https://orcid.org/0000-0001-5900-7424
Summary
The onset of COVID-19 in the spring of 2020 disrupted academic schedules for many colleges, including course registration for the fall of 2020. We solve the problem of salvaging the majority of an existing semester course registration when separating the semester into two terms. The problem was solved using integer programming.
MSC:
Acknowledgements
The authors gratefully acknowledge the significant contributions of Jason Maher, Grinnell College Registrar, throughout the project. They thank Andrew Beveridge, Barry Thomas, and Coralia Cartis for conversations helping with model formulation, perturbation heuristics, and framing the problem in the literature. Lastly, the authors thank the anonymous referee for critique and recommendations that improved the article.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 The typical course at Grinnell College is four or five credits. While the College offers various one and two credit courses, we only consider four and five credit courses in the partitioning problem.
2 We focus here on minimization to match our problem; the intuition applies analogously to maximization.
3 It is possible to formulate problems where rounding the relaxation’s solution is as far from optimal as desired both in terms of the solution and objective value, not to mention the common case where the rounded solution is not even a feasible point.
4 The apparent lopsided nature of the number of fixed courses is related to the 31 First-Year Seminars preassigned to
5 In our implementation, we perturbed the objective with absolute values of samples from Matlab’s normal pseudorandom number generator, randn.
6 In a standard academic year, First-Year Seminars are all assigned the same protected time slot. In this splitting, faculty teaching Seminar were allowed to select their time slot after all other courses were given their designated time slot. This provided a range of time slots for first-year students learning online and spread around the world.
Additional information
Notes on contributors
Jeffrey D. Blanchard
JEFF BLANCHARD (MR Author ID: 856216) is a professor of mathematics at Grinnell College. He has been an NSF International Research Fellow, a Project NExT Fellow, and a Harris Faculty Fellow. His primary research areas are compressed sensing, sparse approximation, and high performance computing with graphics processing units. He currently runs Grinnell’s Wilson Center for Innovation & Leadership.
Marc Chamberland
MARC CHAMBERLAND (MR Author ID: 335642) is the Myra Steele Professor of Mathematics at Grinnell College. He has published in various research areas, including differential equations, number theory, classical analysis, and experimental mathematics. He has also sought to popularize math with a book, mathematical artwork, and a YouTube channel. A current project is writing a book on the number .