Stevens faculty-student duo uses a form of computational simulation to stress-test crew performance for future Mars missions — and learns that putting similar people together doesn't confer an advantage
As the world follows the drama of two American space crew members on the International Space Station whose nine-day mission will unexpectedly stretch to eight months, the pair have worked well together in spite of the uncertainty and stress of their situation.
And space-crew teamwork is suddenly a hot topic.
To help inform that planning, Stevens professor Hao Chen, an expert in space systems, teamed with doctoral student Iser Pena to simulate and understand the crew mixtures that would perform best.
"Missions would take more than two years to get to Mars and back," Chen explains. "These crews need to be able to live and work together in close quarters for all of that time."
"We wanted to derive more insights into how that would play out."
Is difference or sameness more valuable?
In theory, teams of people who were trained similarly, think alike and speak the same languages would seem to be equipped to the task.
Diversity in a planned space mission is not a weakness, the Stevens research found, but rather a strength.
However, previous research has demonstrated that homogenous teams tend to overlook new ideas — don't think outside the box — when they need to cope with unexpected challenges and emergencies. Crews staffed with diverse team members are more likely to introduce, discuss and try new ideas than their counterparts.
"On the other hand, it often takes longer into a task or mission for a diverse group to gel into a smooth-functioning group because of language differences and other communications challenges as everyone learns each other's style," says Pena.
Chen and Pena obviously couldn't put real crews of astronauts through the paces. Instead, they did the next best thing: programmed a computer model to "design" simulated crew members equipped with varying strengths, weaknesses, backgrounds, experience levels, psychological factors (such as stability and social skills), and fitness levels ("health status").
Then they ran a complex simulation of a Mars-based mission for 500 days, pitting crews of six quite similar astronauts against six who were quite different from one another.
"This technique is known as agent-based modeling (ABM)," explains Chen. "It's a common way for scientists and planners to get a handle on a really complex system or challenge, such as an interplanetary mission, where you can't possibly test all the options with real-life, real-time tests beforehand."
The model they designed drew upon a host of psychological, organizational and other data and theory to factor in the various effects of time, stress, fatigue and personal engagement on team performance.
So: once the duo's models ran through the numbers, which group had the right stuff?
"While both teams performed roughly the same on the simulated mission, the 'diverse' team scored slightly higher overall," reports Chen. "This is interesting. It indicates diversity in a planned space mission is not a weakness in any way, but rather a strength."
"Extrapolating, we could conclude that space crews don't necessarily need to be composed of people who all look like each other, talk like each other and have the same life experiences."
"The higher stress stress levels, yet better performance, of those heterogeneous teams was a significant finding that could be attributed to the diverse perspectives and problem-solving approaches they bring to the table and to the mission," adds Pena.
The model also revealed some surprising quirks that could play out during long space missions.
Both crews' health deteriorated significantly due to the considerable physiological challenges (and stress) of such a long mission. Interestingly, however, the health of the "diverse" crew members stayed more robust during the first half of the simulated mission, while the "heterogenous" crew's made the gradual decline you'd expect over time.
What's next? The pair will continue refining their use of ABM to simulate Mars crew staffing.
"We only did a few full-mission simulations to obtain the findings reported in our AIAA conference presentation and paper," notes Pena. "But we will be continuing this project. I will be running it hundreds and hundreds more times to obtain improved data and derive further insights from this work."
"We will also introduce additional skills and variables to the model to make the findings richer and more reflective of real missions," adds Chen.
Findings were presented at the 2024 AIAA ASCEND (Accelerating Space Commerce, Exploration, and New Discovery) conference in August.
