Creating digital twins of chemical processes is becoming more and more essential for improving existing systems and designing new ones in a safe and cost-effective manner. With the increasing computation power, these processes are modeled in greater detail by considering complex phenomena. Yet, finding the best operational conditions or new process routes are very challenging due to model complexities and excessive computational effort required to find a viable solution. That's why data driven modeling is at the heart of our research program, where we explore the untapped potential of using data analytics in process systems engineering.

We focus on:

• Algorithmic developments for data-driven optimization
• Data analytics
• Bi-level optimization
• Multi-objective optimization
• Stochastic analysis
• Integration of time and temporal scales
• Supervised and unsupervised analysis

Several application areas include:

• Integrated planning and scheduling
• Food-energy-water nexus
• Steam cracking for olefins production
• Oil production optimization
• Environmental & biological modeling