Sustainable Medical Supplier Selection Based on Multi-Granularity Probabilistic Linguistic Term Sets

Abstract

The sustainable medical supplier selection (SMSS) is an important issue facing the medical industry in the context of sustainable development, which can be regarded as a typical multi-attribute group decision making (MAGDM) problem. In the MAGDM process, linguistic term set (LTS) is particularly natural and convenient for decision makers (DMs) to express evaluation information. Especially, probabilistic linguistic term set (PLTS) is a very critical and effective tool, which can reflect the importance of different linguistic terms. Due to the different preferences and experience of different DMs, they may use multi-granularity probabilistic linguistic term sets (MGPLTSs) to represent different linguistic information. In this article, in order to study the comparison method of MGPLTSs, a new possibility degree formula is firstly proposed and its properties is proved. Then, in order to build a weight model, a possibility degree-based Best-Worst method (BWM) and a probability degree based-maximizing deviation method are established to calculate the subjective weights and objective weights of attributes, respectively. Where after, a MAGDM method is proposed by combining the ELimination Et Choix Traduisant la REalite (ELECTRE) method with Evaluation based on Distance from Average Solution (EDAS) method in the multi-granularity probabilistic linguistic information environment. Finally, the created MAGDM method is applied to the SMSS, and its effectiveness and advantages compared with other existing methods are verified.

RETRACTED: Online review–driven Product Recommendation Ranking Considering Interactions Among Features and Consumers’ Psychology

“…The possibility degree of PLTS is an indicator to measure the possibility that one PLTS is superior to the other PLTS, based on which the ranking of a list of PLTSs can be further obtained. Compared with the probability degree formulas of Bai et al [41], Chen et al [42], and Mao et al [43], Liu et al [44] considered more comprehensive information between two PLTSs, avoiding loss of information, and provided a more accurate method for comparing calculation. Scale linguistic functions are incorporated to obtain a flexible probabilistic linguistic possible degree formula:…”

A State-of-the-Art Review of the BWM Method and Future Research Agenda

“…■ To handle problems with uncertain, incomplete, and ambiguous information, researchers improved BWM with ordinary fuzzy sets (Ahmad et al, 2023), interval type-2 fuzzy sets (Chen et al, 2022;Aycin et al, 2022), hesitant fuzzy sets (Karbassi Yazdi et al, 2023), hesitant fuzzy linguistic term sets (Liang et al, 2022;Liao et al, 2019), probabilistic linguistic term sets (Xian et al, 2023;Xu et al, 2022;Liu et al, 2022), q-rung orthopair fuzzy sets (Xiao et al, 2022), Z numbers (Abbasi Kamardi et al, 2022), D numbers (Navaei et al, 2023, Bayesian theory (Hashemkhani Zolfani et al, 2022), and rough sets (Huang et al, 2022).…”