Engineers as People: Understanding the human elements of decision making on a design team.

C-SED consultant and Mechanical Engineering Assistant Professor, Jesse Austin-Breneman, conducts research focused on the decision making process of complex system design. He hopes to use his findings to help project teams improve their technical outcome.

The time Jesse Austin-Breneman spent in Peru working in NGOs showed him the challenge of designing for the field. Frustrated by solutions that didn’t fulfill their intended needs, he wondered how to modify their design approach. His current research focus is derived from this question of “how can we get good answers?”

Austin-Breneman wants to close the gap between best practices of complex system design and what decisions actually happen on project teams. He explains his research in the light of problems he often sees with design for development. “We design things and people don’t use them. At the end of the day, I think what we’re trying to do is provide some understanding for what are the type of things we could do to improve that outcome.”

It’s not easy to untangle the many factors that affect system performance. Austin-Breneman is looking specifically at “the human elements of complex system design.” He gives the example of a team working in a cross cultural setting to determine and understand user needs, “we’re interviewing people, we’re observing people. But obviously who you are and how you act as a team, how does that change how good your answer is?” These are questions Austin-Breneman is beginning to answer through a combination of human subject tests and mathematical modeling of human behaviors.

In order to optimize the decisions that teams are making to improve technical performance, they first have to understand what human phenomena are influencing the team and the decisions they make. Austin-Breneman and his research team conduct interviews to gauge what phenomena might be at play in an engineering team, and then use those observations to make the mathematical models. Austin-Breneman explains that the models he’s creating are not so much prescriptive, as they are descriptive of observed behaviors on design teams. The mathematical models represent simplified versions of real world situations, with numbers standing in for different observed behaviors. Changing the ‘rules’ of the mathematical model simulate a change in the way a project team might make decisions.

One finding that’s surfaced so far is the question of how to deal with uncertainty. Austin-Breneman highlights how inherent uncertainty is to the design process. “When you design things, you have incomplete information and then you have to make decisions as if you knew everything.”  He explains that it is impossible to design with complete information due to uncertainty in the models you create, uncertainty in error of measurements, and uncertainty in what factors will have changed by the time your design is implemented. There are formal methods, like uncertainty quantification, for dealing with these unknowns. However, often these methods aren’t actually used in practice. Austin-Breneman thinks that “the real interesting piece is to try and bridge those two worlds” and help teams adjust their design processes to yield better outcomes.