The Neuro-Conceptual Approach to AI: When Deep Learning Meets Conceptual Modeling, Good Things Happen
Dov Dori, Enterprise Systems Modeling Laboratory, Technion - Israel Institute of Technology, Israel
OR for the Nanoworld: Insights into RNA Structure Modeling
Marta Szachniuk, Institute of Computing Science, Poznan University of Technology, Poland
Are Vehicle Routing Problems Truly Well-Known?
Telmo Miguel Pires Pinto, Faculty of Science and Technology, University of Coimbra, Portugal
The Neuro-Conceptual Approach to AI: When Deep Learning Meets Conceptual Modeling, Good Things Happen
Dov Dori
Enterprise Systems Modeling Laboratory, Technion - Israel Institute of Technology
Israel
![]()
Brief Bio
Dov Dori is a Pioneer and Fellow of INCOSE, Life Fellow of IEEE, and Fellow of IAPR and AAIA. He is Professor Emeritus of Systems Engineering and Head of the Enterprise Systems Modeling Laboratory at the Faculty of Data and Decision Sciences, Technion, Israel Institute of Technology. He was a visiting professor at MIT intermittently between 1999 and 2020. In 1993, Dr. Dori invented Object-Process Methodology, OPM, which has become ISO 19450:2024 International Standard. He wrote two books on OPM, which have been central to model-based systems engineering (MBSE) and provided the basis for the MBSE edX certificate program and MOOC series. Prof. Dori has supervised over 60 graduate students and authored 400 publications, cited over 8600 times. He chaired nine international conferences and was Associate Editor of IEEE T-PAMI and Systems Engineering. He was Co-Founder and Co-Chair of the IEEE Society of Systems, Men, and Cybernetics Technical Committee on MBSE. He has received various research, innovation, and teaching awards, including the INCOSE Pioneer Award and INCOSE Propeller Hat Award, and he is a member of Omega Alpha Association – International Honor Society for Systems Engineering.
Abstract
Generative AI has struck the world with its astonishing capabilities to automatically create intelligent text, stunning images, and captivating videos. However, due to their insufficient explainability, large language models (LLMs) cannot as yet be trusted for high-stakes, mission-critical applications in domains like healthcare, finance, and engineering. Conceptual modeling is an activity of formalizing scientific knowledge and specifying engineering systems. The problem with conceptual models is that they must be created by expert humans whose time and intellectual resources are scarce. In this talk, I introduce the neuro-conceptual approach to AI—a specialization of the neuro-symbolic approach, in which the symbolic component is a set of related Object-Process Methodology (OPM, ISO 19450:2024) models. This marriage enhances LLMs’ explainability while providing for automated OPM model construction, ushering in large-scale knowledge formalization and management.
OR for the Nanoworld: Insights into RNA Structure Modeling
Marta Szachniuk
Institute of Computing Science, Poznan University of Technology
Poland
![]()
Brief Bio
Marta Szachniuk is a Professor of Engineering and Technical Sciences at
the Institute of Bioorganic Chemistry, Polish Academy of Sciences, and the
Institute of Computing Science, Poznan University of Technology. After
training in computing science and mathematics, she earned her Ph.D. and
habilitation in computing science, specializing in bioinformatics. She is
the head of the Department of Structural Bioinformatics at IBCH PAS. Her
lab focuses on modeling and analyzing RNA structures, with a core part of
this research dedicated to developing models and computational methods for
RNA structure prediction and analysis.
These efforts culminate in RNApolis, a virtual lab for RNA research, which
now includes over 20 bioinformatics tools dedicated to the study of
nucleic acids. Its flagship tool, RNAComposer, is a world-renowned and
award-winning system for 3D RNA structure prediction. Szachniuk's group
actively participates in RNA 3D structure modeling contests, such as CASP
and RNA-Puzzles. In the first RNA-targeting competition within CASP, her
team achieved 3rd place, ranking as the best RNA structure prediction
group in Europe.
Abstract
In the complex machinery of life, RNAs act as interpreters, compilers, and
regulatory units. The roles of these molecules are directly connected to
their complex architectures, which are studied by structural
bioinformatics, an interdisciplinary area that bridges computation and
molecular biology. In this talk, I will present how techniques of
operations research are applied to RNA research revolutionizing our
understanding of its microscopic world. I will touch on various
representations of molecular data, predicting the 3D shapes from
sequence data, comparing RNA architectures, and assessing their
similarity. The talk will cover case studies highlighting the application
of OR methods in real-world scenarios, such as RNA-targeting
experiments like CASP and RNA-Puzzles.
Are Vehicle Routing Problems Truly Well-Known?
Telmo Miguel Pires Pinto
Faculty of Science and Technology, University of Coimbra
Portugal
![]()
Brief Bio
Telmo Pinto is an Assistant Professor in the Department of Mechanical Engineering at the Faculty of Science and Technology, University of Coimbra. He holds a PhD in Industrial and Systems Engineering from the University of Minho, with a thesis titled "Models and Advanced Optimization Algorithms for the Integrated Management of Logistics Operations”. His research activity is strongly focused on Operations Research & Management Science, as well as related areas such as logistics, production and operations management, and optimization. His scientific contributions include publications in indexed international journals and presentations at leading conferences in his field, as well as several book chapters and oral and panel communications. He has actively participated in academic and competition juries, review activities, editorial coordination of publications, and committees for scientific events. Currently he is member of the Editorial Board of “Boletim APDIO” (semi-annual magazine of Portuguese Operational Research Society).
Abstract
Vehicle Routing Problems (VRPs) are often regarded as canonical combinatorial optimization challenges, yet their ever-expanding array of variants and extensions suggests otherwise. This growth reflects the increasing complexity and evolving nature of these problems, driven by real-world factors such as dynamic traffic conditions, environmental constraints, and data uncertainties. Despite extensive academic attention, a significant gap remains between theoretical models and practical applications. This presentation delves into these complexities, challenging the perception that VRPs are "well-known". By exploring emerging paradigms such as integrated logistics, robust routing, and the environmental impact of routing, we highlight the limitations of classical optimization methods in addressing modern challenges. Ultimately, the aim is to invite the audience to rethink VRPs as a dynamic, multifaceted field that continues to evolve.