Michael Ciarlegli, a fifth year student in the ICES Program, taught himself computer programming in order to create software that has been applied to two different environmental areas: conservation planning for endangered species, and groundwater management in central Texas.
In layman’s terms, what exactly do you do?
My research falls under the category of applied math: a combination of math,
operations research, and computer science, with applications in biology and
hydrology. I have developed novel direct search techniques and software to find efficient
solutions to difficult optimization problems. We encounter optimization
problems daily, whether it's finding the fastest route to work, choosing the
best way to invest money, or creatively multi-tasking to accomplish more at
work or home.
The name of the software is MASTS, which stands for Modular Abstract Self-Learning Tabu Search. My software has been applied to optimization problems in two different areas: conservation planning for endangered species, and groundwater management in central Texas. In each of these efforts, optimization has been critical given the complexity of the projects and the number of stakeholders involved. As a result, my software contains a lot of features that make it ideal for planners and decision support. The software enables ongoing interactive optimization, which can help planners explore and understand the problem from several different angles.
How did you get the idea to study this?
The first summer in graduate school, I was introduced to an optimization technique that randomly stumbled about searching for good solutions. Although this technique is popular, it strikes me as fumbling around in the dark for a light switch. Half epiphany and half frustration, I thought "there has got to be a better way". As luck would have it, Professor J. Wesley Barnes of the Graduate Program in Operations Research and Industrial Engineering within the Cockrell School of Engineering offers a class that studies organized strategies for optimization, called metaheuristics. This class was a perfect match for my interests, and it set me on my dissertation path.
After taking that class, I got involved in groundwater management with encouragement from fellow student Suzanne Pierce (now Dr. Pierce). She set up an amazing collaborative project between Sandia National Labs and UT, which involved several students and several dissertation projects (I always joked that her dissertation was the only that I knew of that had a staff and six figure budget . . . you don't see that in math a lot!). During that project, I met Dr. Sarkar, who proposed another application for my software in conservation planning.
What sort of practical applications does your research have?
I believe that my software can really integrate scientific modeling and optimization into the planning process. Hopefully, the software package ConsNet will become the new standard for systematic conservation planning. It might even be adopted by international non-profit agencies.
What do you hope to do next?
In addition to supporting the conservation planning software, I plan on
extending my techniques to other types of optimization problems. Although there
are lots of opportunities in academia, it also makes sense to continue this type
of work as a small business.
How has being a graduate student at UT enabled your research to reach the level that it has?
The ICES program (Institute for Computational Engineering and Sciences) is the
perfect incubator for interdisciplinary research. They encourage students to
explore and bridge different departments and disciplines. I had a lot of
freedom in choosing my courses and finding my topic.
Also, the quality and sheer size of the UT faculty guarantees that students have
access to experts in every field. My advisor, Dr. Barnes, is not only
familiar with tabu search [the method used in my software], he helped in its early development.
All of my committee members are pioneers in their respective fields.
Finally, the faculty at UT seems to be fairly open to collaborative projects with the community and other institutions.
What advice do you have for people entering graduate school?
Don't be afraid to collaborate with other students on your research projects. You can usually accomplish a lot more with partners.
Q&A by Elizabeth McKetta, December 2007
