Academic journal article American Journal of Entrepreneurship

Building Forecasting Models for Restaurant Owners and Managers: A Case Study

Academic journal article American Journal of Entrepreneurship

Building Forecasting Models for Restaurant Owners and Managers: A Case Study

Article excerpt

Introduction

According to the literature on the restaurant industry business cycles, the US demonstrated three cycles (peak to peak, or trough to trough) for the period of 1970 through 1998. The restaurant industry peaked in 1973, 1979 and 1989, and troughs were at 1970, 1974, 1980 and 1991 Choi et. al. (1999). The mean duration of this industry cycle, calculated either by peak to peak or trough to trough, is 7.3 years. This means the next peak would be 1996 or 1997, followed by 2003 or 2004 and finally 2010 or 2011 if the past trends take the same track. Other related service industries experience similar trends. For example, the hotel industry declined sharply after it reached the peaks. The mean duration for the contraction is about two years (1.7 years) and the mean duration for the expansion is about six years (5.7 years). The hotel industry led the general business cycle peaks by about 0.75 years on average and also led at troughs in the general business cycle by roughly 0.5 years (Jeong-Gil Choi et al., 1999). Given the nature of business cycles, an owner of a small restaurant in a small town would need good forecasts to plan for the changes in demand.

The importance of accurate and timely forecasts of sales is apparent at all levels for restaurant operations. Short-term sales levels are important for daily and weekly employee scheduling, especially where restaurants are dependent on part-time labor. Restaurants deal with very perishable products, and therefore, purchasing and inventory need to be accurately estimated. From a long-term perspective, menu development, employee hiring and training, and capital investment decisions (e.g. equipment, seating capacity and expansion) are directly linked to accurate forecasts. Accurate sales projections also impact the effectiveness and efficiency of marketing strategies and advertising.

Historically, forecasting of restaurant sales in the hospitality industry has been based on the manager or owner's judgment (D.A. Cranage & W.P. Andrew, 1992). The mismatch between the importance of forecasting and the lack of the use of quantitative forecasting methods can be attributed to the fact that the majority of the restaurants are owned by independent restaurant owners who tend to lack the resources for the development and application of more quantitative and accurate forecasting models. Since the trade-off between forecast accuracy and the cost of obtaining accurate forecasts is important for any restaurant owner, we propose different forecasting models for an independently owned bistro in Florence, South Carolina using data from 2006 - 2008. We implement the forecasting models in a spreadsheet package (Microsoft Excel) that is commonly available. We analyze the accuracy of these models using common error terms like Mean Square Error and Mean Absolute Deviation and show that restaurant owners can use these simple models to develop their own forecasts without the need for any external macroeconomic data. Thus, restaurant owners can simply use their own data they collect to make accurate, timely forecasts.

The next section is a brief literature review on the restaurant industry and trends in forecasting methods. Next, we discuss the data and forecasting methods. We then put forth our results, and the last section contains the conclusion and implications of this work.

Literature Review

Most restaurants in the U.S. are owned and operated by independent restaurant owners. Of the 235,701 full-service restaurants operating in the U.S. in 2013, 66% have fewer than twenty employees, and 88 percent of have fewer than fifty employees (US Census Bureau, 2013). In fact, according to the National Restaurant Association, 7 out of 10 restaurants are single-unit operations (National Restaurant Association, 2015). These smaller businesses typically have few resources to dedicate to statistical analysis when compared to their larger counterparts. However, if properly implemented, they could see similar benefits in staffing, inventory control and capital investment by generating accurate forecasts of their sales. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.