As modern factories have grown increasingly complex, companies are relying more and more on computer models to monitor how production processes play out on the factory floor. The manufacturing of a single product could involve numerous models, each with dozens of steps.
One challenge for production managers is how to keep all those computer models in sync as designs change. For example, if the size of a hole changes in one model, the other model would have to be updated manually — a tedious and time-consuming process.
Chris Paredis, professor in the George W. Woodruff School of Mechanical Engineering, walks in the Montgomery Machining Mall, in Georgia Tech’s Manufacturing Related Disciplines Complex.
Researchers who specialize in model-based systems engineering at Georgia Tech are working on ways to link various industrial computer models so they can be kept in sync faster and with less labor, increasing automation and allowing production managers to see the big picture faster.
“Those dependencies between models are a problem right now,” said Chris Paredis, a professor in the George W. Woodruff School of Mechanical Engineering. “These things are becoming a cause for lots of rework and effort to maintain consistency between all of these representations. The chances that something will get missed are pretty high. So anything that we can do to support that process and maintain consistency in an automated fashion is a big step forward.”
Much of Paredis’ work focuses on building the architecture that allows two or more systems to communicate and synchronize information. People in the industry call this consistency management.
“It could be that a diameter becomes a radius in another model. It could be that we have inches becoming meters,” Paredis said. “There are all kinds of things that can be slightly different. Lots of different kinds of information need to come together.”
His team’s research came into play recently on a project with Boeing, where the group considered the manufacturing process of an airplane landing gear assembly.
The goal for the project was to design an integrated system model that could update itself within hours in response to changes in the manufacturing process.
“When we design, we need to understand the consequences of the choices we are making,” Paredis said. “If it takes three or four weeks to update the process plan, then every time you make a change to your design, you have to wait three or four weeks to understand what the consequences are. And you can’t do business like that.”
The goal of the research is to enable factory managers to make decisions faster and with more information — rather than spending time updating models and simulations, Paredis said.
“There are too many people spending their time transcribing information and chasing down information produced by others, and these are really non-value-added tasks,” he said.
Having the computer models updated quickly has helped keep manufacturing companies nimbler, saving them time and money.
“The more inconsistencies you catch, the better, because any one of those can cause a major headache later on,” Paredis said. “The longer businesses wait for all aspects of the system to be updated, the costlier it gets to make changes.”
Josh Brown is a senior science writer at Georgia Tech. A journalist by training, he’s spent the past decade writing about economic development, medical research, and scientific innovation.