Professional Documents
Culture Documents
Demand Forecasting in The Hospitality Industry: HBM491 Operational Analysis
Demand Forecasting in The Hospitality Industry: HBM491 Operational Analysis
• Objectives
• Forecasting Methodology
• Exercises
• Q&A
• Describe the nature and limitations of forecasting and identify the kinds of
patterns that emerge from historical data of hospitality operations.
• Explanation of why collecting and analyzing data about customer demand for lodging products
and services are essential for forecasting sales
• Describe and apply various quantitative forecasting methods and explain how they differ from
qualitative forecasting methods.
• Calculate a forecast using a moving average, weighted moving average, and exponential
smoothing
• Describe the purpose of, and methods used to create, short-term forecasts in
the lodging industry.
Imagine the same situation on other types of the industry i.e. airlines,
hospitals, catering etc.
Demand Forecasting in Hospitality Industry , 9 of 54
A Revenue Management Cycle
There are many components to implementing a
successful revenue management culture
and process in a hotel. Forecasting demand
Optimizing demand
Controlling demand
Monitoring demand
200
150
100 Predicted
demand
50
looking back
Time
Jan Feb Mar Apr May Jun Jul Aug six months
• In revenue management, your goal is to work with demand and the resources
you have to maximize your revenue. To do this effectively, you need to
understand the subtleties of constrained and unconstrained demand and why
accurate forecasting is a necessary part of any revenue management system.
Constrained Demand
• The total of all realized demand, for instance, for rooms at a particular hotel.
Constrained demand is understood to reflect the limitations of inventory and
the influence of availability controls.
Decision on Prices
The forecast
reservations-on-hand
(ROH)
S
t
e
p
M
o
n
i
t
o
Demand Forecasting in Hospitality Industry , 22 of 54
Forecasting in the Hospitality Industry
Hotel Revenue Managers forecast mainly Room
Demand for a future period of time measured Demand (customer)
either in: The number of potential
buyers with the interest
• Number of Rooms and ability to purchase the
• Number of Room Nights products sold by a business
at the specific price offered.
• Number of Guest Nights
• Occupancy % achieved
• By the property
• By room type
• Average number of guests per room
• Average length of guest stay
Average Last 8 Mon Last 8 Tue Last 8 Wed Last 8 Thu Last 8 Fri Last 8 Sat Last 8 Sun
Occ % 88% 91% 78% 67% 45% 51% 29%
ADR $158.75 $188.75 $148.75 $138.75 $155.75 $159.75 $129.75
RevPAR $139.70 $171.76 $116.03 $92.96 $70.09 $81.47 $37.63
1 $1,350
2 $1,320 $1,320
3 $1,390 $1,390 $1,390
4 $1,440 $1,440 $1,440 $1,440
5 $1,420 $1,420 $1,420 $1,420 $1,420
6 $1,458 $1,458 $1,458 $1,458 $1,458 $1,458
7 $1,450 $1,450 $1,450 $1,450 $1,450 $1,450 $1,450
8 $1,460 $1,460 $1,460 $1,460 $1,460 $1,460 $1,460
9 $1,410 $1,410 $1,410 $1,410 $1,410 $1,410
10 $1,440 $1,440 $1,440 $1,440 $1,440
11 $1,470 $1,470 $1,470 $1,470
12 $1,460 $1,460 $1,460
13 $1,418 $1,418
14 $1,494
Total $9,828 $9,938 $10,028 $10,078 $10,108 $10,148.00 $10,108.00 $10,152.00
7-Day $1,404.00 $1,419.71 $1,432.57 $1,439.71 $1,444.00 $1,449.71 $1,444.00 $1,450.29
Rolling
Average
1. Know about special citywide or area-wide events in the area that affect demand
4. Adjust demand forecasts quickly when faced with significant demand- altering events
(i.e., unusual events, inclement weather, power outages and airport or highway
closings)
5. Adjust demand forecasts quickly when faced with significant demand- altering events
(i.e., unusual events, inclement weather, power outages and airport or highway
closings)
Qualitative Forecasting
Qualitative methods emphasize human judgment and may be useful when the data needed for
quantitative methods is not available.
• Time series approaches assume that • Causal approaches assume that the
a pattern recurs over time; pattern value of a certain variable is a
may be used to forecast. function of other variables.
–Advantages:
• Simple to use
• Virtually no cost
• Quick and easy to prepare
• Data analysis is nonexistent –Limitations:
• Easily understandable Do not consider potential effects of
• Can provide accuracy decisions (such as pricing and advertising)
• Can be a standard for accuracy made for future periods;
• must take seasonality into account;
• cannot provide high accuracy