Hotel Pricing Problem

Author
Crystal Ball

Summary
A downtown hotel is considering a major remodeling effort and needs to determine the best combination of rates and
room sizes to maximize revenues. Currently the hotel has 450 rooms that can be segmented into three categories,
each of which has a different price, daily occupancy rate, and price/demand elasticity. Given uncertainty around the
elasticity, the owners want to set the optimal prices for each room type while maximizing their projected revenue.

Keywords: hotel design, pricing, demand, elasticity, planning, revenue, deterministic optimization, stochastic
optimization

Discussion
Currently the hotel has 450 rooms with the following history:

Room Type Rate Daily Avg No. Sold Revenue

Standard $85 250 $21,250
Gold $98 100 $9,800
Platinum $139 50 $6,950

Each market segment has its own price/demand elasticity. Elasticity estimates are:

Standard = -3
Gold = -1
Platinum = -2
This means, for example, that a 1% decrease in the price of a standard room will increase the number of rooms sold
by 3%. Similarly, a 1% increase in the price will decrease the number of rooms sold by 3%. For any proposed set of
prices, the projected number of rooms of a given type sold can be found using the formula:

rooms sold = H + ((E * H * (N - C)) / C)

where:

H = Historical average number of rooms sold.

E = Elasticity
N = New price
C = Current price

The hotel owners want to keep the price of a standard room between $70 and $90, a gold room between $90 and
$110, and a platinum room between $120 and $149. All prices are in whole dollar increments (discrete). Although the
rooms may be renovated and reconfigured, there are no plans to expand beyond the current 450-room capacity.
Using Crystal Ball
Crystal Ball enhances your Excel model by allowing you to create probability distributions that describe the uncertainty
surrounding specific input variables. This model includes three probability distributions, referred to in Crystal Ball as
"assumptions." The three assumptions describe the variation in the elasticity for each room type. Each assumption cell
is colored green and is marked by an Excel note (mouse over the cell to view the note). To view the details of an
assumption, highlight the cell and either select Define Assumption from the Define menu or click on the Define
Assumption button on the Crystal Ball toolbar.

This model also includes two Crystal Ball forecasts, shown in light blue. Forecasts are equations, or outputs, that you
want to analyze after a simulation. During a simulation, Crystal Ball saves the values in the forecast cells and displays
them in a forecast chart, which is a histogram of the simulated values. In this example, you want to analyze the
projected rooms sold and the projected revenue. To view a forecast with Crystal Ball, highlight the cell and either select
Define Forecast from the Define menu or click on the Define Forecast button on the toolbar.

When you run a simulation, Crystal Ball generates a random number for each assumption (based on how the
assumption has been defined) and places that new value in the cell. Excel then recalculates the model. You can test
this by selecting Single Step from the Run menu or clicking on the Single Step button on the toolbar.

Notice that, when the room rate and the new room price are identical, the elasticity has no effect on the model results.
To view the effects of elasticity, increase the new room prices by $1 and run a simulation. When the simulation is
complete, you will see the two forecast charts. What is the mean total room demand? What is the mean total revenue?
Does room capacity ever exceed demand at these new room prices? What happens with when new room prices are all
$5 lower?

To view which of the assumptions had the greatest impact on a particular forecast, use a sensitivity chart. Which of the
three elasticities most affects the demand? Is this the same for the revenue? (Hint: You can change the target forecast
by clicking on the Chart Prefs button.)

Using OptQuest

Now that you have run at least one simulation, you can begin to address optimization using OptQuest. In this model,
you want to set the optimal prices for each room type while maximizing their projected revenue.

OptQuest requires decision variables, which are model variables over which you have control. The three decision
variables defined in this model are the New Price for each type of room. Each decision variable is colored yellow and is
marked by an Excel note (mouse over the cell to view the note). To view the details of a decision variable, highlight the
cell and either select Define Decision from the Define menu or click on the Define Decision button on the Crystal Ball
toolbar. Notice that the bounds in these three variables correspond with the rates described in the final paragraph of
the discussion above.
Start OptQuest from the Run menu and use the OptQuest Wizard to view the settings for the optimization.

This problem has no constraints, one requirement (on the room demand), and one objective: to maximize total
revenue. To ensure that the probability of demand exceeding capacity does not exceed 20%, you can set a
requirement on the total room demand forecast. Specifically, the total room demand is limited by a requirement using
the forecast statistic Percentile (80), with an upper bound of 450.

Run the optimization. For each optimization, OptQuest selects a new value within the defined range of each decision
variable (for example, a new room price) and runs a Crystal Ball simulation (for example, 2000 trials). OptQuest then
saves the mean Total Revenue value and checks to see if the requirement is satisfied. OptQuest then runs another
simulation on a new decision variable value. OptQuest repeats this process, constantly searching for the best total
revenue that also agrees with your requirement.

As OptQuest runs, it uses multiple metaheuristic methods and techniques to analyze past results and improve the
quality and speed of its process. You can watch OptQuest's progress through the Performance Graph, which shows a
flattened line as it converges to an optimal result.
What is the best combination of room prices that results in the highest mean total revenue? Once OptQuest is finished,
you can copy the optimal results back to your spreadsheet through the Copy Best Solution to Spreadsheet option in
the Edit menu. Your spreadsheet now displays the optimal solution, and Crystal Ball displays the forecast chart for the
simulation from the best optimization. You can use OptQuest's Solution Analysis tool to review the other quantities of
products that resulted in high Total Revenue values.

You can also run a non-stochastic optimization of this model by using the Deterministic option in the Advanced tab of
the Options dialog in OptQuest. A deterministic optimization freezes the Crystal Ball assumptions at their original, or
base case, values. Compare the results of the deterministic and stochastic optimizations. What are the effects of
elasticity?

Copyright Information
Copyright 2004, 2010, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Other names
may be trademarks of their respective owners.
 Model

A B C D E F G H I J
1 Hotel Pricing Problem Determine best price
2 of hotel rooms
3
4 Average Projected
5 Daily New Rooms Projected
6 Room type Rate Sold Revenue Elasticity Price Sold Revenue
7 Standard $ 32.00 20 $ 640.00 -3 $ 159.00 (218) $ (34,681.88)
8 Gold $ 42.00 24 $ 1,008.00 -1 $ 239.00 (89) $ (21,168.57)
9 Platinum $ 156.00 50 $ 7,800.00 -2 $ 134.00 64 $ 8,589.74
10
11
12 Total (243) $ (47,260.70)
13 Maximize revenue
subject to capacity Capacity 450
14

Page 4