From ex-post to ex-ante LCA: using machine-learning to evaluate bio-based process technologies ahead of detailed design
Abstract
Life cycle assessment (LCA) is considered as a suitable methodology for evaluating environmental impacts of processes. However, it requires large amount and often inaccessible process data at early design stages. In addition, many of the renewable/innovative processes are still on lab or pilot scale and their sustainability performance once scaled-up is not known yet.
The purpose of the present seminar is to describe a methodology that provides an approach to streamline LCA for biorefinery processes. The proposed method breaks away from conventional LCA work in that the purpose is to support decisions at early stages, assuming minimal use of available data, and points to most dominant LCA impacts, therefore providing useful feedback to process design decisions.
The proposed computational framework applies machine learning techniques, i.e. artificial neural networks (ANN) and classification trees to produce input-output relationships between predictor variables (such as molecular structure of the chemicals and process related characteristics known in early design stages, etc.) and target variables (i.e LCA metrics).
The parallel use of both techniques (ANN and classification trees) demonstrates how they can complement each other to obtain more accurate streamline LCA estimations, at early design stages. The proposed approach provides a set of models in the form of computationally inexpensive and easily interpretable structures that can be used as pre-screening tools in the development of innovative processes, where process inventories are not available to conduct conventional LCA approaches.
Speakers Short CV
Dr. Paraskevi Karka is a tenure-track Assistant Professor in Sustainable Process Design at the Faculty of Science and Engineering at the University of Groningen (RUG). She graduated from the National Technical University of Athens (NTUA), School of Chemical Engineering, in 2006. She holds a PhD degree in Chemical Engineering from NTUA in the field of process design and sustainability (2018). She also holds two MSc degrees in the fields of Energy Production and Management (NTUA) and in Industrial Management (NTUA and University of Piraeus). During 2019-2020, she worked as a Postdoctoral Researcher in Chalmers University of Technology, Sweden, focusing on modelling and sustainability assessment of biofuels production processes. During that period in Chalmers, she also obtained a Diploma of Teaching and Learning in Higher Education. During 2021 she worked as a Research Fellow at the School of Civil Engineering, NTUA, participating in EU funded projects related to circular economy and industrial symbiosis and as a visiting Researcher in Chalmers University of Technology. Her research interests include modelling, process design, sustainability assessment-Life Cycle Assessment (LCA) of industrial processes, circular economy, technoeconomic analysis and data mining/machine learning techniques. She has participated in various EU and national research projects in NTUA, Chalmers University of Technology and RUG. Her teaching duties involve process design and sustainability aspects in the tracks of Chemical Engineering and Industrial Engineering and Management programmes at the University of Groningen.