Chicago wouldn’t last long under zombie invasion, model finds

The light-hearted project by Argonne researcher Chick Macal and his team tested their agent-based modeling simulations on a Halloween-themed slice of epidemiology, following the spread of a zombie virus and testing out "interventions" that could save humanity.

While the worst-case scenario found little resistance against the walking dead, strategies such as training populations to kill zombies or communications from City officials that tell people how to avoid them provided some more optimistic outcomes.

"This offers great promise for countering the zombie apocalypse in Chicago," said Macal, Director Center for Complex Adaptive Agent Systems Simulation Decision and Information Sciences Division at Argonne and a Senior Fellow at the Computation Institute.

The playful experiment draws attention to the power of the group's city model, dubbed ChiSIM, for strategizing during more realistic public health crises. ChiSIM utilizes a computational method called agent-based modeling, which can simulate the complex behavior of millions of individual "agents" representing, in this case, Chicago's 3 million citizens. The model also represents more than 2 million locations in the city, and simulates the movement of Chicagoans through that virtual map, encountering each other and -- occasionally -- transmitting disease.

The results can help officials and policymakers address ongoing crises and prepare for potential ones. For example, the research team conducts simulations of the flu or ebola (and soon, the Zika virus) in case these diseases reach Chicago or surrounding counties. A previous study on MRSA transmission identified distinct hubs for the disease in the Chicago area, such as the Cook County Jail, where health officials can then focus interventions to prevent broader spread of the drug-resistant bacterial infection.

While a zombie invasion is distinctly fictional by comparison to these real-life public health challenges, Macal told the Chicago Tribune that the playful project provided very serious insights for the performance of ChiSIM and the types of strategies that could be used to stop actual diseases.

"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.

The other members of Macal's team at Argonne are Jonathan Ozik, Nick Collier, Emily Rosenblum and Jessica Trail.

[Watch a video of the study’s results here]