Systems Engineering in the Fourth Industrial Revolution

14Hybrid Semiparametric Modeling: A Modular Process Systems Engineering Approach for the Integration of Available Knowledge Sources

Cristiana Rodrigues de Azevedo Victor Grisales Díaz Oscar Andrés Prado‐Rubio Mark J. Willis Véronique Préat Rui Oliveira and Moritz von Stosch

Synopsis

The Quality by Design (QbD) paradigm and novel Process Analytic Technology (PAT) are reshaping the way that manufacturing processes are developed and plants are operated in the future, aiming in the pharmaceutical industry, for instance, at real‐time release of a drug. The application of sensors that can provide real‐time measurements of the process state and processed materials as well as risk assessment and process understanding are key topics of attention. Process understanding is relevant in particular, as it is at the core of process system engineering approaches, where the aim is to provide a better understanding of the involved risks and, therefore, their assessment. In addition, it can also support decisions as to what needs to be analyzed and where (in the process) to place sensors, or adaptive operating strategies. Given these requirements for process understanding, there seems to be a need to change the way we develop, handle, and maintain process understanding. Mathematical modeling is an attractive approach to distill, administrate, and conserve this knowledge in a compact manner.

For mathematical model development, data‐driven and first‐principle approaches can be regarded as the extremes ...