1156. Hierarchical Dominance-based Rough Set Approach for Evaluating Omnichannel Retailing Strategies
Invited abstract in session TA-8: Hierarchical MCDA and MCDA in circular economy, stream Multiple Criteria Decision Aiding.
Tuesday, 8:30-10:00Room: Clarendon SR 2.08
Authors (first author is the speaker)
| 1. | Weronika Mrozek
|
| Poznan University of Technology | |
| 2. | Milosz Kadzinski
|
| Institute of Computing Science, Poznan University of Technology | |
| 3. | Mladen Stamenković
|
| University of Belgrade - Faculty of Economics | |
| 4. | Aleksa Dokić
|
| Department for Business Economics and Management, University of Belgrade-Faculty of Economics |
Abstract
We consider Dominance-based Rough Set Approach (DRSA) incorporating criteria hierarchy. The method enables the partial disaggregation of a base set of criteria into multiple levels, ranging from specific to general. The classification for a given node in the hierarchy is derived based on the outcomes attained or observed only for its child nodes. This approach is advantageous from both structuring and data processing perspectives. We applied the method to a dataset of 74 Serbian retailers' multichannel marketing integration, comprising 17 criteria organized into four hierarchical groups. We compared it against the classical DRSA with a flat criteria structure, using two rule induction algorithms, DOMLEM and DOMApriori, alongside two classification schemes — standard and score-based. The hierarchical DRSA demonstrated competitive accuracy in cross-validation, with the combination of DOMLEM and standard classification yielding the best performance. The results suggest that the inconsistency in top-level criteria values may impact classification quality. However, further improvements in data preprocessing could enhance the method's accuracy, making hierarchical DRSA a promising tool for addressing decision problems with complex criteria structures.
Keywords
- Decision Support Systems
- Decision Analysis
Status: accepted
Back to the list of papers