PITA FY2014 Projects

Variability Control And Analysis In Chemical Process Optimization

PI:Luis Zuluaga, Industrial and Systems Engineering

University:Lehigh University

Co-PI(s):

Industry Affiliate(s):Air Products and Chemicals

Summary:Variability is an inherent characteristic in chemical processes. This is, among others, due to uncertainties associated with environment variables, measurements, and properties of the raw materials. Because variability brings with it uncertainty, it is not surprising that there is typically a desire to have the minimum possible variability in any process. In the 80’s, classical variability reduction techniques like Six Sigma where developed to improve the performance in manufacturing processes. The Six Sigma technique has spread to all areas of industry and has brought with it a vast number of success stories in different industries. However, as the market has rapidly become more demanding, and competitive, a need for more advanced tools to control and analyze volatility have raised. Nowadays, much more complex techniques based on Operational Research, Probability, and Statistics tools are typically used in the manufacturing industry were the Six Sigma was born. Thus, there is a substantial gap between the state-of-the-art techniques used in manufacturing and the ones used in chemical processes to control and analyze variability. For example, classical Six Sigma techniques remain a popular choice in chemical industries. This is likely due to chemical processes being much more difficult to model than manufacturing ones. The main challenge driving this proposal is to close this gap by introducing novel and tailored techniques to control and analyze variability in chemical process optimization. Towards achieving this goal, the main research objectives of this proposal are to: (1) Develop methodologies to control variability based on state-of-the-art Operational Research, Probability, and Statistics techniques. (2) Develop Operations Research techniques that will allow the application of state-of-the-art Operational Risk Management tools to chemical processes. (3) Develop novel optimization or numerical analysis techniques to obtain the shadow prices; that is, the marginal benefits due to parameter’s changes, associated with chemical processes optimization problems.