Environmental Engineering Science Seminar
Friday, April 15, 2005
Lopata 101
Time: 11 am
Application of Experimental Statistical Models in Process and Product Design
Dr. Gennadiy Statyukha
National Technical University of Ukraine “Kiev Polytechnic Institute”, Kiev, Ukraine.
In this presentation, the application of statistical modeling methods to complex technological processes is considered. The proposed approach consists of an experimental design to generate the representative data, which is then used to build statistical models for different processes and products. The experimental data are summarized in the form of polynomial models, which can be used in process or product design and optimization. We utilize the classification methods to differentiate between different classes of processes and products, and use the results of classification to design novel processes and products that are “mixtures” of cases for which experimental data are available. The range of problems that can be solved using the proposed approach include the optimization of operating conditions, determination of the optimal product composition and properties, and minimization of the production costs.
We have developed novel software that facilitates the solution of these problems, and will describe its capabilities during the presentation. We will present illustrative examples from the variety of areas, including chemical technology, building materials, pharmaceuticals, food, and cosmetics. In last time, we use successfully the ESMA for environmental monitoring, especially for indicators and indices of sustainability calculations.
The presentation will summarize the propose approach and argue for its application as a tool that allows us to treat the data and to quickly narrow the available choices to a manageable subset of promising candidates, which warrants further experimental and theoretical investigation.
RETURN TO SEMINAR LISTING - SPRING 2005
RETURN TO
ENVIRONMENTAL ENGINEERING SCIENCE MAINPAGE