Sound Engineering: UW-Madison’s burgeoning optimization community

Share this story:

In the latest Sound Engineering podcast from the UW-Madison College of Engineering, Industrial and Systems Engineering Associate Professor and Wisconsin Institute for Discovery Optimization group member Jim Luedtke discusses why the field of optimization is growing, and why UW-Madison is an exciting place to pursue it. Luedtke also addresses some recent criticism of optimization and stresses that, as with any powerful tool, it’s up to the user to apply it correctly.

Jasmine Sola: With the field of optimization growing rapidly, Jim Luedtke is researching ways to use mathematical optimization to solve some of the world’s everyday problems. As an associate professor in industrial and systems engineering and a member of the Optimization group at the Wisconsin Institute for Discovery, Luedtke is part of a growing, multidisciplinary optimization community at UW-Madison. Luedtke says there’s a good reason so many researchers are drawn to this field.

Jim Luedtke: What we do is we develop math models of some complicated planning and decision problems, and then we try to develop algorithms and methods to try to get the best possible solutions to those math models, ultimately with the goal to try to help decision makers make better decisions and better plans.

Jasmine Sola: For Luedtke, mathematical optimization is so powerful because it can develop solutions that can then be applied in a variety of contexts.

Jim Luedtke: One thing that I think is really a nice feature of optimization is that it provides a good way of thinking about these types of problems. Power systems planning is a good example. A lot of times they have a lot of uncertainty now in how much energy is getting produced from wind sources or solar sources. In optimization, we try to develop tools that would  help them make a plan, make those decisions ahead of time.

Jasmine Sola: Recently, however, optimization has been criticized as an overly technical approach that neglects the human aspect of problems. But Luedtke argues that it doesn’t have to be that way. He says optimization is completely in the hands of the user.

Jim Luedtke: Any time you have a powerful tool, it’s up to the user of it how you will use it, right? Another one that’s come up recently is in doing employee scheduling. And this is one that I’m actually a bit sensitive to, because I actually have done research on using optimization to do employee scheduling. The criticism that’s come out, though they didn’t directly point back to optimization as having done it, but in my mind that’s been an enabling tool of this, is people and companies are trying to do scheduling in such a way to make things so optimal that you get these employees who get a schedule that’s really broken up. They’ll work eight hours and then only five hours off and then another eight-hour shift, and it’s really terrible on the employees, it’s not a predictable schedule. And it’s quite possible that that crazy schedule came out of an optimization model. I don’t know. But it could. The problem is really, it’s not about the optimization tools, it’s about the way the managers decided they wanted to use them. If managers decided that they wanted to provide good predictable schedules to their employees, then optimization tools definitely can do that for you, and they will help you do it in a better way than doing it in other ways. Optimization never makes the decisions. It’s a tool that helps people make decisions. 

Jasmine Sola: As part of a collaborative optimization research community on campus, Luedtke and his colleagues have a unique perspective on the future of the field.

Jim Luedtke: We’re a unique university here in the sense that we have a lot of strength in optimization in various different departments—in industrial and systems engineering where I am, in computer sciences, we have some chemical engineers who do this, people in statistics who are doing things quite related to this as well, electrical engineering. That’s really great, that we have so much strength in mathematical optimization. But prior to the WID, they were all spread out. But with this Optimization group, and this space in WID where we can congregate and be together, you just start having more cross-fertilization of ideas. If I’m unsure about a specific topic that’s not my exact specialty, there’s somebody there who I might ask about it. Rather than have to go look through a bunch of books and get to that, I can just ask a question and they just have the answer right there. 

Jasmine Sola: For more about Luedtke’s research, visit

Jasmine Sola