calculations appear in Figure 2.29. After (manually) entering the parameters of
the equations from the scatterplots into column B, you can proceed as follows.
3UHGLFWHGGHPDQGV Substitute observed prices into the linear, power, and exponential functions to obtain the predicted demands in columns E, F, and G. Specifically, enter
the formulas
%% %
% %A%
and
% (;3% %
in cells E19, F19, and G19, and copy them down their respective columns.

)WXMQEXMRKXLI6IPEXMSRWLMT&IX[IIR4VMGIERH(IQERH

*MKYVI Finding the Best-Fitting Curve Overall

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

A
B
C
Parameters of best-4ng curves
Linear
Intercept
211.31
Slope
-0.3546
Power
Constant
5871064
-1.9082
Exponent
Exponential
Constant
466.51
-0.00491
Exponent

E
Linear
51.74
104.93
55.29
83.65
108.48
51.74
90.75
80.11
34.01
37.56
58.83
73.02

F
Prediction
Power
50.80
110.12
53.02
77.76
117.48
50.80
86.73
73.80
41.55
43.18
55.40
66.75

G
Exponential
51.20
106.94
53.78
79.65
112.32
51.20
87.87
75.84
40.06
42.07
56.49
68.74
MAPE

I
J
Absolute percentage error
Linear
Power Exponential
14.98%
12.89%
13.78%
1.87%
6.91%
3.83%
12.83%
8.21%
9.75%
2.73%
9.58%
7.38%
13.22%
6.01%
10.14%
0.50%
2.31%
1.53%
4.31%
0.32%
1.00%
17.81%
8.53%
11.52%
24.42%
7.67%
10.99%
14.65%
1.86%
4.38%
1.43%
4.48%
2.61%
7.38%
1.84%
1.09%
9.68%

5.88%

6.50%

$YHUDJHSHUFHQWDJHHUURUV Apply Equation (2.1) to calculate APEs in columns H, I,

and J. Specifically, enter the general formula

$%6&(y&

in cell H19 and copy it to the range H19:J30. (Do you see why column C is made absolute?
Remember that this is where the observed demands are stored.)
0$3( Average the APEs in each column with the AVERAGE function to obtain the
MAPEs in row 32.
Evidently, the power curve provides the best fit, with a MAPE of 5.88%. In other words, its
predictions are off, on average, by 5.88%. This power curve predicts that each 1% increase
in price leads to an approximate 1.9% decrease in demand. (Economists call this relationship elasticdemand is quite sensitive to price.)

DEVELOPING THE PROFIT MODEL

Now we move to the profit model, using the best-fitting power curve to predict demand
from price. The key variables appear in Table 2.5. Note there is now one input variable,
unit variable cost, and one decision variable, unit price. (The red background for the decision variable distinguishes it as such.) The profit model is straightforward to develop using
the following steps (see Figure 2.30).

8EFPI Key Variables for Golf Club Problem

,QSXWYDULDEOH
'HFLVLRQYDULDEOH
.H\RXWSXWYDULDEOH
2WKHURXWSXWYDULDEOHV

Unit cost to produce

Unit price
Profit
Predicted demand, total revenue, total cost

*MKYVI
Profit Model

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

A
B
C
D
Prot model
model, using best 4ng power curve for esmang demand

Parameters of best-4ng power curve (from Esmaon sheet)

Constant
5871064
Exponent
-1.9082
p
Monetaryy inputs
Unit cost to produce

$250

Decision variable
Unit price (trial value)

$400

Prot model
P di t d d
Predicted
demand
d
Total revenue
Total cost
Profit

63 601
63.601
$25,441
$15,900
$9,540

3UHGLFWHGGHPDQG Calculate the SUHGLFWHGdemand in cell B14 with the formula

% %A%
This uses the power function that was estimated earlier.
5HYHQXHFRVWSURILW Enter the following formulas in cells B15, B16, and B17:
% %
% %
and
%%
The assumption here is that the company produces exactly enough sets of clubs to meet
customer demand.
0D[LPL]LQJ 3URILW To see which price maximizes profit, you can build the data table
shown in Figure 2.31. Here, the column input cell is B11 and the linking formula in
cell B25 is %. The corresponding scatter chart shows that profit first increases and
then decreases. You can find the maximum profit and corresponding price in at least
three ways. First, you can attempt to read them from the chart. Second, you can scan
down the data table for the maximum profit, which is shown in the figure. The
following Excel Tip describes a third method that uses some of Excels more powerful
features.

([FHO 7LS &RQGLWLRQDO)RUPDWWLQJ

&HOO%LQ)LJXUHLVFRORUHGEHFDXVHLWFRUUHVSRQGVWRWKHPD[LPXPSURILWLQWKHFRO
XPQEXW([FHOV &RQGLWLRQDO)RUPDWWLQJWRROFDQGRWKLVIRU\RXDXWRPDWLFDOO\7 7RFRORU
WKHPD[LPXPSU RILWVHOHFWWKHU DQJHRISU RILWV%%FOLF NRQWKH&RQGLWLRQDO
)RUPDWWLQJGURSGRZQDUURZWKHQ7RS%RWWRP5XOHVDQGWKHQ7RS,WHPVWREULQJXSWKH
GLDORJER[LQ)LJXUH%\DVNLQJIRUWKHWRSLWHPWKHPD[LPXPYDOXHLQWKHU DQJHLV
FRORUHG<RXFDQH[SHULPHQWZLWKWKHPDQ\RWKHU&RQGLWLRQDO)RUPDWWLQJRSWLRQV7KLVLVD
JUHDWWRRO

7The

value in cell B52 also appears to be the maximum, but to two decimals, it is slightly lower.

)WXMQEXMRKXLI6IPEXMSRWLMT&IX[IIR4VMGIERH(IQERH

19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

A
B
C
D
E
Maximum prot
p
from data table below,, with corresponding
p
g best unit p
price
Maximum prot
$10,409
Best price
$530
Data table for Prot as a funcon of Unit price
Unit price
Prot
$9,540
$260
$1,447
$270
$2,693
$280
$3 769
$3,769
$290
$4,699
$300
$5,506
$310
$6,207
$320
$6,815
$330
$7,345
$340
$7,805
$350
$8,206
$360
$8,554
$370
$8
856
$8,856
$380
$9,118
$390
$9,345
$400
$9,540
$410
$9,708
$420
$9,851
$430
$9,973
$440 $10,075
$
$ ,
$450 $10,160
$460 $10,230
$470 $10,286
$480 $10,330
$490 $10,363
$500 $10,387
$510 $10,402
$520 $10,409
$530 $10,409
$540 $10,403
$10 403
$550 $10,391
$560 $10,375
$570 $10,354
$580 $10,329
$590 $10,300
$600 $10,269

Prot versus Price

$12 000
$12,000
$10,000
$8,000
Prot

*MKYVI
Profit as a
Function of Price

$6,000
$4,000
$2,000
$0

Price

Maximum prot

*MKYVI
Conditional
Formatting
Dialog Box

What about the corresponding best price, shown in cell B21 of Figure 2.31? You could
enter this manually, but wouldnt it be nice if you could get Excel to find the maximum
profit in the data table, determine the price in the cell to its left, and report it in cell B21, all
automatically? This is indeed possible. Just enter the formula
,1'(;$$0$7&+%%%
in cell B21, and the best price appears. This formula uses two Excel functions, MATCH
and INDEX. MATCH compares the first argument (the maximum profit in cell B20) to
the range specified in the second argument (the range of profits), and returns the index of
the cell where a match appears. (The third argument, 0, specifies that you want an H[DFW
match.) In this case, the MATCH function returns 28 because the maximum profit is in
the 28th cell of the profits range. Then the INDEX function is called effectively as
=INDEX(A26:A75,28,1). The first argument is the range of prices, the second is a row
index, and the third is a column index. Very simply, this function says to return the value in
the 28th row and first column of the prices range.
To learn more about these functions, you can click on the I[ button and examine the
functions in the Lookup & Reference category. After experimenting, you can see that the

INDEX and MATCH combination solves the problem. You dont have to memorize these
functions, although this combination really does come in handy. Rather, you can often
solve a problem by investigating some of Excels less well-known features. You dont even
need a manualeverything is in online help.

Sensitivity to Variable Cost

We now return to question 2 in the example: How does the best price change as the unit variable cost changes? You can answer this question with a two-way data table. Remember that
this is a data table with two inputsone along the left side and the other across the top row
and a single output. The two inputs for this problem are unit variable cost and unit price, and
the single output is profit. The corresponding data table is in the range A83:F168, the top part
of which appears in Figure 2.33. To develop this table, enter desired inputs in column A and
row 83, enter the linking formula % in cell A83 (it always goes in the top-left corner of a
two-way data table), highlight the entire table, select Data Table from the What-If Analysis
dropdown list, and enter B8 as the Row Input cell and B11 as the Column Input cell.
*MKYVI Profit as a Function of Unit Cost and Unit Price
G

Maximum Prot versus Unit Cost

$20,000
Maximum Prot

A
B
C
D
E
F
77 Maximum prot for dierent unit costs (from data table below)
78 Unit cost
$150
$200
$250
$300
$350
79 Maximum prot
$16,552 $12,748 $10,409
$8,821
$7,669
80 Corresponding best price
$320
$420
$530
$630
$740
81
82 Sensivity of prot to unit cost (unit cost is along the top, unit price is along the side)
$9,540
$150
$200
$250
$300
$350
83
$160
$3,654 -$14,618 -$32,890
-$51,161
-$69,433
84
$170
$6,510
-$9,766 -$26,041
-$42,317
-$58,593
85
$180
$8,756
-$5,838 -$20,432
-$35,026
-$49,620
86
$190 $10,531
-$2,633 -$15,796
-$28,960
-$42,123
87
$200 $11,936
$0 -$11,936
-$23,872
-$35,808
88
$210 $13,050
$2,175
-$8,700
-$19,575
-$30,450
89
90
$220 $13,932
$3,980
-$5,971
-$15,922
-$25,873
91
$230 $14,627
$5,485
-$3,657
-$12,799
-$21,941
92
$240 $15,172
$6,743
-$1,686
-$10,115
-$18,543
93
$250 $15,594
$7,797
$0
-$7,797
-$15,594
94
$260 $15,917
$8,682
$1,447
-$5,788
-$13,023
95
$270 $16,157
$9,425
$2,693
-$4,039
-$10,772
96
$280 $16,330 $10,049
$3,769
-$2,512
-$8,793
97
$290 $16,447 $10,573
$4,699
-$1,175
-$7,049
98
$300 $16,518 $11,012
$5,506
$0
-$5,506
99
$310 $16,551 $11,379
$6,207
$1,034
-$4,138
$6,815
$1,947
-$2,921
100
$320 $16,552 $11,683
101
$330 $16,526
$16 526 $11,935
$11 935
$7 345
$7,345
$2 754
$2,754
-$1,836
$1 836
102
$340 $16,478 $12,142
$7,805
$3,469
-$867
103
$350 $16,412 $12,309
$8,206
$4,103
$0
104
$360 $16,331 $12,442
$8,554
$4,666
$778
105
$370 $16,237 $12,547
$8,856
$5,166
$1,476
106
$380 $16,132 $12,625
$9,118
$5,611
$2,104
107
$390 $16,020 $12,682
$9,345
$6,007
$2,670
108
$400 $15,900 $12,720
$9,540
$6,360
$3,180
109
$410 $15,775 $12,742
$9,708
$6,674
$3,640
$9,851
$6,954
$4,056
110
$420 $15,646 $12,748

$15,000
$10,000
$5,000
$0
$100

$150

$200

$250

$300

$350

Unit Cost

As before, you can scan the columns of the data table for the maximum profits
and enter them (manually) in rows 79 and 80. (Alternatively, you can use the Excel features described in the previous Excel Tip to accomplish these tasks. Take a look at the
finished version of the file for details. This file also explains how conditional formatting is used to color the maximum profit in HDFK column of the table.) Then you can create a chart of maximum profit (or best price) versus unit cost. The chart in Figure 2.33
shows that the maximum profit decreases, but at a decreasing rate as the unit cost
increases.

Limitations of the Model

Question 3 asks you to step back from all these details and evaluate whether the model is
realistic. First, there is no real reason to restrict golf club prices to multiples of $10. This

)WXMQEXMRKXLI6IPEXMSRWLMT&IX[IIR4VMGIERH(IQERH

was only required so that a data table could be used to find the profit-maximizing price.
Ideally, you should search over DOO possible prices to find the profit-maximizing price.
Fortunately, Excels built-in Solver tool enables you to accomplish this task fairly easily.
The problem of finding a profit-maximizing price is an example of an RSWLPL]DWLRQ
PRGHO In optimization models, you try to maximize or minimize a specified output cell by
changing the values of the decision variable cells. Chapters 38 and 16 contain a detailed
discussion of optimization models.
A second possible limitation of the model is the implicit assumption that price is the
RQO\factor that influences demand. In reality, other factors, such as advertising, the state of
the economy, competitors prices, strength of competition, and promotional expenses, also
influence demand. In Chapter 14, you will learn how to use multiple regression to analyze
the dependence of one variable on two or more other variables. This technique allows you
to incorporate other factors into the model for profit.
A final limitation of the model is that demand might not HTXDOsales. For example, if
actual demand for golf clubs during a year is 70,000 but the companys annual capacity is
only 50,000, the company will observe sales of only 50,000. This will cause it to underestimate DFWXDOdemand, and the curve-fitting method will produce biased predictions. (Can
you guess the probable effect on pricing decisions?)

Other Modeling Issues

The layout of the *ROI&OXE'HPDQG[OV[ file is fairly straightforward. However, note that
instead of a single worksheet, there are two worksheets, partly for logical purposes and
partly to reduce clutter. There is one worksheet for estimation of the demand function and
the various scatterplots, and there is another for the profit model.
One last issue is the placement of the data tables for the sensitivity analysis. You might
be inclined to put these on a separate Sensitivity worksheet. However, Excel does not allow
you to build a data table on one worksheet that uses a row or column input cell from
DQRWKHUworksheet. Therefore, you are forced to put the data tables on the same worksheet
as the profit model. Q

PROBLEMS
Skill-Building Problems
Suppose you have an extra six months of data on
demands and prices, in addition to the data in the
example. These extra data points are (350,84), (385,72),
(410,67), (400,62), (330,92), and (480,53). (The price is
shown first and then the demand at that price.) After
adding these points to the original data, use Excels
Trendline tool to find the best-fitting linear, power, and
exponential trend lines. Finally, calculate the MAPE for
each of these, based on all 18 months of data. Does the
power curve still have the smallest MAPE?
Consider the power curve \ 5 10000[22.35. Calculate
\ when [ 5 5; when [ 5 10; and when [ 5 20. For
each of these values of [, find the percentage change
in \ when [ increases by 1%. That is, find the percent-

age change in \when [increases from 5 to 5.05; when

it increases from 10 to 10.1; and when it increases
from 20 to 20.2. Is this percentage change constant?
What number is it very close to? Write a brief memo
on what you have learned about power curves from
these calculations.
Consider the exponential curve \ 5 1000H20.014[.
Calculate \ when [ 5 5; when [ 5 10; and when
[ 5 20. For each of these values of [, find the
percentage change in \ when [ increases by one unit.
That is, find the percentage change in \ when [
increases from 5 to 6; when it increases from 10 to 11;
and when it increases from 20 to 21. Is this percentage
change constant? When expressed as a decimal, what
number is it very close to? Write a brief memo on

what you have learned about exponential curves from

these calculations.

Skill-Extending Problem
In the profit model in this section, we used the power
curve to relate demand and price because it has the

lowest MAPE. However, the exponential curve was

not far behind. Rework the profit model using the
exponential curve to relate demand to price. Write a
brief memo indicating whether you get basically the
same results as with the power curve or you get substantially different results.

2.7 DECISIONS INVOLVING THE TIME VALUE OF MONEY

In many business situations, cash flows are received at different points in time, and a company must determine a course of action that maximizes the value of cash flows. Here are
some examples:
Q
Q
Q

Should a company buy a more expensive machine that lasts for 10 years or a less
expensive machine that lasts for 5 years?
What level of plant capacity is best for the next 20 years?
A company must market one of several midsize cars. Which car should it market?

To make decisions when cash flows are received at different points in time, the key
concept is that the later a dollar is received, the less valuable the dollar is. For example,
suppose you can invest money at a 5% annual interest rate. Then $1.00 received now is
essentially equivalent to $1.05 a year from now. The reason is that if you have $1.00 now,
you can invest it and gain $0.05 in interest in one year. If U 5 0.05 is the interest rate
(expressed as a decimal), we can write this as
$1.00 now 5 $1.05 a year from now 5 $1.00(1 1 U)

Dividing both sides of Equation (2.2) by 1 1 U, we can rewrite it as

$1.00 3 1y(1 1 U) now 5 $1.00 a year from now

The value 1y(1 1 Uin Equation (2.3) is called the GLVFRXQWIDFWRU and it is always less
than 1. The quantity on the left, which evaluates to $0.952 for U5 0.05, is called the SUH
VHQWYDOXHof $1.00 received a year from now. The idea is that if you had $0.952 now, you
could invest it at 5% and have it grow to $1.00 in a year.
In general, if money can be invested at annual rate Ucompounded each year, then $1
received Wyears from now has the same value as 1y(1 1 U)W dollars received todaythat is,
the $1 is discounted by the discount factor raised to the Wpower. If you multiply a cash flow
received Wyears from now by 1y(1 1 U)W to obtain its present value, then the total of these
present values over all years is called the QHWSUHVHQWYDOXH139of the cash flows. Basic
financial theory states that projects with positive NPVs increase the value of the company,
whereas projects with negative NPVs decrease the value of the company.
The rate U (usually called the GLVFRXQWUDWH ) used by major corporations generally
comes from some version of the FDSLWDODVVHWSULFLQJPRGHO The value of U used to evaluate any particular project depends on a number of things and can vary from project to project. Because this is the focus of finance courses, we will not pursue it here. But given a
suitable value of U, the following example illustrates how spreadsheet models and the time
value of money can be used to make complex business decisions.
The GLVFRXQWIDFWRUis 1 divided by (1 plus the GLVFRXQWUDWH). To discount a cash
flow that occurs Wyears from now, multiply it by the discount factor raised to the
W power. The 139is the sum of all discounted cash flows.

(IGMWMSRW-RZSPZMRKXLI8MQI :EPYISJ1SRI]

) 8 1 '$ 0 ( 1 7$ / , 1 6 , * + 7
8LI 8MQI :EPYI SJ 1SRI]
1SRI]IEVRIHMRXLIJYXYVIMWPIWWZEPYEFPIXLERQSRI]
IEVRIHXSHE] JSVXLIWMQTPIVIEWSRXLEXQSRI]IEVRIH
XSHE]GERFIMR ZIWXIHXSIEVRMRXIV IWX 7MQMPEVP] GSWXW
MRGYVVIHMRXLIJYXYV IEV IPIWW GSWXP]XLERGSWXW
MRGYVVIHXSHE] [LMGLMW[L]]SYHSRXWMQTP]WYQYT

)<%140)

2.6 C ALCULATING NPV

VIZIRYIWERHGSWXWMREQYPXMTIVMSHQSHIP=SY MRWXIEH
HMWGSYRX JYXYVIVIZIRYIWERHGSWXWJ SVEJEMVGSQTEVM
WSR[MXLV IZIRYIWERHGSWXWMRGYV VIHXSHE ] 8LI
VIWYPXMRKWYQSJHMWGSYRXIHGEWLJPS [WMWXLIRIXTV I
WIRXZEPYI24: ERHMXJ SVQWXLIGSVRIVWXSRISJ
QYGLSJJMRERGMEPXLISV]ERHETTPMGEXMSRW

A CRON

cron is a large drug company. At the current time, the beginning of year 0, Acron is
trying to decide whether one of its new drugs, Niagra, is worth pursuing. Niagra is in
the final stages of development and will be ready to enter the market one year from now.
The final cost of development, to be incurred at the beginning of year 1, is $9.3 million.
Acron estimates that the demand for Niagra will gradually grow and then decline over its
useful lifetime of 20 years. Specifically, the company expects its gross margin (revenue
minus cost) to be $1.2 million in year 1, then to increase at an annual rate of 10% through
year 8, and finally to decrease at an annual rate of 5% through year 20. Acron wants to
develop a spreadsheet model of its 20-year cash flows, assuming its cash flows, other than
the initial development cost, are incurred at the HQGV of the respective years.8 Using an
annual discount rate of 12% for the purpose of calculating NPV, the drug company wants
to answer the following questions:
Is the drug worth pursuing, or should Acron abandon it now and not incur the $9.3
million development cost?
How do changes in the model inputs change the answer to question 1?
How realistic is the model?
%XVLQHVV2EMHFWLYHV To develop a model that calculates the NPV of Acrons cash flows,
to use this model to determine whether the drug should be developed further and then marketed, and to see how sensitive the answer to this question is to model parameters.
([FHO2EMHFWL YHV To illustrate efficient selection and copying of large ranges and to
learn Excels NPV function.

7SPYXMSR
The key variables in Acrons problem appear in Table 2.6. The first two rows contain the
inputs stated in the problem. We have made a judgment call as to which of these are known
with some certainty and which are uncertain. Although we wont do so in this chapter, a
thorough study of Acrons problem would treat this uncertainty explicitly, probably with
simulation. For now, you can accept the values given in the statement of the problem and
leave the simulation for a later chapter.
8To

simplify the model, taxes are ignored.

8EFPI Key Variables for Acrons Problem

,QSXWYDULDEOHV

Development cost, first year gross margin, rate of increase

during early years, years of growth, rate of decrease in later
years, discount rate
NPV
Yearly gross margins

.H\RXWSXWYDULDEOH
2WKHUFDOFXODWHGYDULDEOHV

The model of Acrons cash flows appears in Figure 2.34. As with many financial
spreadsheet models that extend over a multiyear period, you enter typical formulas in the
first year or two and then copy this logic down to all years. (In a previous edition, we made
the years go across, not down. In that case, splitting the screen is useful so that you can see
the first and last years of data. Splitting the screen is explained in the following Excel Tip.
The main reason we modified the model to have the years go down, not across, is that it
now fits easily on a screen, without needing to split the screen.)
*MKYVI
Acrons Model of
20-Year NPV

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

A
Calculang NPV at Acron
Inputs
Development cost
Gross margin year 1
Rate of increase
Increase through year
Rate of decrease
Discount rate
Cash ows
End of year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
NPV

9.3
1.2
10%
8
5%
12%

D
Range names used:
Development_cost
Discount_rate
Gross_margin_year_1
Gross_margin
Increase_through_year
Rate_of_decrease
Rate_of_increase

=Model!$B$4
=Model!$B$9
=Model!$B$5
=Model!$B$13:$B$32
=Model!$B$7
=Model!$B$8
=Model!$B$6

Gross margin
1.2000
1.3200
1.4520
1.5972
1.7569
1.9326
2.1259
2.3385
2.2215
2.1105
2.0049
1.9047
1.8095
1.7190
1.6330
1.5514
1.4738
1.4001
1.3301
1.2636
3.3003

([FHO 7LS 6SOLWWLQJWKH6FUHHQ

7RVSOLWWKHVFUHHQKRUL]RQWDOO\GUDJWKHVHSDUDWRUMXVWWRWKHULJKWRIWKHERWWRPVFUROOEDU
WRWKHOHIW7RVSOLWWKHVFU HHQYHUWLFDOO\GUDJWKHVHSDU DWRUMXVWDER YHWKHULJKWVFU ROOEDU
GRZQZDUG'UDJHLWKHUVHSDUDWRUEDFNWRLWVRULJLQDOSRVLWLRQWRUHPRYHWKHVSOLW

DEVELOPING THE SPREADSHEET MODEL

To create the model, complete the following steps. (See the file &DOFXODWLQJ139[OV[.)
,QSXWVDQGUDQJHQDPHV Enter the given input data in the blue cells, and name the
ranges as shown. As usual, note that the range names for cells B4 through B9 can be created all at once with the Create from Selection shortcut, as can the range name for the gross
margins in column B. In the latter case, highlight the whole range B12:B32 and then use
the Create from Selection shortcut.
(IGMWMSRW-RZSPZMRKXLI8MQI :EPYISJ1SRI]

&DVKIORZV Start by entering the formula

*URVVBPDUJLQB\HDUB
in cell B13 for the year 1 gross margin. Then enter the general formula
,)$ ,QFUHDVHBWKURXJKB\HDU% 5DWHBRIBLQFUHDVH
% 5DWHBRIBGHFUHDVH
in cell B14 and copy it down to cell B32 to calculate the other yearly gross margins. Note
how this IF function checks the year index in column A to see whether sales are still
increasing or have started to decrease. Of course, by using the (range-named) input cells in
this formula, you can change any of these inputs in cells B6 through B8, and the calculated
cells will automatically update. This is a PXFKbetter practice than embedding the numbers
in the formula itself.

([FHO 7LS (IILFLHQW6HOHFWLRQ

$QHDV\ZD\WRVHOHFWDODUJHUDQJHDVVXPLQJWKDWWKHILUVWDQGODVWFHOOVRIWKHUDQJHDUHYLVLEOH
LVWRVHOHFWWKHILUVWFHOODQGWKHQZLWK\RXUILQJHURQWKH6KLIWNH\VHOHFWWKHODVWFHOO'RQWIRU
JHWWKDW\RXFDQVSOLWWKHVFU HHQKRUL]RQWDOO\DQGRUYHUWLFDOO\WRPDN HWKHVHILUVWDQGODVWFHOOV
YLVLEOHZKHQWKHUDQJHLVODUJH7KLVVHOHFWVWKHHQWLUHUDQJHDQGLVHDVLHUWKDQVFUROOLQJ9

9WIXLI'XVP)RXIV
WLSVXGYXXSIRXIVE
JSVQYPEMREVERKI
EPP EXSRGI -XMW
IUYMZEPIRXXS
GST]MRK

([FHO 7LS (IILFLHQW&RS\LQJZLWK&WUO(QWHU

$QHDV\ZD\WRHQWHUWKHVDPHIRUPXODLQDUDQJHDOODWRQFHLVWRVHOHFWWKHUDQJHDVLQWKH
SUHFHGLQJ([FHO7LSW\SHWKHIRUPXODDQGSUHVV&WUO(QWHUERWKNH\VDWRQFH$IWHU\RX
JHWXVHGWRWKLVVKRUWFXW\RXZLOOSUREDEO\XVHLWDOOWKHWLPH

1HWSUHVHQWYDOXH The NPV is based on the sequence of cash flows in column B.

From the general discussion of NPV, to discount everything back to the beginning of year
1, the value in cell B13 should be multiplied by 1y(1 1 U)1, the value in cell B14 should be
multiplied by 1y(1 1 U)2, and so on, and these quantities should be summed to obtain the
NPV. (Here, U5 0.12 is the discount rate.) Fortunately, however, Excel has a built-in NPV
function to accomplish this calculation. To use it, enter the formula
'HYHORSPHQWBFRVW139'LVFRXQWBUDWH*URVVBPDUJLQ

8LIWXVIEQSJGEWL
JPS[WMRXLI24:
JYRGXMSRQYWXSGGYV
EX XLIIRHWSJ]IEV
]IEV ERHWSSR -J
XLI XMQMRKMWMVVIKYPEV
]SYGERHMWGSYRX
QERYEPP] SV]SY
GERYWI)\GIPW<24:
JYRGXMSR

in cell B34. The NPV function takes two arguments: the discount rate and a range of cash
flows. Furthermore, it assumes that the first cell in this range is the cash flow at the HQGof
year 1, the second cell is the cash flow at the end of year 2, and so on. This explains why
the development cost is subtracted RXWVLGHof the NPV functionit is incurred at the EHJLQ
QLQJof year 1. In general, any cash flow incurred at the beginning of year 1 must be placed
outside the NPV function.
To get some understanding of NPV, note that the VXPof the cash flows in column B is
slightly more than $34.14 million, but the NPV (aside from the development cost) is only
about $12.60 million. This is because values further into the future are discounted so heavily. At the extreme, the $1.2636 million cash flow in year 20 is equivalent to only
$1.2636[1y(1 1 0.12)20] 5 $0.131 million now.

([FHO )XQFWLRQ 139

7KH 139 IXQFWLRQ WDNHV WZR DUJXPHQWV WKH GLVFRXQW UDWH HQWHUHG DV D GHFLPDO VXFK DV
IRU DQG D VWUHDP RI FDVK IORZV 7KHVH FDVK IORZV DUH DVVXPHG WR RFFXU LQ FRQVHFXWLYH
\HDUV VWDUWLQJ DW WKH HQG RI \HDU ,I WKHUH LV DQ LQLWLDO FDVK IORZ DW WKH EHJLQQLQJ RI \HDU
VXFK DV DQ LQLWLDO LQYHVWPHQW LW VKRXOG EH HQWHUHG RXWVLGH WKH 139 IXQFWLRQ 7KHUH LV DOVR DQ
;139 IXQFWLRQ WKDW KDV three DUJXPHQWV D GLVFRXQW UDWH D VHULHV RI FDVK IORZV DQG D VHULHV
RI GDWHV ZKHQ WKH FDVK IORZV RFFXU %HFDXVH WKHVH GDWHV GR QRW KDYH WR EH HTXDOO\ VSDFHG

9You can find other tips like this for increasing your efficiency in the ([FHO7XWRULDO[OV[ file on this textbooks
essential resource Web site.

WKURXJK WLPH WKLV IXQFWLRQ LV FRQVLGHUDEO\ PRUH IOH[LEOH WKDQ WKH 139 IXQFWLRQ
:H ZLOO QRW XVH WKH ;139 IXQFWLRQ LQ WKLV ERRN EXW \RX FDQ OHDUQ PRUH DERXW LW LQ ([FHOV
RQOLQH KHOS

Deciding Whether to Continue with the Drug

NPV calculations are typically used to see whether a certain project should be undertaken.
If the NPV is positive, the project is worth pursuing. If the NPV is negative, the company
should look for other places to invest its money. Figure 2.34 shows that the NPV for this
drug is positive, over $3 million.10 Therefore, if Acron is comfortable with its predictions
of future cash flows, it should continue with the development and marketing of the drug.
However, Acron might first want to see how sensitive the NPV is to changes in the sales
predictions. After all, these predictions are intelligent guesses at best.
One possible sensitivity analysis appears in Figure 2.35. Here you can build a one-way
data table to see how the NPV changes when the number of years of increase (the input in
cell B7) changes. Again, the important question is whether the NPV stays positive. It certainly does when the input variable is greater than its current value of 8. However, if sales
start decreasing soon enoughthat is, if the value in B7 is 3 or lessthe NPV turns negative. This should probably not concern Acron, because its best guess for the years of
increase is considerably greater than 3.
*MKYVI
Sensitivity of
NPV to Years of
Sales Increase

D
E
11 Sensivity to years of increase (cell B7)
12
3.3003
13
3
-0.7190
4
0.1374
14
5
0.9687
15
6
1.7739
16
7
2.5516
17
8
3.3003
18
9
4.0181
19
10
4.7027
20

Another possibility is to see how long DQGhow good the good years are. To do this,
you can create the two-way data table shown in Figure 2.36, where cell B6 is the row input
cell and cell B7 is the column input cell. Now you can see that if sales increase through
year 6, all reasonable yearly increases result in a positive NPV. However, if sales increase
only through year 5, then a low enough yearly increase can produce a negative NPV. Acron
might want to step back and estimate how likely these bad scenarios are before proceeding
with the drug.
*MKYVI
Sensitivity of
NPV to Years of
Increase and
Yearly Increase

D
E
F
G
H
I
22 Sensivity to rate of increase in early years (cell B6) and years of increase (cell B7)
23
3.3003
5%
6%
7%
8%
9%
24
3
-1.3405
-1.2184
-1.0951
-0.9708
-0.8454
25
4
-0.8203
-0.6352
-0.4469
-0.2554
-0.0606
26
5
-0.3383
-0.0897
0.1652
0.4265
0.6943
27
6
0.1074
0.4195
0.7419
1.0750
1.4189
28
7
0.5182
0.8934
1.2838
1.6899
2.1123
29
8
0.8958
1.3330
1.7912
2.2711
2.7738
30
9
1.2413
1.7392
2.2643
2.8182
3.4023
31
10
1.5559
2.1125
2.7033
3.3306
3.9963

J
10%
-0.7190
0.1374
0.9687
1.7739
2.5516
3.3003
4.0181
4.7027

10You

might wonder why we didnt discount back to the beginning of the current year, year 0, instead of year 1.
This is a fairly arbitrary decision on our part. To discount back to year 0, you would simply divide the current
NPV by 1.12. The important point, however, is that this would have no bearing on Acrons decision: A positive NPV
would stay positive, and a negative NPV would stay negative.

(IGMWMSRW-RZSPZMRKXLI8MQI :EPYISJ1SRI]

Limitations of the Model

Probably the major flaw in this model is that it ignores uncertainty, and future cash flows
are highly uncertain, due mainly to uncertain demand for the drug. Incorporating uncertainty into this type of model will be covered when we discuss simulation in Chapters 10
and 11. Aside from this uncertainty, there are almost always ways to make DQ\model more
realisticat the cost of increased complexity. For example, you could model the impact of
competition on Niagras profitability. Alternatively, you could allow Acron to treat its
prices as decision variables. However, this might influence the likelihood of competition
entering the market, which would certainly complicate the model. The point is that this
model is only a start. When millions of dollars are at stake, a more thorough analysis is certainly warranted. Q

PROBLEMS
Skill-Building Problems
Modify Acrons model so that development lasts for
an extra year. Specifically, assume that development
costs of $7.2 million and $2.1 million are incurred at
the beginnings of years 1 and 2, and then the sales in
the current model occur one year later, that is, from
year 2 until year 21. Again, calculate the NPV discounted back to the beginning of year 1, and perform
the same sensitivity analyses. Comment on the effects
of this change in timing.
Modify Acrons model so that sales increase, then
stay steady, and finally decrease. Specifically, assume
that the gross margin is $1.2 million in year 1, then
increases by 10% annually through year 6, then stays
constant through year 10, and finally decreases by 5%
annually through year 20. Perform a sensitivity analysis with a two-way data table to see how NPV varies
with the length of the increase period (currently 6
years) and the length of the constant period (currently
4 years). Comment on whether Acron should pursue
the drug, given your results.
Create a one-way data table in the Acron model to see
how the NPV varies with discount rate, which is
allowed to vary from 8% to 18% in increments of
0.5%. Explain intuitively why the results go in the
direction they gothat is, the NPV decreases as the
discount rate increases. Should Acron pursue the drug
for all of these discount rates?

Skill-Extending Problems
The NPV function automatically discounts each of the
cash flows and sums the discounted values. Verify that

it does this correctly for Acrons model by calculating

the NPV the long way. That is, discount each cash
flow and then sum these discounted values. Use Excel
formulas to do this, but dont use the NPV function.
(+LQW Remember that the discounted value of $1
received Wyears from now is 1y(1 1 U)W dollars today.)
In a situation such as Acrons, where a one-time cost
is followed by a sequence of cash flows, the LQWHUQDO
UDWHRIUHWXUQ (IRR) is the discount rate that makes
the NPV equal to 0. The idea is that if the discount
rate is greater than the IRR, the company will not pursue the project, but if the discount rate is less than the
IRR, the project is financially attractive.
D Use Excels Goal Seek tool to find the IRR for the
Acron model.
E Excel also has an IRR function. Look it up in
online help to see how it works, and then use it on
Acrons model. Of course, you should get the same
IRR as in part D
F Verify that the NPV is negative when the discount
rate is slightly greater than the IRR, and that it is
positive when the discount rate is slightly less than
the IRR.
The XNPV function can calculate NPV for any
(possibly irregular) series of cash flows. Look this
function up in Excels online help. Then use it to
develop a spreadsheet model that finds the NPV of
the following series: a payment of $25,000 today
(assumed to be June 15, 2010), and cash inflows of
$10,000 on March 1, 2011; $15,000 on September 15,
2011; $8000 on January 20, 2012; $20,000 on April 1,
2012; and $10,000 on May 15, 2012. Discount these
back to today using a discount rate of 12%.

2.8 CONCLUSION
The examples in this chapter provide a glimpse of things to come in later chapters. You have
seen the spreadsheet modeling approach to realistic business problems, learned how to
design spreadsheet models for readability, and explored some of Excels powerful tools, particularly data tables. In addition, at least three important themes have emerged from these
examples: relating inputs and decision variables to outputs by means of appropriate formulas, optimization (for example, finding a best order quantity), and the role of uncertainty
(uncertain response rate or demand). Although you have not yet learned the tools to explore
these themes fully, you will have plenty of opportunities to do so in the rest of this book.

Summary of Key Management Science Terms

7HUP
Model inputs
Decision variables
Model outputs

Net present value (NPV)

Discount rate

([SODQDWLRQ
The numeric values that are given in any
problem statement
The variables a decision maker has control over
to obtain better solutions
The numeric values that result from combinations
of inputs and decision variables through the use
of logical formulas
The current worth of a stream of cash flows that
occur in the future
Interest rate used for discounting future cash flows
to get the net present value

3DJH
22
22
22

55
55

Summary of Key Excel Terms

7HUP
IF function

([SODQDWLRQ
Useful for implementing logic

Relative, absolute
cell addresses

Useful for copying formulas;

absolute row or column stays fixed,
relative row or column moves
Useful for making formulas more
meaningful

Range names

Pasting range names

Cell comments
One-way data table
Goal Seek
Formula Auditing
toolbar
I[ button

Provides a list of all range names in

the current workbook
Useful for documenting contents
of the cell
Shows how one or more outputs
vary as a single input varies
Solves one equation in one unknown
Useful for checking which cells are
related to other cells through formulas
Useful for getting help on
Excel functions

([FHO
=IF(FRQGLWLRQUHVXOW,I7UXH
UHVXOW,I)DOVH)
A1 (relative), $A1 or A$1 (mixed),
$A$1 (absolute); press F4 to
cycle through possibilities
Type name in Name box, or use
Create from Selection shortcut
(Ctrl+Shift+F3)
Use Paste List from Use
in Formula dropdown list (F3)
Right-click on cell, select Insert
Comment menu item
Use Data Table from What-If
Analysis dropdown list
Use Goal Seek from What-If
Analysis dropdown list
Use Formula Auditing buttons
on Formulas ribbon
On Formula Bar

3DJH
24
29

33
34
36
37
38
42
(FRQWLQXHG)

'SRGPYWMSR

Summary of Key Excel Terms

7HUP
VLOOKUP function
Two-way data table
SUMPRODUCT
function
Trendline tool

Conditional
formatting
Splitting screen
Efficient selection

Efficient copying
NPV function

(Continued)

([SODQDWLRQ
Useful for finding a particular value
based on a comparison
Shows how a single output varies
as two inputs vary
Calculates the sum of products of
values in two (or more) similarsized ranges
Superimposes the best-fitting line
or curve of a particular type on
a scatter chart or time series graph
Formats cells depending on whether
specified conditions hold
Useful for separating the screen
horizontally and/or vertically
Useful for selecting a large
rectangular range
Shortcut for copying a formula
to a range
Calculates NPV of a stream of cash
flows at the ends of consecutive
years, starting in year 1

([FHO
=VLOOKUP(YDOXH7R&RPSDUH
ORRNXS7DEOHFROXPQ7R5HWXUQ)
Use Data Table from What-If
Analysis dropdown list
=SUMPRODUCT(UDQJHUDQJH)

3DJH
42
43
44

With chart selected, right-click on

any point and select Add Trendline

Use Conditional Formatting

on Home ribbon
Use screen splitters at top and right
of scrollbars
While pressing the Shift key, click
on upper-left and bottom-right cells
of range
Select the range, enter the formula,
and press Ctrl+Enter
=NPV(GLVFRXQW5DWHFDVK)ORZV)

51
57
58

58
58

PROBLEMS
Skill-Building Problems
Julie James is opening a lemonade stand. She believes
the fixed cost per week of running the stand is $50.00.
Her best guess is that she can sell 300 cups per week
at $0.50 per cup. The variable cost of producing a cup
of lemonade is $0.20.
D Given her other assumptions, what level of sales
volume will enable Julie to break even?
E Given her other assumptions, discuss how a change
in sales volume affects profit.
F Given her other assumptions, discuss how a change
in sales volume and variable cost jointly affect
profit.
G Use Excels Formula Auditing tools to show which
cells in your spreadsheet affect profit directly.
You are thinking of opening a Broadway play, ,/RYH
<RX<RXUH0HGLRFUH1RZ*HW%HWWHUIt will cost
$5 million to develop the show. There are 8 shows
per week, and you project the show will run for
100 weeks. It costs $1000 to open the theater each
night. Tickets sell for $50.00, and you earn an average
of $1.50 profit per ticket holder from concessions. The
theater holds 800, and you expect 80% of the seats to
be full.

D Given your other assumptions, how many weeks

will the play have to run for you to earn a 100%
return on the plays development cost?
E Given your other assumptions, how does an
increase in the percentage of seats full affect profit?
F Given your other assumptions, determine how a
joint change in the average ticket price and number
of weeks the play runs influence profit.
G Use Excels Formula Auditing tools to show which
cells in the spreadsheet are directly affected by the
percentage of seats full.
You are thinking of opening a small copy shop. It
costs $5000 to rent a copier for a year, and it costs
$0.03 per copy to operate the copier. Other fixed costs
of running the store will amount to $400 per month.
You plan to charge an average of $0.10 per copy, and
the store will be open 365 days per year. Each copier
can make up to 100,000 copies per year.
D For one to five copiers rented and daily demands
of 500, 1000, 1500, and 2000 copies per day, find
annual profit. That is, find annual profit for HDFKof
these combinations of copiers rented and daily
demand.
E If you rent three copiers, what daily demand for
copies will allow you to break even?

F Graph profit as a function of the number of copiers

for a daily demand of 500 copies; for a daily
demand of 2000 copies. Interpret your graphs.
Georgia McBeal is trying to save for her retirement.
She believes she can earn 10% on average each year
on her retirement fund. Assume that at the beginning
of each of the next 40 years, Georgia will allocate [
dollars to her retirement fund. If at the beginning of a
year Georgia has \dollars in her fund, by the end of
the year, it will grow to 1.1\dollars. How much
should Georgia allocate to her retirement fund each
year to ensure that she will have $1 million at the end
of 40 years? What key factors are being ignored in this
analysis of the amount saved for retirement?
A European call option on a stock earns the owner an
amount equal to the price at expiration minus the exercise price, if the price of the stock on which the call is
written exceeds the exercise price. Otherwise, the call
pays nothing. A European put option earns the owner
an amount equal to the exercise price minus the price
at expiration, if the price at expiration is less than the
exercise price. Otherwise, the put pays nothing. The
file 3B[OV[ contains a template that finds (based
on the well-known BlackScholes formula) the price
of a European call and put based on the following
inputs: todays stock price, the duration of the option
(in years), the options exercise price, the risk-free rate
of interest (per year), and the annual volatility in stock
price. For example, a 40% volatility means approximately that the standard deviation of annual percentage changes in the stock price is 40%.
D Consider a six-month European call option with
exercise price $40. Assume a current stock price of
$35, a risk-free rate of 5%, and an annual volatility
of 40%. Determine the price of the call option.
E Use a data table to show how a change in volatility
changes the value of the option. Give an intuitive
explanation for your results.
F Use a data table to show how a change in todays
stock price changes the options value. Give an
intuitive explanation for your results.
G Use a data table to show how a change in the
options duration changes the options value. Give
an intuitive explanation for your results.
Repeat parts DGof the previous problem for a
six-month European put option with exercise price
$40. Again, assume a current stock price of $35, a
risk-free rate of 5%, and an annual volatility of 40%.
The file 3B[OV[ lists sales (in millions of dollars)
of Dell Computer during the period 19871997 (where
year 1 corresponds to 1987).
D Fit a power and an exponential trend curve to these
data. Which fits the data better?
E Use your part Danswer to predict 1999 sales for
Dell.

F Use your part Danswer to describe how the sales of

Dell have grown from year to year.
G Search the Web for more recent Dell sales data.
Then repeat the preceding parts using all of the data.
Dataware is trying to determine whether to give a
$10 rebate, cut the price $6, or have no price change
on a software product. Currently, 40,000 units of the
product are sold each week for $45 apiece. The variable cost of the product is $5. The most likely case
appears to be that a $10 rebate will increase sales
30%, and half of all people will claim the rebate.
For the price cut, the most likely case is that sales
will increase 20%.
D Given all other assumptions, what increase in sales
from the rebate would make the rebate and price
cut equally desirable?
E Dataware does not really know the increase in sales
that will result from a rebate or price cut. However,
the company is sure that the rebate will increase
sales by between 15% and 40% and that the price
cut will increase sales by between 10% and 30%.
Perform a sensitivity analysis that could be used to
help determine Datawares best decision.
The file 3B[OV[ gives the annual sales for
Microsoft (in millions of dollars) for the years
19841993, where 1984 = year 1.
D Fit an exponential curve to these data.
E Assuming you are back in 1993, by what percentage do you estimate that Microsoft has grown each
year, based on this historical data?
F Why cant a high rate of exponential growth
continue for a long time?
G Rather than an exponential curve, what curve might
better represent the growth of a new technology?
H Search the Web for more recent Microsoft sales
data. Then repeat the preceding parts using all
the data.
Assume that the number of units sold of a product is
given by 100 2 0.53 1 26$
w, where 3is the price
(in dollars) charged for the product and $is the
amount spent on advertising (in thousands of dollars).
Each unit of the product costs $5 to produce. Use a
data table to find the combination of price and advertising that maximizes profit.
A company manufacturers a product in the U.S. and
sells it in England. The unit cost of manufacturing is
$50. The current exchange rate (dollars per pound) is
1.51. The demand function, which indicates how many
units the company can sell in England as a function of
price (in pounds) is of the power type, with constant
27556759 and exponent 22.4.
D Develop a model for the companys profit (in dollars) as a function of the price it charges (in
pounds). Then use a data table to find the profitmaximizing price to the nearest pound.

'SRGPYWMSR

E If the exchange rate varies from its current value, does

the profit-maximizing price increase or decrease?
Does the maximum profit increase or decrease?
The yield of a chemical reaction is defined as the ratio
(expressed as a percentage) of usable output to the
amount of raw material input. Suppose the yield of a
chemical reaction depends on the length of time the
process is run and the temperature at which the process
is run. The yield can be expressed as follows:
Yield 5 90.79 2 1.095[1 2 1.045[2 2 2.78121
2 2.524[22 2 0.775[1[2
Here [1 5 (Temperature 2 125)y10 and [2 5 (Time
2 300)y30, where temperature is measured in degrees
Fahrenheit, and time is measured in seconds. Use a
data table to find the temperature and time settings that
maximize the yield of this process.
A bond is currently selling for $1040. It pays the
amounts listed in the file 3B[OV[ at the ends of
the next six years. The yield of the bond is the interest
rate that would make the NPV of the bonds payments
equal to the bonds price. Use Excels Goal Seek tool
to find the yield of the bond.
Assume the demand for a companys drug Wozac during
the current year is 50,000, and assume demand will grow
at 5% a year. If the company builds a plant that can produce [units of Wozac per year, it will cost $16[. Each
unit of Wozac is sold for $3. Each unit of Wozac produced incurs a variable production cost of $0.20. It costs
$0.40 per year to operate a unit of capacity. Determine
how large a Wozac plant the company should build to
maximize its expected profit over the next 10 years.
Consider a project with the following cash flows: year
1, 2$400; year 2, $200; year 3, $600; year 4, 2$900;
year 5, $1000; year 6, $250; year 7, $230. Assume a
discount rate of 15% per year.
D Find the projects NPV if cash flows occur at the
ends of the respective years.
E Find the projects NPV if cash flows occur at the
beginnings of the respective years.
F Find the projects NPV if cash flows occur at the
middles of the respective years.
A software company is considering translating its
program into French. Each unit of the program sells
for $50 and incurs a variable cost of $10 to produce.
Currently, the size of the market for the product is
300,000 units per year, and the English version of the
software has a 30% share of the market. The company
estimates that the market size will grow by 10% a year
for the next five years, and at 5% per year after that. It
will cost the company $6 million to create a French
version of the program. The translation will increase
its market share to 40%. Given a 10-year planning
horizon, for what discount rates is it profitable to
create the French version of the software?

The payback of a project is the number of years it takes

before the projects total cash flow is positive. Payback
ignores the time value of money. It is interesting, however, to see how differing assumptions on project
growth impact payback. Suppose, for example, that a
project requires a $300 million investment at year 0
(right now). The project yields cash flows for 10 years,
and the year 1 cash flow will be between $30 million
and $100 million. The annual cash flow growth will be
between 5% and 25% per year. (Assume that this
growth is the VDPHeach year.) Use a data table to see
how the project payback depends on the year 1 cash
flow and the cash flow growth rate.

Skill-Extending Problems
You are entering the widget business. It costs $500,000,
payable in year 1, to develop a prototype. This cost can
be depreciated on a straight-line basis during years 15.
Each widget sells for $40 and incurs a variable cost of
$20. During year 1, the market size is 100,000, and the
market is growing at 10% per year. You believe you
will attain a 30% market share. Profits are taxed at
40%, but there are no taxes on QHJDWLYH profits.
D Given your other assumptions, what market share is
needed to ensure a total free cash flow (FCF) of $0
over years 1 to 5? (1RWHFCF during a year equals
after-tax profits plus depreciation minus fixed
costs, if any.)
E Explain how an increase in market share changes
profit.
F Explain how an increase in market size growth
changes profit.
G Use Excels auditing tool to show how the market
growth assumption influences your spreadsheet.
Suppose you are borrowing $25,000 and making
monthly payments with 1% interest. Show that the
monthly payments should equal $556.11. The key
relationships are that for any month W
(Ending month Wbalance)
5 (Ending month W2 1 balance)
2 ((Monthly payment) 2 (Month Winterest))
(Month Winterest) 5 (Beginning month Wbalance) 3
(Monthly interest rate)
Of course, the ending month 60 balance must equal 0.
You are thinking of starting Peaco, which will produce
Peakbabies, a product that competes with Tys Beanie
Babies. In year 0 (right now), you will incur costs of $4
million to build a plant. In year 1, you expect to sell
80,000 Peakbabies for a unit price of $25. The price of
$25 will remain unchanged through years 1 to 5. Unit
sales are expected to grow by the same percentage (J)
each year. During years 1 to 5, Peaco incurs two types
of costs: variable costs and SG&A (selling, general, and
administrative) costs. Each year, variable costs equal

half of revenue. During year 1, SG&A costs equal 40%

of revenue. This percentage is assumed to drop 2% per
year, so during year 2, SG&A costs will equal 38% of
revenue, and so on. Peacos goal is to have profits for
years 0 to 5 sum to 0 (ignoring the time value of
money). This will ensure that the $4 million investment
in year 0 is paid back by the end of year 5. What annual
percentage growth rate Jdoes Peaco require to pay
back the plant cost by the end of year 5?
Suppose the demand (in thousands) for a toaster is
given by 100S22, where S is the price in dollars
charged for the toaster.
D If the variable cost of producing a toaster is $10,
what price maximizes profit?
E The elasticity of demand is defined as the percentage change in demand created by a 1% change in
price. Using a data table, show that the demand for
toasters has constant elasticity, that is, the elasticity
doesnt depend on the price. Would this be true if
the demand for toasters were linear in price?

hedging expenses depends on the number of puts

purchased and the final $/yen exchange rate.
Assume final exchange rates between 0.006 $/yen
and 0.015 $/yen are possible.
The file 3B[OV[ contains a template for a car
loan. Specifically, once values are entered in the blue
cells, you need to enter formulas in the gray cells to
calculate the amount financed, the monthly payment
(assuming that monthly payments stay the same
throughout the term of the loan), the total interest paid,
and an amortization schedule. For the latter, fill in the
HQWLUH gray area with formulas, but use IF functions so
that blanks appear past the term of the loan.
The IRR is the discount rate Uthat makes a project have
an NPV of $0. You can find IRR in Excel with the
built-in IRR function, using the syntax =IRR(range of
cash flows). However, it can be tricky. In fact, if the
IRR is not near 10%, this function might not find an
answer, and you would get an error message. Then you
must try the syntax =IRR(range of cash flows, guess),
where guess is your best guess for the IRR. It is best
to try a range of guesses (say, 290% to 100%). Find
the IRR of the project described in Problem 34.

The file 3B[OV[ contains the cumulative

number of bits (in trillions) of DRAM (a type of
computer memory) produced and the price per bit
(in thousandths of a cent).
D Fit a power curve that can be used to show how
price per bit drops with increased production. This
relationship is known as the learning curve.
E Suppose the cumulative number of bits doubles. Create
a prediction for the price per bit. Does the change in
the price per bit depend on the current price?

A project does not necessarily have a unique IRR.

(Refer to the previous problem for more information
on IRR.) Show that a project with the following cash
flows has two IRRs: year 1, 2$20; year 2, $82; year 3,
2$60; year 4, $2. (1RWH It can be shown that if the
cash flow of a project changes sign only once, the
project is guaranteed to have a unique IRR.)

A large U.S. drug company, Pharmco, has 100 million

yen coming due in one year. Currently the yen is
worth $0.01. Because the value of the yen in U.S.
dollars in one year is unknown, the value of this
100 million yen in U.S. dollars is highly uncertain. To
hedge its risk, Pharmco is thinking of buying one-year
put options on the yen with an exercise price of
$0.008. For example, if the yen falls in value a year
from now to $0.007, the owner of the put receives
$0.001. The price of such a put is $0.00007. Show
how the dollar value of Pharmcos receipts and

The file 3B[OV[ contains data on prices of products for several of a chain stores locations, a discount
schedule offered to customers depending on how
much they spend, and commission rates of the salespeople at the various stores. Your job is to develop an
invoice form. Specifically, you should enter formulas
in the gray cells so that whenever data are entered in
the blue cells, the formulas in the gray cells calculate
automatically. As an extra, use data validation in cell
B23 so that the user can choose a city from a list of
cities where the chain has its stores.

%44)2(-< TIPS FOR EDITING AND

DOCUMENTING SPREADSHEETS
Editing and documenting your spreadsheet models is crucial, and the following tips make
these tasks much easier.

Format Appropriately
Appropriate formatting can make a spreadsheet model much easier to read. To boldface, for
example, select one or more cells and click on the % button on the Home ribbon (or press Ctrl+B).
Similarly, to italicize, indent, increase or decrease the number of decimal places, right-justify, or
perform other common formatting tasks, use the buttons on the Home ribbon or shortcut keys.
%TTIRHM\ 8MTWJSV)HMXMRKERH(SGYQIRXMRK7TVIEHWLIIXW

Use Range Names

Naming ranges takes time but makes formulas much easier to read and understand. To
enter a range name, highlight any cell or range of cells and enter a name for the range in the
Name box (just to the left of the Formula Bar). If you want to edit or delete range names,
select Name Manager on the Formulas ribbon. Here are some other options you have from
the Defined Names group on the Formulas ribbon.
Q

After you have named some ranges, you can get a list of them in your spreadsheet by
placing the cursor at the top of the range where you want the list to be placed,
selecting the Use in Formula dropdown list on the Formulas ribbon, and clicking on
the Paste List option. Alternatively, you can press the F3 button.
Suppose you have labels such as Fixed Cost, Variable Cost, Revenue, and Profit in the
range A3:A6, with their values next to them in column B. If you want to name the
cells in column B with the labels in column A, highlight the range A3:B6, select
Create from Selection on the Formulas ribbon (or press Ctrl+Shift+F3), and make
sure the Left Column box is checked. This creates the range names you want. A similar
trick works if you have descriptive labels DERYH columns of data you want
to name.
If you have a formula, such as =SUM(A10:A20), and then you name the range
A10:A20 Costs, say, the formula does QRWchange automatically to =SUM(Costs).
However, you can make it adapt to your new range name by selecting Apply Names
from the Define Name dropdown list on the Formulas ribbon.
Sometimes you might want to use the VDPHrange name, such as Total_cost, on
multiple worksheets of a workbook. For example, you might want Total_cost to refer
to cell B26 in Sheet1 and to cell C59 in Sheet2. The trick is to use a VKHHW-level name
rather than a ZRUNERRNlevel name for one or both versions of Total_cost. This is easy
to do from the Name Manager. When you define a new name, just select a worksheet
as the Scope of the name.

Use Text Boxes

Text boxes are very useful for documenting your work. To enter an explanation or any
other text into a text box, click on the Text Box button on the Insert ribbon, drag a box, and
start typing. This technique is PXFK better than typing explanations into cells because text
boxes have word wrap. Therefore, text in text boxes is much easier to edit than text in cells.

Use Cell Comments

Cell comments provide another good way to document your work. To enter a comment in
a cell, select the cell and right-click. This brings up a dialog box (which is also useful for
other tasks such as formatting). Click on the Insert Comment item to enter a comment. If a
comment is already in the cell, this menu will contain Edit Comment and Delete Comment
items. The cells with comments should have small red triangles in their corners. When you
hover the cursor over the cell, the comment appears.

Other Tips
Finally, we urge you once again to open the ([FHO7XWRULDO[OV[ file on the Essential
Resource Web site and work through it. The file includes a number of techniques that will
make you a better and more efficient Excel user.

Flight Crew Operating Manual: AVRO 146-RJ
100% (1)
Flight Crew Operating Manual: AVRO 146-RJ
520 pages
2 ASSIGNMENT 2 (Beginning Superstore)
0% (1)
2 ASSIGNMENT 2 (Beginning Superstore)
1 page
Excel Practice
No ratings yet
Excel Practice
40 pages
Excel Practice Sheet
No ratings yet
Excel Practice Sheet
156 pages
BISM7202 Excel Assignment 26 Mar 2019
No ratings yet
BISM7202 Excel Assignment 26 Mar 2019
9 pages
Beginning Exercises With VBA
No ratings yet
Beginning Exercises With VBA
3 pages
Beyond Pie Charts Tutorial Dari Excel
No ratings yet
Beyond Pie Charts Tutorial Dari Excel
82 pages
Naked Statistics: Stripping The Dread From The Data Practical Business Statistics, Sixth Edition
No ratings yet
Naked Statistics: Stripping The Dread From The Data Practical Business Statistics, Sixth Edition
2 pages
Data Analysis with Excel: Tips and tricks to kick start your excel skills
From Everand
Data Analysis with Excel: Tips and tricks to kick start your excel skills
Manisha Nigam
No ratings yet
CARTY BINN MASTERING EXCEL POWER PIVOT - POWER QUERY - Step by Step Practical Guide To Master Excel P
No ratings yet
CARTY BINN MASTERING EXCEL POWER PIVOT - POWER QUERY - Step by Step Practical Guide To Master Excel P
138 pages
Tutorial On Feasibility Study
No ratings yet
Tutorial On Feasibility Study
4 pages
XLOOKUP Guide
No ratings yet
XLOOKUP Guide
3 pages
Edp Report Learning Worksheet Fundamentals
No ratings yet
Edp Report Learning Worksheet Fundamentals
27 pages
Resume - Rajat Chaturvedi
No ratings yet
Resume - Rajat Chaturvedi
3 pages
Excel Practice Data For Pivot Table 1
No ratings yet
Excel Practice Data For Pivot Table 1
47 pages
Mastering MS Excel Your Ultimate Productivity Toolkit
No ratings yet
Mastering MS Excel Your Ultimate Productivity Toolkit
10 pages
TIB Bwce 2.6.2 Migration
0% (1)
TIB Bwce 2.6.2 Migration
96 pages
Excel Array Formulas
No ratings yet
Excel Array Formulas
12 pages
Unit 3 Introduction To Ms-Excel
No ratings yet
Unit 3 Introduction To Ms-Excel
12 pages
Proc SQL
100% (1)
Proc SQL
7 pages
Advanced Excel
No ratings yet
Advanced Excel
3 pages
Content 4 CURSO EXCEL
No ratings yet
Content 4 CURSO EXCEL
25 pages
Cloud Lab
No ratings yet
Cloud Lab
39 pages
Qlikview Interview Questions Asked in Accenture
No ratings yet
Qlikview Interview Questions Asked in Accenture
1 page
Excel Dax Tutorial PDF
No ratings yet
Excel Dax Tutorial PDF
23 pages
Power BI
No ratings yet
Power BI
108 pages
What If Analysis
No ratings yet
What If Analysis
7 pages
Excel 2016 Charts and Graphs
No ratings yet
Excel 2016 Charts and Graphs
38 pages
(M) BROCHURE - Data Science Learning Path
No ratings yet
(M) BROCHURE - Data Science Learning Path
33 pages
Access Part 2 Bangla Book
0% (1)
Access Part 2 Bangla Book
16 pages
Case Study 4
No ratings yet
Case Study 4
10 pages
Linear Algebra Using R
No ratings yet
Linear Algebra Using R
17 pages
Excel Shortcutss
No ratings yet
Excel Shortcutss
2 pages
modelling with excel (online)
100% (2)
modelling with excel (online)
159 pages
Step by Step Business Math and Statistics Sneak Preview
No ratings yet
Step by Step Business Math and Statistics Sneak Preview
42 pages
Cheat Sheet: With Stata 15
No ratings yet
Cheat Sheet: With Stata 15
1 page
Data Analyst With MS Excel
No ratings yet
Data Analyst With MS Excel
9 pages
Excel 2016 Intermediate
0% (1)
Excel 2016 Intermediate
157 pages
Beginning Excel
0% (1)
Beginning Excel
336 pages
Excel Interview Questions Answers
No ratings yet
Excel Interview Questions Answers
7 pages
Things You Can Make in Excel: All About Numbers
No ratings yet
Things You Can Make in Excel: All About Numbers
8 pages
Powerful Forecasting With MS Excel Sample
No ratings yet
Powerful Forecasting With MS Excel Sample
257 pages
Power BI Interview Questions 1717598715
No ratings yet
Power BI Interview Questions 1717598715
24 pages
Solver Frontline Solvers Reference Guide
No ratings yet
Solver Frontline Solvers Reference Guide
305 pages
100 SQL Formulas Each Student Should Know
No ratings yet
100 SQL Formulas Each Student Should Know
10 pages
Introduction To Spreadsheet Modeling
No ratings yet
Introduction To Spreadsheet Modeling
32 pages
SQL Queries and PL/SQL
No ratings yet
SQL Queries and PL/SQL
92 pages
Solving Sudoku Using Excel
No ratings yet
Solving Sudoku Using Excel
8 pages
Hands-On Lab 5 - Cleaning Data
No ratings yet
Hands-On Lab 5 - Cleaning Data
5 pages
Stats Excel 2013 Xlstat Card PDF
No ratings yet
Stats Excel 2013 Xlstat Card PDF
6 pages
Essential of Microsoft Excel
No ratings yet
Essential of Microsoft Excel
1 page
Anil Nahar - Ready To Use 101 Powerful Excel VBA Code Just Copy - Paste - Run (For Functional Users) (2017) PDF
100% (1)
Anil Nahar - Ready To Use 101 Powerful Excel VBA Code Just Copy - Paste - Run (For Functional Users) (2017) PDF
123 pages
Using Excel For Problem Solving - Part II - Pivot Tables: Learning Objectives
No ratings yet
Using Excel For Problem Solving - Part II - Pivot Tables: Learning Objectives
5 pages
Infosys Placement Paper at Vaddeswara1
No ratings yet
Infosys Placement Paper at Vaddeswara1
5 pages
Introduction To R: Arin Basu MD MPH Dataanalytics
No ratings yet
Introduction To R: Arin Basu MD MPH Dataanalytics
33 pages
Keyboard Shortcuts
100% (1)
Keyboard Shortcuts
1 page
Microsoft Office Excel 2003 Intermediate III: Formulas and Worksheets
No ratings yet
Microsoft Office Excel 2003 Intermediate III: Formulas and Worksheets
8 pages
Week 1: Practice Challenge - Taking Charge of Excel
No ratings yet
Week 1: Practice Challenge - Taking Charge of Excel
5 pages
Prezi Cookbook
From Everand
Prezi Cookbook
Charlotte Olsson
No ratings yet
Mastering Prezi for Business Presentations
From Everand
Mastering Prezi for Business Presentations
Russell Anderson-Williams
No ratings yet
Excel 2007 Power Programming with VBA
From Everand
Excel 2007 Power Programming with VBA
John Walkenbach
4/5 (9)
Writing A Business Plan
100% (1)
Writing A Business Plan
6 pages
Birla Cements Marketing
90% (10)
Birla Cements Marketing
45 pages
Des, Double Des Aes
No ratings yet
Des, Double Des Aes
11 pages
Pro4801 02 Mark079100
No ratings yet
Pro4801 02 Mark079100
12 pages
Organic Agriculture Gr12 Module8.Final For Student
100% (1)
Organic Agriculture Gr12 Module8.Final For Student
25 pages
English: Two-Way Radio Model MT 550
No ratings yet
English: Two-Way Radio Model MT 550
9 pages
Aquatec Fortuna
No ratings yet
Aquatec Fortuna
59 pages
Final Research
No ratings yet
Final Research
65 pages
Heavy Oil Upgrading by Supercritical Water Treatment
No ratings yet
Heavy Oil Upgrading by Supercritical Water Treatment
20 pages
Individual Performance Commitment and Review Form (Ipcrf) For Teacher I-Iii
No ratings yet
Individual Performance Commitment and Review Form (Ipcrf) For Teacher I-Iii
22 pages
Lecture Notes On Operations Research 3OR
No ratings yet
Lecture Notes On Operations Research 3OR
51 pages
En Banc: Upreme !court
No ratings yet
En Banc: Upreme !court
13 pages
Elle UK October 2020
No ratings yet
Elle UK October 2020
192 pages
SolarPower Europe Apricum
No ratings yet
SolarPower Europe Apricum
25 pages
Aa 036335 001
No ratings yet
Aa 036335 001
1 page
Supernova ZLND E
0% (1)
Supernova ZLND E
26 pages
New STW PVT LTD GST Certi
No ratings yet
New STW PVT LTD GST Certi
3 pages
UCF All Questions
No ratings yet
UCF All Questions
132 pages
Ramanan Ramanathan
No ratings yet
Ramanan Ramanathan
6 pages
Hollow Core Diaphragm Design PDF
No ratings yet
Hollow Core Diaphragm Design PDF
49 pages
Cervical Spondylotic Myelopathy: A Guide To Diagnosis and Management
No ratings yet
Cervical Spondylotic Myelopathy: A Guide To Diagnosis and Management
11 pages
John Carlo C. Tolentino March 3, 2022 Bsma-3A Mr. Jefferson Cruz 3
No ratings yet
John Carlo C. Tolentino March 3, 2022 Bsma-3A Mr. Jefferson Cruz 3
5 pages
Government of Andhra Pradesh Abstract Allowances
No ratings yet
Government of Andhra Pradesh Abstract Allowances
3 pages
CPD Plan
No ratings yet
CPD Plan
1 page
Fundamentals of Logistics Fundamentals Final-Coverage
100% (1)
Fundamentals of Logistics Fundamentals Final-Coverage
39 pages
Dec. 29, 1970 Jean-Claude Asscher 3,550,228: Filed Aug. 6, 1968 4 Sheets-Sheet L
No ratings yet
Dec. 29, 1970 Jean-Claude Asscher 3,550,228: Filed Aug. 6, 1968 4 Sheets-Sheet L
6 pages
Philippine Laws, Statutes and Codes - Chan Robles Virtual Law Library
No ratings yet
Philippine Laws, Statutes and Codes - Chan Robles Virtual Law Library
3 pages
Airblue - View Reservation
No ratings yet
Airblue - View Reservation
1 page

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.

Introduction To Spreadsheet Modeling - Winston Albright

Uploaded by

Introduction To Spreadsheet Modeling - Winston Albright

Uploaded by

',%48)6

Introduction to Spreadsheet Modeling

ER] SJ QEREKIQIRX WGMIRGIW QSWX WYGGIWWJYP ETTPMGEXMSRW EVI XVEHMXMSREP

4EPXMIPERH*VIIHFIVK  MRZIWXMKEXIXLIGSWXWERHFIRIJMXWSJHIZIPSTMRKERH

2.2 BASIC SPREADSHEET MODELING:

A clear, logical layout to the overall model

2.1 O RDERING NCAA T-S HIRTS

,) $1' 6FRUH6FRUH )DLO3DVV

Range names used

Fixed order cost

Range names used

 &EWMG7TVIEHWLIIX1SHIPMRK 'SRGITXWERH&IWX4VEGXMGIW

Range names used

2.3 COST PROJECTIONS

he Woodworks Company produces a variety of custom-designed wood furniture for its

8EFPI Input Data for Manufacturing a Bookshelf

%XVLQHVV2EMHFWLYHV To build a model that allows Woodworks to see, numerically and

8EFPI Key Variables for the Bookshelf Manufacturing Example

Projected Bookshelf Costs

Years from Now

Cost per board-foot

Cost per hour

Cost per bookshelf

To calculate projected hourly labor costs, enter the

in cell D19. Then enter the formula

Using the Model for What-If Questions

The model in Figure 2.6 can now be used to answer

*MKYVI Effect of Higher Increases in Wood Costs

Projected Bookshelf Costs

Years from Now

Cost per board-foot

Cost per hour

Cost per bookshelf

2.4 BREAKEVEN ANALYSIS

2.3 B REAKEVEN A NALYSIS

he Quality Sweaters Company sells hand-knitted sweaters. The company is planning to

8EFPI Key Variables in Quality Sweaters Problem

Various unit costs, average order size, response rate

*MKYVI Quality Sweaters Model

([FHO 7LS &UHDWLQJ5DQJH1DPHV

7RFUHDWHDU DQJHQDPHIRUDU DQJHRIFHOOV ZKLF KFRXOGEHDVLQJOHFHOO KLJKOLJKWWKH

([FHO 7LS 3DVWLQJ5DQJH1DPHV

,QFOXGLQJDOLVWRIWKHU DQJHQDPHVLQ\RXUVSU HDGVKHHWLVXVHIXOIRUGRFXPHQWDWLRQSXU

DEVELOPING THE SPREADSHEET MODEL

Model of revenue, costs, and profit

([FHO 7LS ,QVHUWLQJ&HOO&RPPHQWV

,QVHUWLQJFRPPHQWVLQFHOOVLVDJU HDWZD\WRGRFXPHQW\RXUVSU HDGVKHHWPRGHOVZLWKRXW

',%2+)7 -2 )<')0 

Forming a One-Way Data Table

Prot versus Response Rate

([FHO 7RRO 2QH:D\'DWD7DEOH

97-2+ +3%0 7))/

([FHO 7RRO *RDO6HHN

Limitations of the Model

Using the Formula Auditing Tool

Model of revenue, costs, and profit

([FHO 7RRO )RUPXOD$XGLWLQJ7RROEDU

that allows the company to break even. Then chart

In the Quality Sweaters model, the range E9:E11 does

As the Quality Sweaters problem is now modeled, if

2.5 ORDERING WITH QUANTITY DISCOUNTS

Unit prices, table of unit costs specifying quantity discount structure

DEVELOPING THE SPREADSHEET MODEL

*MKYVI Sams Profit Model

Quantity discount structure

([FHO 7RRO I[ %XWWRQDQG)XQFWLRQ/LEUDU\*URXS

,I\RXZDQWWROHDUQPRUHDERXWKRZDQ([FHO IXQFWLRQRSHUDWHVFOLFNRQWKHI[ EXWWRQQH[W

Two-Way Data Table

*MKYVI Profit as a Function of Order Quantity and Demand

([FHO 7RRO 7ZR:D\'DWD7DEOH

$WZRZD\GDWDWDEOHDOORZV\RXWRVHHKRZDVLQJOH output FHOOYDULHVDV\RXYDU\WZR

*MKYVI Comparison of Expected Profits

Expected Prot versus Order Quanty

([FHO )XQFWLRQ 680352'8&7

7KH680352'8&7IXQFWLRQWDNHVWZRUDQJHDUJXPHQWVZKLFKPXVWEHexactly the same

In some ordering problems, like the one for Sams

4EPXMIPERH*VIIHFIVK MRZIWXMKEXIXLIGSWXWERHFIRIJMXWSJHIZIPSTMRKERH

,)$1'6FRUH6FRUH)DLO3DVV

&EWMG7TVIEHWLIIX1SHIPMRK 'SRGITXWERH&IWX4VEGXMGIW

8EFPI Input Data for Manufacturing a Bookshelf

%XVLQHVV2EMHFWLYHV To build a model that allows Woodworks to see, numerically and

8EFPI Key Variables for the Bookshelf Manufacturing Example

*MKYVI Effect of Higher Increases in Wood Costs

8EFPI Key Variables in Quality Sweaters Problem

*MKYVI Quality Sweaters Model

([FHO 7LS &UHDWLQJ5DQJH1DPHV

7RFUHDWHDU DQJHQDPHIRUDU DQJHRIFHOOVZKLF KFRXOGEHDVLQJOHFHOOKLJKOLJKWWKH

([FHO 7LS 3DVWLQJ5DQJH1DPHV

,QFOXGLQJDOLVWRIWKHU DQJHQDPHVLQ\RXUVSU HDGVKHHWLVXVHIXOIRUGRFXPHQWDWLRQSXU

([FHO 7LS ,QVHUWLQJ&HOO&RPPHQWV

',%2+)7 -2 )<')0

([FHO 7RRO 2QH:D\'DWD7DEOH

([FHO 7RRO *RDO6HHN

([FHO 7RRO )RUPXOD$XGLWLQJ7RROEDU

*MKYVI Sams Profit Model

([FHO 7RRO I[ %XWWRQDQG)XQFWLRQ/LEUDU\*URXS

,I\RXZDQWWROHDUQPRUHDERXWKRZDQ([FHO IXQFWLRQRSHUDWHVFOLFNRQWKHI[ EXWWRQQH[W

*MKYVI Profit as a Function of Order Quantity and Demand

([FHO 7RRO 7ZR:D\'DWD7DEOH

$WZRZD\GDWDWDEOHDOORZV\RXWRVHHKRZDVLQJOH output FHOOYDULHVDV\RXYDU\WZR

*MKYVI Comparison of Expected Profits

Expected Prot versus Order Quanty

([FHO )XQFWLRQ 680352'8&7

7KH680352'8&7IXQFWLRQWDNHVWZRUDQJHDUJXPHQWVZKLFKPXVWEHexactly the same

*MKYVI Best-Fitting Exponential Curve

*MKYVI Finding the Best-Fitting Curve Overall

$YHUDJHSHUFHQWDJHHUURUV Apply Equation (2.1) to calculate APEs in columns H, I,

$%6&(y&

8EFPI Key Variables for Golf Club Problem

Parameters of best-4ng power curve (from Esmaon sheet)

3UHGLFWHGGHPDQG Calculate the SUHGLFWHGdemand in cell B14 with the formula

([FHO 7LS &RQGLWLRQDO)RUPDWWLQJ

(IGMWMSRW-RZSPZMRKXLI8MQI :EPYISJ1SRI]

8EFPI Key Variables for Acrons Problem

([FHO 7LS 6SOLWWLQJWKH6FUHHQ

&DVKIORZV Start by entering the formula

([FHO 7LS (IILFLHQW6HOHFWLRQ

([FHO 7LS (IILFLHQW&RS\LQJZLWK&WUO(QWHU

1HWSUHVHQWYDOXH The NPV is based on the sequence of cash flows in column B.

([FHO )XQFWLRQ 139

(IGMWMSRW-RZSPZMRKXLI8MQI :EPYISJ1SRI]

D Given your other assumptions, how many weeks

F Graph profit as a function of the number of copiers

F Use your part Danswer to describe how the sales of

E If the exchange rate varies from its current value, does

The payback of a project is the number of years it takes

The file 3B[OV[ contains the cumulative

A project does not necessarily have a unique IRR.

A large U.S. drug company, Pharmco, has 100 million