Date of Graduation

5-2025

Document Type

Thesis

Degree Name

Bachelor of Science in Industrial Engineering and Operations Analytics

Degree Level

Undergraduate

Department

Industrial Engineering

Advisor/Mentor

Sullivan, Kelly

Committee Member

Curry, Robert

Abstract

Collegiate football teams often compete in groups of 8-20 teams known as conferences. One such conference, the Atlantic Coastal Conference (ACC), added three new schools for the 2024-25 season, bringing their total to 17 teams. Currently, each ACC team plays eight games against others within the ACC. At the season’s conclusion, the two ACC teams with the best intraconference record compete in a conference championship game. With 17 total teams each playing eight games, the ACC could have a three-way tie for the best record where none of top the three teams play one another. To avoid this situation, we introduce the Minimizing Uncovered Triples (MUT) problem, in which we build a conference schedule for all teams that minimizes the number of uncovered triples in which none of the three teams in the triple play one another. To solve the MUT, we present two integer programs (IPs). The first IP is a straightforward approach that determines each team’s opponents according to predetermined scheduling constraints while the second IP is a divisional approach that builds upon the first model by forming groups of teams as divisions and ensuring that teams within the same division play each other. We first explore the computational efficacy of each approach for the ACC conference schedule scenario and extend our models to larger examples with more teams and games.

Keywords

integer programming; maximal covering location problem; college football; sports scheduling

Share

COinS