A Conceptual Framework for AI-based Operational Digital Twin in Chemical Process Engineering
Evrim Örs, Robin Schmidt, Moein Mighani, Marwan Shalaby
Abstract
As digitalization is becoming more and more an integral part of business in all sectors, the digital twin paradigm starts to play a more crucial role. This paper primarily aims at describing a generic framework for digital twin development in chemical process industry from an operational perspective. The main building blocks of the operational digital twin are presented, namely data management, process modeling, process optimization, production scheduling, and process control, as well as the deployment. Strong emphasis is put on the advanced process control hierarchy. Additionally, the role of artificial intelligence in the development and deployment of operational digital twin in process industry is presented, particularly regarding surrogate modeling, predictive modeling and AI supported optimization and control. Consequently the potential for new business models induced by digitalization is discussed, and an outlook for prospective research topics is provided.