That’s according to a team of scientists at Argonne National Laboratory, who came to the conclusion using a high-powered computer model that’s normally used for much more serious work on infectious diseases.
“The people of Chicago could suffer dire consequences at the hands of a zombie invasion,” said Chick Macal, a senior systems engineer in Argonne’s Global Security Sciences Division. “No part of the city would be spared.”
While the scenario is obviously fictional, Macal and his colleagues on the Argonne Infectious Disease Modeling Team calculated the timeline by running the numbers through an intricately designed system that was originally created to analyze the spread of real diseases.
The lab began working on the project about a year ago, inspired by Halloween and "The Walking Dead," to illustrate how the lab uses that computer model, called ChiSIM.
ChiSIM was originally developed in 2009 to figure out how to deal with MRSA, a strain of bacteria that causes difficult-to-treat and potentially deadly infections.
MRSA, short for methicillin-resistant Staphylococcus aureus, originally gained attention as a problem in hospitals. But then, people who hadn’t been exposed to the health care system began developing infections.
So Argonne partnered with the National Institutes of Health to figure out how the bacteria were spreading, building an incredibly detailed computer model of Chicago, its citizens and how they interact.
Macal and his team used Census data to create individual “agents” — anonymized representations of census data which, as a whole, reflect the 2.9 million people who were in Chicago as of the 2000 Census.
The researchers built simulations for how people move throughout the city over the course of days, weeks and years, mapping their movements and interactions in 2 million real locations.
Then, they threw in MRSA, and figured out how it spread.
When ChiSIM was finally ready to go in 2011, the Infectious Disease Modeling Team ran a 10-year simulation, starting with MRSA’s prevalence in 2001 and running 2 trillion interactions — and found the numbers came out right on the other end.
“The models re-created the past, and we could look into the model and look and see what places were important, what age groups were important in transmitting the disease,” Macal said.
Since then, the ChiSIM’s been adapted for other diseases like the flu and Ebola, and they’re looking into using it for Zika. They can add in more areas, like larger Cook County or DuPage County, or swap in Census and location data from other areas.
And, of course, they can use it to model a zombie apocalypse.
Macal and the Infectious Disease Modeling Team took ChiSIM and randomly spread a small set of zombies throughout the city (slow zombies, more like “The Walking Dead” than “28 Days Later”). After someone’s bitten by a zombie in the ChiSIM model, there’s a short incubation period of one to two days before the infection takes over.
The team added two parameters if a person encounters a zombie: the likelihood of a person being bitten and the likelihood of the person being able to kill the zombie.
The end result, in ChiSIM’s main simulation: total takeover within 60 days, with no living humans, close to 2 million living (or undead, if you prefer) zombies and close to 500,000 dead zombies.
The Infectious Disease Modeling Team ran several variations of the models, though, and found there were some ways that humans could slow or stop the zombie invasion, mostly by improving defenses and using better offensive tactics.
“A small minority of computer simulation runs really gave us a lot of hope that we could affect the outcome to counter the zombie apocalypse,” Macal said. “Zombies were contained, and there wasn’t massive spreading beyond a small number of places where the zombie outbreak began. So I think there’s a lot of hope here.”
Realistically, Macal said, running the zombie model might actually help the Infectious Disease Modeling Team in the future, inspiring them to make the model better at finding the best outcome.
“This work allowed us to understand and do a better job for finding interventions that result in better outcomes, or even optimal outcomes — better solutions,” Macal said.