• Hybrid modeling and chemical-engineering-informed machine learning
• Data-driven and simulation-based global optimization
• Carbon capture and sustainable energy systems
• Circular Economies: Modeling and optimization for plastics and fertilizer recycling
• Bioprocess modeling and optimization

Dr. Boukouvala is the PI of the Data-Driven Process Systems Engineering (DDPSE) Lab, founded in 2016. The lab develops hybrid mathematical models and optimization tools to understand, design, and operate complex process systems that face constraints and competing objectives. The group combines machine learning and operations research with chemical engineering fundamentals. Its work spans scales from particle- and molecular-level phenomena to single processes, integrated systems, and enterprise-wide operations.

On the methods side, the DDPSE lab members are working on developing hybrid approaches that merge mechanistic knowledge with machine learning, as well as methods that use AI to speed up the discovery of phenomenological models. The DDPSE lab also develops data-driven optimization algorithms for expensive simulation-based problems. These methods are applied to pressing problems in energy and sustainable manufacturing, often with experimental and industrial collaborators. Examples include designing carbon capture processes for flue gas and direct air capture; modeling mechanochemical plastics recycling from ball-mill particle dynamics through techno-economic, life cycle, and supply chain analysis; hybrid modeling of electrochemical systems for fertilizer recycling; and hybrid modeling for bioprocess monitoring and control. Emerging directions include self-improving physics-informed digital twin models.