As a modeling and simulation engineer, Bianica Pires, Ph.D., investigates artificial societies, which are simulations that use social science research to replicate human and social communities. Here, she spoke with us about virtual work that solves real problems.
From a very early age, my family circumstances forced me to pay attention to the impact of federal and local policies on people. Although I didn’t know it at the time, I was developing my interest in computational social science.
Now I’m supporting MITRE’s mission of solving problems for a safer world by generating synthetic communities—a method of artificial intelligence [AI] that simulates statistically representative real-world communities.
Artificial Societies: Not Just a Game
If you’ve ever played SimCity, you already have an idea of what it means to develop a simulated city or region for analysis. We create these artificial societies that are representative—not identical—to a geographic region, such as a city, county, or state. We’re not replicating the actual individuals living in these locations. Instead, we’re looking at the demographic breakdown, including race, education, economics, and family units.
In these simulated cities, there are social networks just like in the real world, including household members, colleagues, classmates, neighbors, and friends. Then we add to this simulated society the issues we’re trying to study and solve.
For example, stay-at-home orders early in the pandemic had the unintended consequence of an increase in opioid overdoses. Before the pandemic, we had increased social isolation in a model—and it predicted this effect. In future national emergencies, the Centers for Disease Control and Prevention, and even state and local governments, will have this data at hand to drive policy related to support services for opioid misuse.
Growing up, I saw firsthand how external factors drive life-changing decisions and pathways, and the importance of social and support systems to improve outcomes.
Real-World Applications to Social Simulations
In May 2021, President Biden issued an executive order on climate change, partly to ensure that insurance policies don’t violate principles of equity and create hardship for minorities and underserved communities.
As a result, our group looked at the risks of floods, wildfires, and other climate events, and affordability of insurance in underserved communities. Using artificial societies, we’re able to assess how major disruptions in private insurance coverage in U.S. markets might impact those most vulnerable to climate-change impacts—from both a protection and financial standpoint.
Taking data from the U.S. Census Bureau, First Street Foundation, the Department of Agriculture, and local governments, we built out the environment to understand individual properties: the construction methods, materials used, and how people live in them. For example, when we look at local data regarding wildfires, we examine the likelihood of whether these properties would burn.
Then we go back to the people aspect—what’s the household income in this area? What are monthly household expenses? Do they rent or own, which may change whether they personally make changes to their homes so they’re more adaptable to climate change or risks?
Personal Experiences Pave the Way to Computational Social Science
While growing up, I saw firsthand how external factors drive life-changing decisions and pathways, and the importance of social and support systems to improve outcomes.
My Brazilian mother’s schooling was frequently interrupted in her youth because she was needed at home, or she had to go to work to get food on the table for her family. And when I was just 19, my father had a brain injury that left him with a permanent cognitive disability. Suddenly, our family had no health insurance or the financial ability to purchase a policy.
My mother was determined I get the education she hadn’t been able to. As an undergrad, I studied engineering, but a sociology course I took purely for a writing-intensive requirement really stuck with me. At the time, computational social science was just emerging as a field. I realized the potential of combining soft and hard sciences to look at societal problems and find solutions.
Our MITRE team strives to employ science for the better good of all people. I’m thankful for the opportunity to highlight the social side of solving worldwide problems.
This year, Pires was honored with BEYA’s Research Leadership Award in recognition of her role as a thought leader in shaping research to demystify the underlying mechanisms behind societal outcomes. Learn about all of MITRE’s 2023 BEYA Award winners.
Join our community of innovators, learners, knowledge-sharers, and risk takers. View our Job Openings.