Close Menu

Lulu Kang: Uncertainty Quantification from Data Collection and Analysis

Lulu Kang, associate professor of applied mathematics, is interested in the development of statistical theories and methodologies and their applications in science and engineering. Kang has a Ph.D. in Industrial Engineering and an M.S. in Operations Research from Georgia Tech as well as a B.S. in Mathematics.

Associate Professor of Applied Mathematics Lulu Kang


Lulu Kang


“We’re living in a world of big data,” said Kang. “Big data means big challenges, particularly in abstracting the most useful and accurate information from the gigantic amount of data.

“On the other hand, we live in a world of small data,” she continued. “Sometimes data are very expensive or time consuming to collect – such as the genetic information from patients, or climate simulation that is extremely computational and time-consuming to run. So then you need a clever data collection scheme, that is, design of experiments (DOE), to collect the most useful data with the least amount of resources or time.”

She noted that complexity of the data structure is another challenging problem that statisticians need to tackle today, even more so than the volume of data. One such complexity issue is that there can be various kinds of data types describing the same process or system. For instance, data collected by thermal sensors (numerical type), high-speed cameras (functional type), and quality control test (pass/fail binary type) can all be available to detect the same manufacturing process. “As there is more and more complex data available, we have to keep coming up with new statistical theories and methodologies for the proper data analysis--in other words, to quantify the uncertainty in the underlying random system/process so as to uncover the rules that govern such a system/process,” said Kang.

Among other projects, Kang has been funded by the National Science Foundation (NSF) Civil, Mechanical, and Manufacturing Innovation (CMMI) division for “Collaborative Research: Experimental Design and Analysis of Quantitative/Qualitative Responses in Manufacturing and Biomedical Systems.” She also was funded by the Environmental Protection Agency (EPA) for “Combining Measurements and Models to Predict the Impacts of Climate Change and Weatherization on Indoor Air Quality and Chronic Health Effects in U.S. Residences”; by Illinois Tech’s Wanger Institute for Sustainable Energy Research (WISER) for “Stochastic Search for Optimal Wind Farm Layout”; and more. Recently, she received a $40,000 seed grant from the Center for Interdisciplinary Scientific Computation (CISC) at Illinois Tech with Sonja Petrovic, associate professor of applied mathematics, and Mahima Saxena, assistant professor of psychology for a project that combines statistics, computational mathematics, and psychological sciences to study worker experiences specific to occupational health psychology. Kang’s research has included applications in manufacturing, mechanical, civil engineering, and more. She also has closely collaborated with scientists from industry, such as Proctor & Gamble.

Kang has supervised seven master’s students on their thesis and advised two Ph.D. students. All worked on various research topics in statistical modeling and experimental design methods. One Ph.D. student graduated in 2014 and is now an assistant professor at DePaul University. The other will soon graduate. While earning his Ph.D., he gained useful practical experience through one summer internship in industry as well as research experience through one summer internship at Argonne National Lab and one semester visit at the Statistical and Applied Mathematical Sciences Institute (SAMSI) in Durham, NC. Kang has also worked with several undergraduate students in two summers and acted as mentor for two local high school students. With the need for statisticians growing rapidly, and more and more students interested in the field, it is certain that she will guide many more to come.