A Quantum Annealing Approach to Reduce Covid-19 Spread on College Campuses

TL;DR

This paper demonstrates a proof-of-concept quantum annealing method to optimize student groupings on college campuses, aiming to minimize COVID-19 spread through network-based cohorting, showing promising simulation results.

Contribution

It introduces a novel quantum annealing approach for cohorting students to reduce disease transmission, a method not previously explored in this context.

Findings

Quantum grouping reduced total infected students in simulations

The approach decreased peak infection percentages compared to random groupings

Potential for practical advantages in larger networks is suggested

Abstract

Disruptions of university campuses caused by COVID-19 have motivated strategies to prevent the spread of infectious diseases while maintaining some level of in person learning. In response, the proposed approach recursively applied a quantum annealing algorithm for Max-Cut optimization on D-Wave Systems, which grouped students into cohorts such that the number of possible infection events via shared classrooms was minimized. To test this approach, available coursework data was used to generate highly clustered course enrollment networks representing students and the classes they share. The algorithm was then recursively called on these networks to group students, and a disease model was applied to forecast disease spread. Simulation results showed that under some assumptions on disease statistics and methods of spread, the quantum grouping method reduced both the total and peak percentage of infected students when compared against random groupings of students. Scaling to larger networks, it is possible that this quantum annealer-assisted grouping approach may provide practical advantage over classical approaches. This paper, however, is strictly a proof-of-concept demonstration of the approach and is not intended to argue for a quantum speedup.