Statistics Seminar - 03/21/23

Abstract: 

Response surface methodology (RSM) is widely used in engineering and industrial processes. RSM aims to obtain a combination of input variables such that the response is optimal. Many RSM strategies have been developed. They are highly efficient and useful in studying the underlying relationship between the input variables and the corresponding responses. The conventional RSM is limited and focuses on point-valued responses. However, in many practical scenarios, the response would be in other formats, such as an interval. Studying the response surface with interval response is rather challenging. Instead of building models based on the point observations, we propose a strategy that models the interval responses as a whole. The proposed method can provide the optimal solution in revealing the contributing variables, estimating the effects of the variables, and obtaining the optimal response. In addition, it provides a solution when the endpoint of the interval response is infinity while the conventional methods have failed. More detailed comparisons will be provided in the case studies. 

Bio:

Dr. Ching-Chi Yang is an assistant professor of Mathematical Sciences at the University of Memphis. He received a doctoral degree in statistics from the Pennsylvania State University in 2019. His primary interests include response surface methodology, interval data, statistical learning, and dimensional analysis. His research projects have covered a range of topics such as using response surface methodology to design a motorcycle, determining the optimal treatment for diabetic patients, and predicting stock prices via stock research reports.