EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An m-Way Pythagorean Fuzzy Decision-Theoretic Rough Set Model with Admissible Loss Tables

Muhammad Imran Shahid, Muhammad Tahir, Sami Ullah Khan

Journal of Contemporary Decision Science · 26.10.2026 · 3 (1) · 1-22

DOI: 10.67334/cds31202648 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Contemporary Decision Science, online erschienen am 26.10.2026. Verfasst von Muhammad Imran Shahid und 2 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Decision-theoretic rough set models for Pythagorean fuzzy information stop at three outcomes, although credit, purchasing and clinical decisions are routinely graded into more. We develop an m-way Pythagorean fuzzy decision-theoretic rough set model for any m ≥ 2, in which a power-weighted conditional degree is turned into m ordered regions, a chain of graded probabilistic approximations, by minimum expected loss. Using the risk-set geometry of Bayesian decisions we show that all m actions survive exactly when the loss points form a strictly convex chain; otherwise some regions vanish, and the common repair of separating tied thresholds is not Bayes consistent. A two-exponent loss family, specified by a cost ratio and two risk coefficients, guarantees a genuine m-way partition whenever both coefficients are at most one half (and not both equal to it), contains the standard three-way table as a special case, and has a deferral region that shrinks strictly as the coefficient grows. A bound on the loss of offering only m actions gives a tolerance rule for choosing m. On the German credit data, under the published cost matrix and repeated cross-validation, the calibrated five-way model cut the realised cost by 6.3% relative to three-way and 19.9% relative to two-way decisions, while seven regions gave no further gain. A supplier classification example, Monte Carlo perturbation and Sobol' indices show that the risk coefficient and the cost ratio, not the attribute weights, drive the classification.

Zitieren

APA

Shahid, M. I., Tahir, M., & Khan, S. U. (2026). An m-Way Pythagorean Fuzzy Decision-Theoretic Rough Set Model with Admissible Loss Tables. Journal of Contemporary Decision Science, 3(1), 1-22. https://doi.org/10.67334/cds31202648

Vancouver

Shahid MI, Tahir M, Khan SU. An m-Way Pythagorean Fuzzy Decision-Theoretic Rough Set Model with Admissible Loss Tables. Journal of Contemporary Decision Science. 2026;3(1):1-22. doi:10.67334/cds31202648

BibTeX
@article{Shahid2026,
  author = {Shahid, Muhammad Imran and Tahir, Muhammad and Khan, Sami Ullah},
  title = {An m-Way Pythagorean Fuzzy Decision-Theoretic Rough Set Model with Admissible Loss Tables},
  journal = {Journal of Contemporary Decision Science},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {1--22},
  doi = {10.67334/cds31202648}
}
Kostenlos starten →