لخّصلي

RM goals, data and information gathering, analysis RM process starts with the goals setting by the revenue manager with specifi c strategic (several years), tactical (weeks/months) and operational (days) time horizon (Ivanov & Zhechev, 2011, p. 304). On the other hand, they rely on the fact that knowing how certain variable has changed over time (e.g. what was the occupancy of the hotel during the last couple of months) can provide information on how this variable will change in future, i.e. as if the variable has memory, similarly to technical analysis in fi nancial markets forecasting.Review of available literature on hotel RM reveals that most papers deal with 2 main topics: forecasting demand (e.g. Frechtling, 2001; Lim & Chan, 2011; Song, Witt & Li, 2009) and forecasting RM metrics and operational data (El Gayar et al., 2011; Haensel & Koole, 2011; Morales & Wang, 2010; Zakhary, Atiya, El-Shishiny & El Gayarm 2011).Mostly used (or analysed) by researchers historical methods are: moving average (Burger et al., 2001; Weatherford & Kimes, 2003; Yuksel, 2007), exponential smoothing (Burger et al., 2001; Chen & Kachani, 2007; Rajopadhye, Ghalia, Wang, Baker & Eister, 2001; Weatherford & Kimes, 2003; Yuksel, 2007) and other autoregressive models (Burger et al., 2001; Lim & Chan, 2011; Lim, Chang & McAleer, 2009; Yuksel, 2007).Th erefore, a high proportion of the research literature is dedicated to forecasting from theoretical and methodological perspective (Burger, Dohnal, Kathrada & Law, 2001; Frechtling, 2001; Tranter et al., 2008; Weatherford, Kimes & Scott, 2001; Weatherford & Kimes, 2003, among others), summarized in Table 4.Th is is justifi ed since volume, structure and characteristics of demand and forecasts for occupancy rate, number of arrivals, cancellations, no shows, RevPAR, ADR and other operational statistics are of utmost importance to hotel's RM system.Weatherford and Kimes (2003) divide the methods to historical, advanced booking and combined methods.

RM goals, data and information gathering, analysis
RM process starts with the goals setting by the revenue manager with specifi c strategic (several years),
tactical (weeks/months) and operational (days) time horizon (Ivanov & Zhechev, 2011, p. 304). Th ey
relate to the values of the diff erent RM metrics discussed above (RevPAR, ADR, occupancy, target
profi t per available room). Th e RM software gathers the necessary operational data and information
provided by the hotel’s marketing information system. Th e operational data is analyzed to provide the
revenue manager with clues about the trends in hotel’s RM metrics for the forthcoming days/weeks.
Th e third stage also involves analysis of demand (on the level of individual hotel, chain properties in
the destination and on destination level) and the supply in the destination (opening/closing/refl agging
of properties).
Forecasting
Forecasting involves the application of diff erent forecasting methods in order to provide the revenue
manager with prognoses about the future development of RM metrics, demand and supply. Successful
application of revenue management requires hotels being able to forecast demand. Th erefore, a high
proportion of the research literature is dedicated to forecasting from theoretical and methodological
perspective (Burger, Dohnal, Kathrada & Law, 2001; Frechtling, 2001; Tranter et al., 2008; Weatherford, Kimes & Scott, 2001; Weatherford & Kimes, 2003, among others), summarized in Table 4.
Review of available literature on hotel RM reveals that most papers deal with 2 main topics: forecasting demand (e.g. Frechtling, 2001; Lim & Chan, 2011; Song, Witt & Li, 2009) and forecasting
RM metrics and operational data (El Gayar et al., 2011; Haensel & Koole, 2011; Morales & Wang,
2010; Zakhary, Atiya, El-Shishiny & El Gayarm 2011). Th is is justifi ed since volume, structure and
characteristics of demand and forecasts for occupancy rate, number of arrivals, cancellations, no shows,
RevPAR, ADR and other operational statistics are of utmost importance to hotel’s RM system. However,
RM decisions in a particular hotel experience the infl uence of its competitors’ decisions and actions
and developments in the external environment. In this regard it is surprising that a limited number of
papers, most notably Yüksel (2007), discuss issues related to forecasting competitive actions and the
external environment which remains a neglected fi eld.
Proper forecasting procedure requires the application of suitable forecasting methods. Weatherford and
Kimes (2003) divide the methods to historical, advanced booking and combined methods. Mostly used
(or analysed) by researchers historical methods are: moving average (Burger et al., 2001; Weatherford
& Kimes, 2003; Yüksel, 2007), exponential smoothing (Burger et al., 2001; Chen & Kachani, 2007;
Rajopadhye, Ghalia, Wang, Baker & Eister, 2001; Weatherford & Kimes, 2003; Yüksel, 2007) and
other autoregressive models (Burger et al., 2001; Lim & Chan, 2011; Lim, Chang & McAleer, 2009;
Yüksel, 2007). It is evident that historical methods are based on time series analysis. Th eir advantage is
the relatively easy application and low data requirements. On the other hand, they rely on the fact that
knowing how certain variable has changed over time (e.g. what was the occupancy of the hotel during
the last couple of months) can provide information on how this variable will change in future, i.e. as
if the variable has memory, similarly to technical analysis in fi nancial markets forecasting. Th is is the
main disadvantage of time series forecasting – they disregard other variables – demand, competitors’
TOURISM 187 Review
Stanislav Ivanov / Vladimir Zhechev
Vol. 60/ No. 2/ 2012/ 175 -197
actions or special events in the destinations that stimulate demand. However, albeit their shortcomings
time series methods remain widely used.