Public Release: 

Professors' model outperforms movie screen exhibitors at box office

Institute for Operations Research and the Management Sciences

LINTHICUM, MD, January 4, 2000 - Three professors' mathematical model for choosing films can do between 30% and 120% better at the box office than movie exhibitors' own choices, according to a study in a journal of the Institute for Operations Research and the Management Sciences (INFORMS®).

Although movies are a glamour industry, the authors based their model on research into scheduling parallel machines at industrial plants. "Our model leads to an improvement in profitability for the exhibitor under a broad range of parameter estimates, cost assumptions, contract terms, and decision-making structures," the authors write.

The study, "SilverScreener: A Modeling Approach to Movie Screens Management," appears in the current issue of Marketing Science, an INFORMS publication. The authors are Jehoshua Eliashberg, Wharton Business School, Sanjeev Swami, Indian Institute of Technology, and Charles B. Weinberg, University of British Columbia. The authors use operations research and other quantitative techniques to reach their conclusions.

Much of the authors' research can be extended to retailing operations that share characteristics with movie theaters. These include books, video games, and fashion good industries.

Two-Part Model

The improvement in profitability over actual decisions apparently results from a combination of better selection and scheduling of movies. Picking movies and deciding how long to run them at the theater are the areas of concentration in the model.

"In brief, our findings show that managers of movie theaters should choose fewer 'right' movies and run them longer," explains Prof. Eliashberg. "Considering all the business pressures that theater managers experience, the model serves as a decision tool that can supplement their experience and business sense."

The authors' model contains two tiers.

The first tier provides a movie selection plan that a manager can use to prepare for an entire season and bid for movies before the season's start. It employs a prediction system the researchers developed based on matching upcoming movies with similar movies that have played in the theater in the previous seasons. The model takes into account five attributes of a movie: genre, MPAA rating, sequel, stars, and distributor.

The manager begins with a two-dimensional, week by screen chart. After a bid plan is developed and finalized with distributors, the exhibitor fills in some of the empty cells in the chart. The remaining cells represent slots that can be decided during the season, either by extending movies that the exhibitor booked before the season or by scheduling other movies that may become available later in the season.

The second tier is an adaptive scheduling approach that helps the exhibitor in weekly decision making during the season.

The authors' interviews with theater managers and analysis of data show that the human element - particularly the obligation that managers feel to run films pushed by distributors - often hurts the bottom line. Their research shows that exhibitors need to be selective in their choice of movies and may suffer a substantial loss in profitability if they place too much emphasis on accommodating distributors.

Batman and Manhattan Theater

The authors tested their model by matching their results with data compiled by Variety. The authors studied sales figures and other information for the 84th Street Sixplex on New York's Upper West Side, chose the theater for their comparison because of the theater's size and feature of first-run films. Data available for the year 1989 provided the best information for test purposes about contract terms, schedule, gross revenues, forecasts for each movie at the theater, costs and concession profits, and the amount of the "house nut" - the amount guaranteed to the theater for each film. Contract information about one of the year's popular movies, Batman, allowed the researchers to incorporate legal considerations into their model.

By developing an optimization model, commonly employed by operations researchers, the authors proposed an alternate schedule for the year that would have screened only 27 of the 43 movies actually scheduled by the exhibitor. The model's results would have yielded approximately 35% higher cumulative profit than the Manhattan theater's actual receipts.

In their analysis, the researchers found that distributors' net revenue remained about the same, implying the possibility of a win-win situation for exhibitors and distributors.


The Institute for Operations Research and the Management Sciences (INFORMS®) is an international scientific society with 12,000 members, including Nobel Prize laureates, dedicated to applying scientific methods to help improve decision-making, management, and operations. Members of INFORMS work in business, government, and academia. They are represented in fields as diverse as airlines, health care, law enforcement, the military, the stock market, and telecommunications. The INFORMS website is at .

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.