Revenue Management Takes Its Next Step - Case Study Example

Forecasting Report Incorporating Computer Analysis Based on a Case Study Provided Name Affiliation Table of content Table of content 2 Executive Summary 2 Introduction 3 The Aim and Purpose of the study 4 Background 4 Limitation of the study 4 Forecasting procedures 5 Forecast methods 5 DATA ANALYSIS 7 The demands forecasting in hotel 9 Short-term forecast 9 Long-term forecast 10 Using pickup ratio forecast 10 THE APPROPRIATE FORECASTING METHOD 11 The forecast for Saturday 11 CONCLUSION 13 References 14 Executive Summary In this report the forecasting procedure of Hamilton hotel will be the most important thing to take note of. This will be guided through the report by also finding the appropriate forecasting method to be used in order that the Marriot Group of company to be able to maximize the revenue allocation. Over the years the company had been facing problems in the revenue management in the hotels and resorts all over its territorial regions in the globe. The report seeks to find the best procedural and computerized system that can be used in the forecasting so as to be able to manage the large database used in the company since the manual analysis on data is the biggest problem in the Marriott Hotels. I am going to use both long term and short term methods of forecasting in this report. The forecast concentrated with the booking and demands of the hotels and the resorts. Introduction Marriott Corporation manages Hamilton hotel which has 180 restaurants and hotels around the world. It’s a business hotel with meeting space for conservations in the city suburbs and also for group activities. The hotel has also 1877 rooms that can be used for bookings for tourists. In order to meet to the targets of the Hotel, the corporation has expanded the hotels all over the globe even to unexpected areas in Africa; in Kenya Marriott group of hotels is one of the best top class hotels. Over the years the company has been undertaking forecasting reports which has been helping the company to schedule the work assignment to the staff, helps the company in procurement duties such as purchasing and financial analysis, revenue collection and disbursement, planning and determination of the hotel costs and at most generating and consolidation of the house book keeping. All the above statistical information obtained from the forecast has used by the management to revenue collected from the hotel branches, the occupancy rate of the rooms in the resorts and hotels, and as a measuring kit for the awarding of the best mangers in the company. The Aim and Purpose of the study The aim of this report is to make all days demand forecast using Tuesday as the determining day by also using the best forecasting procedures. This will be done by determining days demand for Saturday form Friday. The best methods to be used for the forecasting by using the employment level as the benchmark is also very important and is a component in this study. With the use of this forecast the managers of the hotels will be able to make proper decisions. Background Being the manager of the hotel (Hamilton Hotel reservation) I will advice Snow to adjust some of the situations that led had an influence to the since most of the clients did not show up on weekends, some also cancelled the reservations before their arrivals and also to those that stayed in the hotel extra days sine this and an impact in the employee’s works and the revenue departments. The managers through the use of better forecasting procedures and methods will be able to run the hotels and resorts effectively in order to attain reputable revenue collections and to have good relationship with the employees since without them the company cannot attain anything at the end of the financial period. Limitation of the study The limitation of this study was that the data available were only primarily provided to the company using mails and there was no personal visitation to the hotels to get the data. The manager at the hotel was now not able to give the correct overlaying booking and demand forecast of other hotels in the globe since the managers to this hotels avoided or gave wrong data this leads to a fluctuations in the pick up ratios that will be used to determine weeks 14 forecast. Since the data was collected within a short life span of four months the data was collected using Saturday as the master day the study only used short term data for the display on the booking and demand trends. The data collection within the few days could note give the exert forecast for the company. Forecasting procedures Another name for revenue management is yield management. This is a process whereby we use historical data as well as current reservations to determine the demands for the future accurately in order to maximize revenue. Is the process of applying records of historical data and current reservations to predict future demand as accurately as possible to maximize revenue? There are two forecasting procedures to be used in this study that is statistical forecasting and judgemental forecasting. When using the statistical for casting in the case of Hamilton Hotel there is the use of the computer system for the mechanical support the managers to use the databases more effectively in maximization of the revenue from the company business. While using the judgmental forecasting, the computer system helps in the analysis of the decision support so as to encourage the management to more effective decision making. Forecast methods With reference to this report, there is going to be the use of four methods of forecasting used to analyse and compute the demands of the hotel from May 23rd to August 22nd, the year 1987. Simple exponential smoothing method is the first method that uses the following constant value α between 0.01 and 0.3. This forecasting method is mostly used for forecasting techniques in the hotel lodging industry. This method is useful when the data is of horizontal pattern. The formula used to calculate the exponential smoothing is as below; Ft=At-1+(1-)Ft-1 Whereby Ft is the forecast period At-1 this is the actual value of the time series Ft-1 this is the forecast from the prior period is the smoothing constant The value of the will determine the smoothing degree and shows how the responsive model is responding to the fluctuation of time series data. The value of the arbitrary and hence determined by the feelings of the forecaster and the nature of the data. The second method is the Moving average method whereby the period average is 5 days that ranges from day 3 to day 7. The simple forecasting method uses the last values of the time series that provides the forecast for the given data up to a time t. the more the observations included in the moving average the greater the value of the smoothing effect. There forecaster must choose the value of the periods (k+1) in the moving average. A linear regression method was also used that used the day of the week and the final demands as the variables, and the last is the series decomposition method that broke constituents and later reassembled parts to construct the forecast. This method was favourable used for the analysis of the revenue allocations from the resorts and hotels. DATA ANALYSIS The increase of data that is predicted in this study on the demand and the booking data obtained from the historical data hand an increase to about 1900. This was so as portrayed on Saturday whereby it was recorded that the reservations went higher up to 2000 bookings. Using the regression analysis to compute this data there was an evidence of increased revenue but low level of employee’s moral attributed to less showups of the clients. All through the weeks there was low demand on Thursday and Saturday of the week it ranged from 700 to 870. As shown in figure 1 below there was similarities between the demands and the bookings all through the 13 weeks. ORIGINAL DATA CHART Every day of the week was as important as the customers who came to the hotels and the resorts since it was the pivot for the forecast. The trend of the pick up ratios is as follows. The overall ratio for the pick is as follows 1.476 to 1.441 on Thursday and 0.846 to 0.977 on Saturday. The differences between demands and bookings are larger, the pickup ratio will higher or lower than 1. The pick up ratio was very important in the decision making for the managers and the management of the company. Figure 2 The same information can be represented using the scatter diagram to plot the same data and add a line of best fit (trend line) to it. Figure 3, 4, 5 The demands forecasting in hotel There are major forecast methods that can be used to determine the demands of the 14th week these are moving average method, simple exponential smoothing method, linear regression method, and time-series decomposition method. Short-term forecast The data was given on a weekly basis hence led to the selection of the short term method to determine the demand of the 14th week since the weeks is the unit for the forecast. When using this method the average of the pick up ratio is important. Its going to use the 5 day moving average as shown the excel sheet. There is the use of MAD to calculate the errors. There are three smoothing constants are set to compute the forecast: α=0.1, 0.2 and 0.3 as shown in the Excel sheet. We can find out that the error of smoothing constant of α=0.3 is the smallest among these three constants is 0.124 (0.130 for α=0.1 and 0.127 for α=0.2 and 0.124 for α=0.3). Long-term forecast The forecast for the demands data of the week 14 are forecasted using time-series decomposition method and regression forecast method. Using pickup ratio forecast We first determine the seasonal index of the original pickup ratio by finding the average pickup ratio that is done by determining the mean of the data period then multiply the data by 100/the index of the week. The above data is a form of form a linear regression. After calculating the regression data we have to reseasonalisation the information so as to shape the is needed to shape the future pattern. The results are shown in Figure 8 as below. In computing the error in between the regression and reseasonalisation Deseasonlised data is used to form a linear regression, in order to more accurately forecast the demands of the 14th week. After computing the regression data, reseasonalisation is needed to shape the future pattern. The results are shown in Figure 11 as below. In computing the error in between the regression and reseasonalisation it was found that the error was regression is 268, and error of reseasonalisation is 239 THE APPROPRIATE FORECASTING METHOD We can easily find the smaller errors in the computation of the forecast by comparing the values of the bookings to those of the demands using the simple moving average method. The comparison between the figures of the Tuesday bookings and the demands provides the best error results. Therefore, I recommend that the best forecast method is the simple moving average method of time-series decomposition method used pickup ratio data to make the 14th week demands and booking forecast. The forecast for Saturday According to the forecast method of time-series decomposition to compute the data of demand on Saturday in the 14th week is 1662, the data of bookings is 1839 and pickup ratio is 0.904(pickup ratio = demands/ bookings). The manager of hotel’s reservation Snow can accept the extra 60 rooms’ request. Although the bookings number on Saturday is 1839 which is close to 96% of room occupancy rate, to consider some reasons of customer no-show reservation, stay in hotel with extra period and check out rooms in advance, the hotel still have capacity to adopt request for up to 60 rooms. Determination of the Employment Levels Using the Pick Up Ratio The employment level in the hotel is related to the pickup ratio and should follow the trend of pickup ratio, since the extent at which the bookings and the demands determine the assignments that is to be given to the other areas within the company. Some of the departments that are closely affected by changes in demands and bookings are staff restaurant and the staff desk. To be able to organize the trend of pickup ratio, I will use the simple moving average method since the data is of historical nature to forecast week 14. Figure 14 CONCLUSION In order to have a good and human resource with the best working environment in the hotel and to increase the revenue allocation for the company I propose the use of the simple moving average method of data forecasting that is very simple to be used using the historical data provided. The use of the following two forecasting procedures statistical forecasting and judgemental forecasting. When using the statistical for casting in the case of Hamilton Hotel there is the use of the computer system for the mechanical support the managers to use the databases more effectively in maximization of the revenue from the company business. While using the judgmental forecasting, the computer system helps in the analysis of the decision support so as to encourage the management to more effective decision making. This will improve the productivity of the company to. References Albright, B. 2008. Revenue management takes its next step. Hotel & Motel Management, 223(12), 42-42. Retrieved from http://ezproxy.library.unlv.edu/login?url=http://search.ebscohost.com/login.aspx rue&db=buh&AN=32968151&site=bsi-live Andreassen, P. B.,1990. Judgmental extrapolation and the salience of change. Journal of Forecasting, 9(4), 347-372. doi:10.1002/for.3980090405 Armstrong, J. S. (Ed.). 2001. Principles of forecasting: A handbook for researchers andpractitioners. New York, NY: Springer Science+Business Media, Inc. Retrieved fromhttp://books.google.com/books?id=XdE4m_xMfL8C&lpg=PR5&ots=QnHf Read More