This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Notes on Excel Forecasting Tools Data Table, Scenario
Manager, Goal Seek,
& Solver 2001-2002
1 Contents Overview ................................................................................................................1
Data Table ....................................................................................................2
Scenario Manager ........................................................................................8
Goal Seek ......................................................................................................11
Solver ............................................................................................................12 Appendix
Views of the worksheets in the ForecastingTools.xls workbook 2 Overview Data Table
Excel’s Data Table is a powerful sensitivity analysis tool that shows how
changing certain values in a model’s formulas might affect critical elements of
the model. Data tables provide a shortcut for generating multiple views for a
model in a single operation as well as a way to view and compare the results of
all of the variations together on a single worksheet. There are two varieties of
Data Table: one-input and two-input. To run a Data Table, establish the proper
Data Table layout and data and then use the commands Data, Table to open the
“Table” dialog. Using the prompts in the “Table” dialog, link the Data Table
input values to the model and click OK to run.
A scenario is a set of values that Excel saves and can substitute on command in a
worksheet model. You can create and save different groups of values on a
worksheet and then switch to any of these new scenarios to view different model
results. For example, if you create a budget worksheet but are uncertain what
revenue value to include, you can define different values for the revenue and
then switch between the scenarios to perform what-if analyses. To build
scenarios, choose Tools, Scenarios to open the “Scenario Manager” dialog. Follow
When you know the result you want from a single formula but not the input
value the formula needs to determine the result, use Excel’s Goal Seek. When
goal seeking, Excel varies the value in a worksheet cell you specify until the
formula that's dependent on that cell returns the result you want.
Excel’s Solver is a problem-solving tool like Goal Seek; however, the Solver
provides a much more powerful and flexible approach. Use Solver to determine
the maximum or minimum value of one cell by changing other cells— for
example, the maximum profit you can generate by changing advertising
expenditures. You specify one or more “changing cells” which must be related
through formulas on the worksheet. In addition, you can establish model
constraints, and Solver will search for a solution without violating the
constraints. Solver adjusts the values in the changing cells you specify to produce
the result you want from the formula. Choose Tools, Solver to open the “Solver
Parameters” dialog. Solver is an Excel add-in, but is part of Excel. If you don’t
find Solver on Excel’s Tools menu, add it with Tools, Add-Ins or return to your
Excel software media to add it as an option to your Excel installation. 1 Examples
The Data Table
The ForecastingTools.xls workbook contains an “Income Statement” worksheet that
shows a monthly income statement for Triangle Widgets, Inc. Obtain a copy of this
workbook if you want to follow along in Excel. The Income Statement worksheet. As you might expect, the income statement model uses formulas, not static values, for
many cells. For example, Revenue, Total Variable Expenses, and Total Fixed Expenses
are formulas. One can change data values in the model and see how those changes affect
Here are some examples of questions you might answer using this model:1
What happens to operating income if Triangle Widgets sells 2000, 2500, or 3000
units per month instead of 1000?
If the company sells 2000 units how do expenses change? What’s the impact on
expenses if 3000 units are sold? 4000?
What happens if Triangle Widget’s leasing costs go up by 20%? By 25% By 30%? As the model is currently constructed there are also some kinds of questions (that involve more
complex calculations and assumptions) that you can not answer. For example, in this model the
fixed expense value for “Advertising” is a static number. While changing the static advertising
value will affect the bottom line, the model can’t show how a change might affect “Units Sold”. So
you can’t use this model to answer the question “How would an increase in advertising affect
unit sales?” 1 2 What happens if manufacturing costs go down by an eighth? By a quarter? By a
Many questions of this kind can be quickly answered for a wide range of possible values
by using one of Excel’s powerful forecasting tools, the Data Table. Some of the
advantages of using a Data Table for this kind of task instead of just changing values in
the model itself are:
You can include any number of substitute (changing) values in the Data Table.
For example, with a Data Table it’s easy how a wide range of production level
values affects operating income instead of viewing the effect of each possible
production level in the model viewed one at a time.
The results are visible in a conveniently small matrix.
Data Tables can be a bit tricky to work with only because you must understand the Data
Table layout Excel requires and how to execute the Data Table properly. In addition to
these notes, review Introduction to Data Tables and Data Table Exercises. Excel’s online help
is also a good source for information about Data Tables.
Building a One input Data Table
We’ll locate a one input Data Table to the right of the income statement model on the
“Income Statement” worksheet. The single input to the Data Table will be “Units Sold”.
We locate these values in the first Data Table column, varying the “Units Sold” values
from 800 to 1500 in 100 increments. We include in the Data Table three formulas, one for
each of the other three columns in the table. These other columns will show the effect of
a change in “Units Sold” on the values for total revenue (Cell C7), “Total Expenses” (Cell
22), and “Operating Income” (Cell C23).
$29,500 In the completed Data Table above, by reading across the table you can see that when
Triangle Widgets sells 1400 units, total revenue is $133,000, “Total Expenses” are
$109,300, and “Operating Income” is $23,700. 3 The top cells each of the total revenue, “Total Expenses” and “Operating Income”
columns appear like the other values in the matrix. However, these first row cells are
different. Because of their position they’re actually part of the structure of the Data Table,
and not part of the Data Table results. They are formulas that either directly or indirectly
reference the Data Table input value (“Units Sold”) in the model. As the Data Table is
being built these formulas are either entered in the Data Table or references are made to
the formulas in the model (e.g., =C7, =C22, and =C23).
Working backward, before the Data Table is executed (before it has any data in its
matrix), its structure looks like this:
values $94,500 800
1,500 $500 Three
formulas The shaded area in the illustration above shows the area in which Excel puts the values
the Data Table generates. If you execute your Data Table once and then want to reexecute it2, you can do so by erasing the values in the shaded area first and then rerunning the Data Table commands.
To execute the Data Table and fill it with results, select the entire table (the empty cells,
the input values down the left-hand-column, and the formulas across the top) and
choose the Excel menu commands Data, Table. The “Table” dialog opens. This dialog
facilitates relating the Data Table input values to the model itself.
When prompted to supply the cell reference
from your model that the Data Table needs for
calculations, indicate the “Units Sold” cell
from the model (C5). This value should be
treated as column input; that is, enter its
reference in the text box with the label
“Column input cell”3. Since this is a one input
Data Table, there is no row input cell. The
“Row input cell” text box remains blank. Why might you want to re-execute a Data Table? Perhaps because you’ve changed or expanded
the inputs and want to see the new results or because you made an error in the Data
Table execution (usually choosing “Row input cell” when you should have chosen
“Column input cell” or vice versa) and need to correct it.
3 Column, because that’s how the “what-if” values for Units Sold are arranged in the One-Input
Data Table: Down the first column.
2 4 Click OK to start the Data Table processing.
When the Data Table is finished processing the Data Table matrix is filled with results
data. The completed Data Table shows a summary of eight by three “what-if” scenarios.
That is, for each one of the eight possibilities of “Units Sold” (800, 900, 1000, 1100, etc.)
you see the effect of those inputs in the three formulas ranged across the top of the Data
Table: revenue, “Total Expenses” and “Operating Income”.
The Data Table matrix makes it easy to see the results of many possible substitutions for
“Units Sold”. This is convenient if you want to compare possible options. It’s much more
convenient to use a Data Table than it would be to change the “Units Sold” value in the
model itself once for each of the eight scenarios you wanted to study and then had to
remember (or jot down) the way the other key values changed.
One input Data Table Summary
A One input Data Table varies one model value to see the effect on one or more other
values in the model. You can focus on the change to a single variable (such as revenue).
Or, you can show the changes to multiple variables (revenue, “Total Expenses”, and
“Operating Income”), as in this example. You can also make your one input Data Table
larger (or smaller) by extending (or reducing) the column of input values. Note: In the illustration above, the three cells holding formulas have been reformatted to
display text instead: Revenue, Total Expenses, and Operating Income. The
formulas remain in place, however. They’re part of the one input Data Table
structure. Building a Two input Data Table
As you might expect, a two input Data Table allows you to supply two input values and
vary them both. The two inputs must be variables used in a single formula. The output is
displayed just as it is for the one input Data Table. However, the layout of the two input
Data Table is, of course, a bit different. One input is located down the left-hand column
of the Data Table, as it is for the one input Data Table. But in the two input Data Table
the second input is located across the top row of the table. This arrangement of inputs
permits only a single formula, which is located in the upper-left-hand corner of the two
input Data Table. 5 Here, we again use the data in the “Income Statement” worksheet and construct a two
input Data Table to show the effect of varying “Units Sold” and “Price per Unit” values
on “Operating Income”.
In the Data Table we locate the range of “Units Sold” values down the left-hand column
and the range of “Price per Unit” values across the top row. For “Units Sold”, we begin
with a value of 800 and end with a value of 1500, incrementing by 100. For “Price per
Unit”, we begin with a value of $88 and end with a value of $98, incrementing by $2.
Execute the Two input Data Table the same way you did the one-input table but when
prompted supply the appropriate model cell references for both row and column input.
For example, when looking at the completed Data Table, at 1000 units sold the breakeven point is at a unit price of $96. The completed two input Data Table. Tip
In the illustration above, I’ve substituted
“Operating Income” as a label in place of 500
(that you see in the layout at right). However,
this cell refers to the formula for “Operating
Income” in the model. Having “Operating
Income” appear in the formula cell for the
two input Data Table is a nicety and some find
that it makes the Data Table easier to read (others
do not). 500 $
1100 88.00 $ 90.00 Here’s how to use this labeling technique:
1. Click the upper left-hand cell (the formula cell) of the two input Data
2. Right-click and choose Format Cells to open the “Format Cells” dialog. 6 3. In the “Format Cells” dialog choose the Custom option in the Category list. 4. In the Type list, choose General.
5. In the text box beneath Type, enter the label you want in the current cell.
Note that the label does not affect the actual contents of the cell. It affects
only what Excel displays to the viewer. In this case, enter “Operating
Income” (use double quotes) and click OK. The label you enter appears
in the current cell of the
spreadsheet (and is added to
the Custom types in your
copy of Excel, so you can reuse it if you care to). 7 Scenario Manager
For an example of the Scenario Manager we again return to the “Income Statement”
worksheet in the ForecastingTools.xls workbook. Above, we generated two kinds of
Data Tables to summarize how a change in one or two values affect other values in the
model. Here, we do something similar. However, instead of showing the changed values
in a tabular format we show a number of possible variations individually as different
worksheet scenarios. This method has the advantage of allowing the viewer to see the
changes within the worksheet itself and is often useful if you have a small number of
scenarios you want to work with or present.
We use Excel’s Scenario Manager to create three scenarios: Low Cost, Competitive, and
For each one of the three scenarios, we changes to these three values in the model:
“Units Sold”, “Price per Unit”, and “Material Cost per Unit” as indicated in the table
Material_Cost_per_Unit Competitive High Cost 1,400
$26 To generate the scenarios, use Tools, Scenarios to open the “Scenario Manager” dialog.
Click the Add button on that dialog to open the “Add Scenario” dialog, where we can
define the first scenario. Provide a name and indicate which
cells should be the designated as
the “changing cells”.
Click OK to go to the “Scenario
Values” dialog that allows you to
enter the values you want to apply
to these changing cells under the
“Low Cost” scenario. 8 Click the Add button to name another scenario. Provide values for it in the “Scenario
Values” dialog. Continue until you’ve named and defined all the scenarios you want for
When you’re through defining scenarios, the main “Scenario Manager” dialog lists each
scenario by name. To see the values for a scenario plugged
into the spreadsheet model, choose the
scenario name and click the Show button. You can add a new scenario at any time
or delete an existing scenario using the
Add or Delete buttons. Edit an existing
scenario using the Edit button. Choose the Summary button to have Excel
generate a scenario summary report. When you ask for a scenario summary, you have the chance to
identify what Excel terms “result cells”. These are cells in your
model affected by the changing scenario values. Usually,
they’re values critical to your model that you want to focus on.
When Excel generates the summary report, it shows you the
values these variables take on under each scenario.
In this example, we identify Units Sold (C5), Total Expenses
(C22), and Operating Income (C23) as the cells to track. 9 Excel puts the report on a new worksheet tab and names the tab “Scenario Summary”. If you’ve assigned any range names in your
worksheet for changing or result cells, the report uses
those names. Otherwise, the report uses cell
references. The report shows values for the three scenarios we created, plus a “Current Values”
report, which shows the values in the worksheet at the time the scenarios were created. Use the outline symbols to display just the changing cells, just the result cells, or both.
When you save your worksheet, the scenarios are saved automatically. 10 Goal Seek
When you know the result you want from a single formula but not the input value the
formula needs to determine that result, use Excel’s Goal Seek. When goal seeking, Excel
varies the value in the cell you specify until a formula that's dependent on that cell
returns the result you want.
To follow along in Excel, see the “Goal Seek” tab in the ForecastingTools.xls workbook.
For example, suppose you want the Payment value in the illustration below to equal
$900. Use Goal Seek and identify the interest rate in cell B3 as the value that Goal Seek
can change. Excel changes the interest rate incrementally until the payment value in B4
equals $900.00. To open the “Goal Seek” dialog, choose Tools, Goal
Seek from Excel’s menus. Identify your “set cell”
(must hold a formula) and the value you want this
formula to resolve to, and identify one cell that
affects the formula that Goal Seek can change.
Click OK. Excel provides a status dialog you can use
to preview its solution. Keep the solution
by clicking OK, or discard the solution by
clicking Cancel. If you keep the solution, the
changed set cell (payment value)
and changing cell (interest rate)
display in the worksheet. 11 Solver
Excel’s Solver is a powerful tool for forecasting and what-if analysis. For an example of
Solver we use the data on the “Optimize” worksheet in the Forecasting Tools.xls
workbook. Start by taking a look at that worksheet. T riangle Widge ts, Inc.
October Production Schedule
Tota l Profit: $ 3,150.00 Mode ls
Re d W i dge t Bl ue W i dge t Gre e n W i dge t Ye ll ow W idge t Ora nge W i dge t
Profit pe r Unit
Production Le ve l
10 Widge t
Com pone nts
G - W idge t Com pone nts Re quire d for Ea ch Mode l 5
0 Pa rts on
200 Pa rts
Re quire d
200 The worksheet shows the five different widget models that Triangle Widgets makes
(Red, Blue, etc.) along with the profit per unit for each model, and the production levels
for October. A sumproduct formula near the top of the worksheet calculates the Total
Profit that will be generated by the current model’s Production Level.
As written, however, there’s a problem with the values in this model. In the section of
the worksheet that details widget components are two columns that track the number of
widget parts required for the current level of production (“Parts Required”) and the
number of widget parts on hand (“Parts on Hand”). At the production levels currently
in the model -- 10 widgets of each type -- the company has exceeded parts-on-hand
quantities for components B (by 30) and E (by 40). So the Production Level values in the
model show an impossible scenario; we can’t achieve $3,150.00 in Total Profit.
To change the Production Level values to get the highest possible Total Profit without
violating any of the Parts on Hand and Parts Required constraints, one could change the
five Production Level values in the model manually. This would require experimenting
with different combinations until a combination that maximizes profit is found, while
keeping within the constraints. Unfortunately, to be sure to maximize profits this
experimental, manual process could take quite a bit of time. In a large problem, a
manual method might prove to be simply impractical.
As an alternative to a manual method, Excel’s Solver tool can maximize total profit in
this model, quickly finding the best solution. 12 Solver is an Excel add-in that’s part of the Excel program. That is, when you purchase
Excel, you get Solver automatically (whether or not it’s actually installed). If installed,
Solver appears on Excel’s Tools menu. If you don’t see a Solver entry on your home PC’s
Excel Tools menu, you may need to do one of two things: 1) Use Tools, Add-Ins to open
Excel’s “Add-Ins” dialog box and add a check to the left of the “Solver Add-in” entry, or
if you don’t see the Solver Add-in as an option in that dialog, 2) Return to your MS
Office installation CD and the Excel installation and request Solver as an installed
option. To invoke Solver, choose Tools, Solver. The “Solver Parameters” dialog opens. There are three main segments in this dialog box. They define the “set” cell, the
changing cells, and any constraints.
The “Set” Cell or Target
Begin with the “set” cell or target cell. In our example, the target is Total Profit, in Cell
B4. We want to maximize total profit, so click the “Max” option. Notice that you also
have the choice to minimize or to find a particular value (like Goal Seek).
The Changing Cells
Second, identify the cells that Solver can change to find its solution. The changing cells
should be related to the target cell, of course. In our example, the changing cells are the
number of widgets to make, or the Production Level values. Your changing cells need
not be contiguous in the worksheet. You can have an unlimited number of changing
Finally, identify any constraints Solver should take into account when finding a solution.
Although it appears that you have a text box in which to enter constraints, that’s not the
case. To enter a constraint, click inside the “Subject to the Constraints” box and click the
Add button. Solver then presents you with a special mini-dialog in which to enter the 13 constraint in the correct format. For example, we must specify several constraints for this
The Production Level values
must be greater than or
equal to zero. E.g., not
negative production. The Parts Required range
values are less than or equal
to the Parts on Hand range. The Production Level values
must be integers. No
fractional widgets. With these constraints in place, the “Solver Parameters” dialog looks like this: Click the Solve button to start the solving process. 14 While Solver may not be able to find a solution for every problem, in most cases, Solver
finds a solution and returns with a dialog that looks like the below: Choose Keep Solver Solution and click OK here to keep the Solver solution in the
worksheet, or Restore Original Values to discard the Solver solution. Note that you can
choose to generate one or more of Solver’s reports (Answer, Sensitivity, Limits) and/or
to apply the Scenario Manager to record the Solver’s solution.
Our solved model looks like this: Solver found that we could achieve maximum total profit by eliminating the red and
blue widgets from the production schedule and adjusting the production levels of the
three other widget types considerably.
Note that Solver respected the constraints we defined:
No production level is less than zero.
The values in the Parts Required column are less than or equal to the Parts on
All the production level recommendations are for integer quantities. 15 A ppendix
Views of the worksheets in the accompanying ForecastingTools.xls workbook. Data Tables: the Income Statement Tab The Scenario Report: The Scenario Summary Tab 16 The Goal Seek Example Problem: The Goal Seek Tab The Model with Solver’s Solution Applied: The “Optimize” Tab 17 ...
View Full Document
- Spring '01