Meng, Q; Bogle, IDL; Charitopoulos, VM; (2024) Data-Driven Chance-Constrained Optimization for Minimizing the Influence of Material Uncertainty on Product Quality. In: Computer Aided Chemical Engineering. (pp. 1579-1584). Elsevier

Text Bogle_ESCAPE34QingboFinal.pdf Access restricted to UCL open access staff Download (541kB)

Abstract

Minimizing the impact of process uncertainties, caused by estimation and measurement errors, unplanned disturbances, or environmental changes, is one of the crucial practical challenges in the pharmaceutical industry. In this work we propose an approach using data-driven chance constraints to eliminate the influence caused by physical property uncertainty in the raw materials via model-based optimization. A flowsheet for the pharmaceutical tableting manufacturing process of the Diamond Pilot Plant (DiPP) at the University of Sheffield is used to test the methodology. Firstly, the Kernel Density Estimation (KDE) technique is applied to generate the inverse cumulative density function for the uncertain variable using historical raw material quality data obtained from the supplier. Different uncertainty risk levels are then considered when obtaining the optimal operating conditions to explore the trade-off between product quality and economic performance. Process operating limitations are addressed for different risk levels to guarantee the desired product quality. Results indicate that the proposed approach can effectively reduce the uncertainty in the product quality caused by raw material physical property changes.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item