Vassilis M. Charitopoulos - Uncertainty-aware Integration of Control with Process Operations and Multi-parametric Programming Under Global Uncertainty

Aims and Scope

The series Springer Theses brings together a selection of the very best Ph.D. theses from around the world and across the physical sciences. Nominated and endorsed by two recognized specialists, each published volume has been selected for its scientific excellence and the high impact of its contents for the pertinent field of research. For greater accessibility to non-specialists, the published versions include an extended introduction, as well as a foreword by the students supervisor explaining the special relevance of the work for the field. As a whole, the series will provide a valuable resource both for newcomers to the research fields described, and for other scientists seeking detailed background information on special questions. Finally, it provides an accredited documentation of the valuable contributions made by todays younger generation of scientists.

More information about this series at http://www.springer.com/series/8790

This Springer imprint is published by the registered company Springer Nature Switzerland AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Vassilis research has focused on developing novel methods for mitigating the impact of uncertainty for optimal decision-making in process and energy systems engineering problems. The unique advances presented in this Thesis signal a major step towards the systematic treatment of uncertainty for highly interconnected industrial and energy systems that are emerging currently and in the foreseeable future. Uncertainty is ubiquitous in decision-making problems and in the face of the constantly increasing sustainability and economic considerations, the research outcomes of the present Thesis are of paramount importance. The work comprises key algorithmic developments related to the fields of process systems engineering, operations research, control engineering and decision-making under uncertainty and its timeliness and relevance is underlined by the Industry 4.0 inspired applications demonstrated within the thesis. In the first part of the thesis, novel computer algebra-based algorithms are presented for the solution of (mixed integer) optimisation problems which are susceptible to global uncertainty from a multi-parametric programming viewpoint. For the first time, theorems are established for the mathematical nature of the explicit solutions in multi-parametric linear programs under global uncertainty which provide transformative insights for mitigating uncertainty considerations in optimisation problems. The proposed algorithms are tested and showcased in a rich portfolio of flagship process systems engineering case studies such as energy planning, process scheduling and process design and model-based control. In the second part of thesis, the problem of integrating process planning, scheduling and control is examined and a rigorous mathematical framework for its uncertainty-aware solution is presented. Integrating control with operations is central to Industry 4.0 and smart manufacturing considerations. Complex, multi-scale, model-based optimisation problem is formulated and solved. It is showcased how by integrating control decisions with other operational levels, manufacturing processes can become more flexible, responsive and resilient to externalities. Using state-of-the-art chance-constrained programming and robust optimisation techniques a novel hybrid-framework is developed and applied to chemical industry case studies. The research work has been presented in six journal publications, several international conference proceedings and presentations. For his scientific excellence, Vassilis has received several awards and distinctions including the: UCL Dennis Rooke Prize (2016), UCL Davids Newton Prize (2019) and he has been selected as one of the best young researchers, globally, by the Institute of Chemical Engineers (2018;2019).

Advances in computing power along with the widespread availability of data have led process industries to consider a new paradigm for automated and more efficient operations. This trend is commonly referred to as Industry 4.0 and many reports indicate that it can result in cost reductions of up to 20% while improving asset utilisation by 3040%. Despite the research effort dedicated to techniques for the interpretation and efficient employment of big-data within the industry, the development of mathematical modelling and optimisation techniques that serve the requirements for seamless communication among the different levels of decision making has received considerably less attention. McKinsey Global Institute estimates that operational optimisation within the Industry 4.0 can generate up to $3.7T of value in 2025. It is the goal of this project to present models and methodologies that can be employed towards the realisation of the Industry 4.0 vision within the process industries while at the same time investigate the impact of uncertainties in such highly integrated settings.