Supply Chain Frontiers issue #12. Read all articles in this issue.
John C.W. Parsons, a 2004 graduate of CTL's Master of Engineering in Logistics (MLOG) program, was selected as co-winner of the case competition at the Institute for Operations Research and the Management Sciences (INFORMS) annual meeting held last month in San Francisco.
Parsons, currently an associate at McKinsey & Company, presented a case with MIT Professor Stephen Graves based on research they had done in 2004 in partnership with Reebok International Ltd. on the company's demand forecasting of NFL team replica jerseys. Parsons conducted the research under the direction of Dr. Graves as part of his Master of Engineering in Logistics degree at MIT.
Through the research, Parsons attempted to tackle the notoriously difficult task of predicting demand for football jerseys, which is often driven by uncertainties like team popularity and player trades. His thesis attempted to solve this problem by pinpointing a modeling approach that would allow Reebok to plan and manage inventory in a way that managed costs while providing the flexibility required to meet demand.
First Parsons had to consider Reebok's supply chain. The jerseys were being made by contract manufacturers and then shipped to Reebok distribution centers. Some jerseys were "dressed" with players' names and numbers; others were shipped as "blanks" without names and numbers, which were added later in the company's Indianapolis, U.S. distribution center.
Parsons tested out two modeling approaches. First, the Simple Newsvendor Model, which considered demand for the selected players separately from the demand for blank jerseys. In this model, blanks were only used for non-selected players. But it was the second approach--the Newsvendor Model with Risk Pooling--that yielded the most favorable results. By combining the simple model described above with the risk-pooling opportunities afforded by including blanks as an option, blank jerseys could be stocked to fill demand for both selected and non-selected players.
In February, following retailers' initial order placements, enough information is available to generate a team and player level forecast. Using this forecast the Planning Manager can determine an optimal quantity for each player and team blank jersey, Parsons explained. Over the next several months purchasing programs can target these optimal quantities. As new information is available the forecast - and the model - must be updated. Player movements, increased early sales, or heightened expectations must all be incorporated into the forecast and managed to ensure that proper planning is carried out.
The analysis is not a cure-all, but it does show how establishing the optimum balance between the use of dressed and blank jerseys can maximize profits. Neither is the model a substitute for industry expertise. As Parsons pointed out, "It cannot and should not entirely replace the experience, gut feeling, and art that every member of the Reebok team must have to understand and react in the professional sports business."
Parsons' thesis is entitled "Using A Newsvendor Model for Demand Planning of NFL Replica Jerseys." Anyone interested in reading the paper in its entirety can purchase a copy by contacting MIT Document Services.