As cities and other municipalities prepare for a 2020 presidential election that will be unlike any before it, set against the backdrop of a pandemic, they face dilemmas in how to structure and staff their in-person polling places: How can they best protect voters and poll workers from COVID-19? How do they minimize wait times? What if there are shortages of poll workers? Should they consolidate polling locations or turn to sports arenas as voting sites?
Laura Albert, a professor of industrial and systems engineering at the University of Wisconsin-Madison, and PhD student Adam Schmidt have analyzed some of the options in a report they’ve shared with the Wisconsin Elections Commission ahead of the Nov. 3 election.
Albert and Schmidt used Milwaukee as a case study and modeled the city’s election operations with a method called discrete event simulation. They produced five broad recommendations for election officials across the country:
1. Municipalities should ensure they have more poll workers than previous presidential elections. These additional volunteers can operate more check-in booths, sanitize voting booths and manage lines. The researchers say officials should plan to add one more check-in booth per polling place than what election toolkits call for.
“Even if we have record levels of early or absentee voting, you still need a tremendously large number of poll workers on Election Day,” says Albert. “In part that’s because it takes longer to check in because of the PPE and sanitation procedures that slow down the check-in process. This is the bottleneck.”
2. A priority queue for those who self-identify as high risk related to COVID-19 could help protect particularly vulnerable voters.
3. Rather than consolidating polling locations, which would require substantially more check-in booths and poll workers and could confuse voters in the run-up to Election Day, municipalities should use their standard polling places.
“We did not feel that consolidating polling locations was ultimately going to work to keep the lines at a reasonable length,” says Albert. “When we looked at how many check-in booths would be needed to support consolidated locations, we determined it would create confusion and require more staff.”
4. While sports arenas offer the promise of more space, they also may divert resources—workers and equipment—away from high-demand polling locations and further complicate voters’ choices on where to vote. Albert and Schmidt recommend sticking to regular polling locations.
“It’s best to allocate scarce resources, like poll workers and voting booths, to the polling places where there are the longest wait times and the resources are most needed,” says Albert. “And the NBA arena doesn’t do that at all.”
5. Expanding early voting through longer timeframes and in-person locations, increasing the number of drop boxes for absentee ballots, and enhancing education about the absentee voting process are crucial for preventing long lines at the polls this year. The researchers say officials should aim to get one-half to three-quarters of all votes in early.
Albert, who is a Harvey D. Spangler Faculty Scholar, took up the project after a conversation with Barry Burden, the Lyons Family Professor in political science and director of the Elections Research Center at UW-Madison, about the importance of queuing theory and location analysis in elections systems. Both methods are in Albert’s wheelhouse, and she also sees the work as a timely continuation of her efforts to use operations research to safeguard critical infrastructure in the public sector.
“Election systems are critical infrastructure,” she says. “With the pandemic, we’re seeing that a lot of our processes are different, and we have a chance to redesign them.”
Author: Tom Ziemer