The Evolution of Multiple Criteria Decision Aiding under the Influence of the Artificial Intelligence Paradigm

Abstract: The old challenge of Operational Research was how to make better decisions based on optimization techniques. In recent years, the abundance of data about human choices changed the paradigm of Operational Research from ‘optimization’ to ‘analytics’. Furthermore, operational research users bacame increasingly aware that realistic decision support requires considering multiple conflicting criteria. ‘Optimum’ has thus been replaced by ‘best compromise’ determined by preferences of Decision Makers (DMs). For the development of personalized computing, the concept of preference has also become relevant for Machine Learning and Artificial Intelligence able to analyze vast amounts of user data to make predictions and recommendations. Preferences provide a means for specifying desires in a declarative and intelligible way, a key element for the effective representation of knowledge and reasoning respecting the value systems of DMs [1].

We present a constructive preference learning methodology, called robust ordinal regression (ROR), for multiple criteria decision aiding [2,3]. This methodology links Operational Research with Artificial intelligence, and as such, it confirms the current trend in mutual relations between these disciplines.

Multiple criteria decision problems concern a set of actions considered in one of three settings: classification, ranking, or choice, with multiobjective optimization constituting a particular case of choice. To identify a best-compromise solution to these problems, decision aiding methods require preference information reflecting the value system of one or more DMs. In ROR, the preference information has the form of training examples. They may either be provided by the DM on a set of real or hypothetical actions, called reference actions, or may come from observation of DM’s past decisions. This information is used to build a preference model, which is then applied on a non-dominated set of possible actions to arrive at a recommendation presented to the DM(s). In practical decision aiding, the process composed of preference elicitation, preference modeling, and DM’s analysis of a recommendation, loops until the DM (or a group of DMs) accepts the recommendation or decides to change the problem setting. Such an interactive process is called constructive preference learning. We describe this process for three types of preference models: (i) utility functions, (ii) outranking relations, and (iii) sets of monotonic decision rules.

References

• Hüllermeier, E., Słowiński, R.: Preference learning and multiple criteria decision aiding: differences, commonalities, and synergies – part part I. 4OR – A Quarterly Journal of Operations Research, 22 (2024) 179–209, https://doi.org/10.1007/s10288-023-00560-6, and part II. 4OR, 22 (2024) 313-349, https://doi.org/10.1007/s10288-023-00561-5
• Corrente, S., Greco, S., Kadziński, M., Słowiński, R.: Robust ordinal regression in preference learning and ranking. Machine Learning, 93 (2013) 381-422, https://link.springer.com/article/10.1007/s10994-013-5365-4
• Greco, S., Słowiński, R., Wallenius J.: Fifty years of multiple criteria decision analysis: From classical methods to robust ordinal regression. European Journal of Operational Research, 323 (2025) 351-377, https://doi.org/10.1016/j.ejor.2024.07.038

Roman Słowiński, a full member of the Polish Academy of Sciences, is a Professor of the Laboratory of Intelligent Decision Support Systems at the Institute of Computing Science, Poznań University of Technology, Poland and a Professor at the Systems Research Institute of the Polish Academy of Sciences in Warsaw.

Roman Słowiński has conducted extensive research on methodology and techniques of intelligent decision support, combining Operational Research and Artificial Intelligence. His interests are particularly focused on: addressing robustness in decision analysis; mining ordinal data; preference learning; and modeling uncertainty and imprecision in decision problems. He has also made significant contributions to rough set theory, fuzzy set theory, and evolutionary multiobjective optimization. His work extends to project scheduling, which encompasses multiple category resources, job modes, multiple criteria, and uncertainties. Furthermore, he applies his decision support expertise in medicine, technology, economics, and environmental studies.

He has published 16 books and more than 550 articles, including 350 papers in major scientific journals

Roman Słowiński is highly ranked in the Stanford University World’s TOP 2% scientists of the World. He is no. 413 in the world, and no. 1 in Poland.

Roman Słowiński is recipient of the EURO Gold Medal (1991), and Doctor Honoris Causa of six universities: Faculté Polytechnique de Mons (2000), Université Paris Dauphine (2001), Technical University of Crete (2008), Nanjing University of Aeronautics and Astronautics (2018),  Hellenic Mediterranean University in Heraklion (2022), and University of West Attica in Athens (2024).

Roman Słowiński was honored with the Scientific Award of the Prime Minister of Poland for creating
a scientific school of Intelligent Decision Support Systems. In 2021, he got the Richard Price Award in Data Science from the International Academy of Information Technology and Quantitative Management. In 2022, he received the Humboldt Research Award by Alexander von Humboldt Foundation.

He was awarded Fellow grade by IEEE (the Institute of Electrical and Electronics Engineers – 2017), IRSS (International Rough Set Society – 2015, INFORMS (the Institute for Operations Research and the Management Sciences – 2019), IFIP (International Federation for Information Processing – 2019), IFORS (International Federation of Operational Research Societies – 2022), IAITQM (International Academy of Information Technology and Quantitative Management – 2022), AAIA (Asia-Pacific Artificial Intelligence Association – 2022), and AIIA (International Artificial Intelligence Industry Alliance – 2024.