1677. An Explainable Composite Indicator Based on Decision Rules
Invited abstract in session MC-8: Preference Learning 2, stream Multiple Criteria Decision Aiding.
Monday, 12:30-14:00Room: Clarendon SR 2.08
Authors (first author is the speaker)
| 1. | Silvano Zappalà
|
| Department of Economics and Business, University of Catania | |
| 2. | Salvatore Corrente
|
| Department of Economics and Business, University of Catania | |
| 3. | Salvatore Greco
|
| Department of Economics and Business, University of Catania | |
| 4. | Roman Slowinski
|
| Institute of Computing Science, Poznan University of Technology |
Abstract
Composite indicators are widely employed in various fields to classify or rank alternatives evaluated on multiple criteria. Their construction falls within Multiple Criteria Decision Analysis (MCDA), where different methods are chosen based on key technical aspects, like measurement scales of criteria, degree of acceptable compensation between them, and potential interactions among criteria. However, beyond the final classification or score assigned to each alternative, ensuring result explainability and procedural transparency is crucial.
This paper proposes a method for constructing an explainable composite indicator using decision rules. The Decision Maker (DM) provides simple preference information, either in the form of classifications (e.g., alternative a is assigned to class 1, while b is assigned to class 2) or pairwise comparisons (e.g., a is preferred to b) on some reference alternatives. We then apply the Dominance-Based Rough Set Approach (DRSA) to derive decision rules that systematically explain these preferences by relating class assignments of reference alternatives with conditions concerning threshold performances on some selected criteria. These rules explicitly link the DM’s judgments to the performance of alternatives on subsets of criteria, clarifying the underlying rationale. Moreover, they serve as a basis to classify or rank new alternatives of interest.
To illustrate our approach applicability, we present a real-world decision-making case study.
Keywords
- Decision Support Systems
- Decision Analysis
Status: accepted
Back to the list of papers