An integrated decision-making framework for prioritizing the barriers and applications in medical digital twins under hesitant fuzzy linguistic environment with stochastic probabilistic distribution

Vollständiger Abstract

Worum geht es in dieser Arbeit?

The rapid evolution of digital twins (DTs) technology is reshaping the medical service system. As medical DTs (MDTs) are themselves complex, existing decisionmaking framework often ignored decision-makers’ (DMs’) hesitant fuzzy preference information and subjective cognitive biases. This research develops an integrated decision-making framework for prioritizing the MDTs barriers and applications with the hesitant fuzzy linguistic term sets (HFLTSs) incorporating stochastic probabilistic information. Aligned with the taxonomy of distributed linguistic representations, we first introduce a generalized probabilistic distance metric tailored for HFLTSs. Second, an integrated decision-making framework is established. On the one hand, three different hesitant fuzzy linguistic best-worst method (HFL-BWM) models with stochastic probabilistic distribution are provided for MDTs barriers’ prioritization. On the other, we develop a hesitant fuzzy linguistic generalized TODIM (HFL-GTODIM) method that incorporates loss-aversion and the proposed distance measure for MDTs applications’ prioritization. Our method innovatively introduces a stochastic hesitant fuzzy linguistic distance metric that reduces aggregation-induced information loss while retaining interpretability. The proposed decision-making framework is validated on an MDT case derived from expert elicitation. Under the representative weighting model, the derived MDTs barriers’ weights are [0.3251, 0.2019, 0.1269, 0.1031, 0.0555, 0.1876], highlighting data privacy and security as the principal barrier. In the process of MDTs applications’ prioritization, surgical simulation and development attains the highest overall dominance degree of 0.5736, indicating a clear priority among alternatives. Sensitivity analysis clearly shows that different parameter values can influence the results of MDTs applications’ prioritization. DMs can address parameter issues by continuously collecting feedback and adjusting parameters to fit current decision requirements. These quantitative findings demonstrate that the proposed decision-making framework can preserve the integrity of uncertain and hesitant preference information, indicate stability in high-stakes healthcare environments, and provide actionable decision support for the related departments and unit.