Choosing a COVID-19 surveillance testing site: Game theory
http://nautil.us/blog/let-game-theory-tell-you-when-its-time-to-go-shopping
Part of our new normal since returning back to Cornell’s campus during the COVID-19 pandemic involves getting 2 weekly surveillance tests, which are offered at several locations around campus. Despite the fact that there are multiple locations, it seems to be the case that some testing sites are much more popular than others and there are obvious peak times of day when the lines for these tests are much longer and slower.
In order to avoid the excruciatingly long waits, some Cornell students have been unknowingly applying game theory and strategizing which location and time of day is best. Most students, of course, simply choose based on convenience, choosing the location closest to where they live and during times of the day when they don’t have classes. Because of this common behavior driven by convenience, two of the busiest locations seem to be RPCC and the CollegeTown testing sites, both of which are in the hearts of the housing centers for freshmen and upperclassmen, respectively. And peak times of day seem to be around lunchtime or the late afternoon when I would guess most classes end. However, because most students opt for the convenience, the convenient locations end up facing an influx of individuals, leading to longer lines(whose length are exaggerated by the 6 ft social distancing measure), longer wait times, and more time wasted in areas that are essentially large gatherings of people.
Even before learning about game theory, I noticed this pattern and have opted to sacrifice some convenience in order to save time. Instead of choosing one these convenient locations, I opt for locations that are a bit out of the way such as Bartels or the ILR Conference Center and/or I try to go in the mornings, a time when I know many people aren’t even awake yet and would much rather sleep in that go get their surveillance test.
Although I couldn’t find any articles specifically related to this example of game theory on Cornell’s campus, I found an article from Nautil discussing a similar phenomenon related to choosing a time to go to the grocery store. The article mentioned the El Farol Bar Problem, which involves 100 people who want to go the bar but if 60% of the people go to the bar, the evening is unenjoyable, but if less than 60% go to the bar, the night is fun. While we’re definitely not in a scenario where we’re considering going to the bar, the payoff matrix for this game seems pretty similar to what I may except for our testing site example.
payoff matrix for El Farol Bar Problem