EUVIMEDEUVIMED entdecken
Crossref

A dynamic interval type-2 fuzzy TOPSIS with hybrid adaptive weighting and Q-Learning for real-time healthcare decision-making

Laxminarayan Sahoo, Sumanta Lal Ghosh

Healthcare Engineering · 2027 · 3 (3) · 143-158

DOI: 10.5267/j.ijdns.2026.10 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Healthcare Engineering, 2027. Verfasst von Laxminarayan Sahoo und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 1 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The process of making decisions in today’s intelligent systems, including smart healthcare, wireless sensor networks, Internet of Things (IoT), and cyber-physical systems, must occur in a condition of uncertainty and ambiguity as well as continuous change; however, traditional multi-criteria decision-making (MCDM) techniques, like TOPSIS, AHP, and VIKOR, use static matrices and static criteria weights. In this study, the concept of the Deep Reinforcement Learning-Assisted Dynamic Interval Type-2 Fuzzy TOPSIS (DRL-DIT2F-TOPSIS) model is introduced by considering the following techniques: the usage of Dynamic Interval Type-2 Fuzzy Sets for dealing with uncertainties; Dynamic Entropy Weighting and Adaptive Fuzzy Weighting for time-dependent criterion weights; Interval Type-2 Fuzzy AHP for hierarchical reasoning of the experts’ judgment; Deep Reinforcement Learning for autonomous reward-based policy optimization; and Explainable Artificial Intelligence (XAI) for ranking decision explanation. The proposed approach's efficiency is demonstrated through a case study of intelligent healthcare monitoring, using a smart dynamic patient prioritization method based on several physiological risk criteria. Statistical metrics (mean closeness coefficient, standard deviation, and Spearman correlation) and visual analysis methods (weight dynamics, PCA, sensitivity analysis, and reward convergence) support the effectiveness. The experiments showed that the introduced DRL-Hybrid approach outperforms classical approaches such as TOPSIS, entropy-weighted, AHP-weighted, and static hybrid alternatives by approximately 46.8% and 11%, respectively. The research lays the groundwork for future research in intelligent decision support systems.

Zitieren

APA

Sahoo, L., & Ghosh, S. L. (2027). A dynamic interval type-2 fuzzy TOPSIS with hybrid adaptive weighting and Q-Learning for real-time healthcare decision-making. Healthcare Engineering, 3(3), 143-158. https://doi.org/10.5267/j.ijdns.2026.10

Vancouver

Sahoo L, Ghosh SL. A dynamic interval type-2 fuzzy TOPSIS with hybrid adaptive weighting and Q-Learning for real-time healthcare decision-making. Healthcare Engineering. 2027;3(3):143-158. doi:10.5267/j.ijdns.2026.10

BibTeX
@article{Sahoo2027,
  author = {Sahoo, Laxminarayan and Ghosh, Sumanta Lal},
  title = {A dynamic interval type-2 fuzzy TOPSIS with hybrid adaptive weighting and Q-Learning for real-time healthcare decision-making},
  journal = {Healthcare Engineering},
  year = {2027},
  volume = {3},
  number = {3},
  pages = {143--158},
  doi = {10.5267/j.ijdns.2026.10}
}
Kostenlos starten →