Premium
A cumulative belief degree approach for group decision‐making problems with heterogeneous information
Author(s) -
Ervural Bilal,
Kabak Özgür
Publication year - 2019
Publication title -
expert systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 38
eISSN - 1468-0394
pISSN - 0266-4720
DOI - 10.1111/exsy.12458
Subject(s) - computer science , preference , group decision making , representation (politics) , set (abstract data type) , aggregate (composite) , fuzzy set , fuzzy logic , degree (music) , data mining , tuple , group (periodic table) , artificial intelligence , machine learning , mathematics , statistics , materials science , physics , discrete mathematics , politics , political science , acoustics , law , composite material , programming language , chemistry , organic chemistry
In some complex group decision‐making (GDM) problems, the information needing to be processed may be heterogeneous. This may involve consideration of objective and subjective criteria by experts who have their own particular set of criteria, their own preference format for assessing alternatives under these criteria, and who may themselves be assigned differing importance weights as experts. This paper presents a cumulative belief degree approach to cope with heterogeneous information in multiple attribute GDM problems. The proposed approach focuses to aggregate subjective expert assessments and objective criteria that are presented in various representation formats and scales. The methodology employs transformation formulae for several preference representation scales to belief structure, including 2‐tuple representation, classical fuzzy sets, hesitant fuzzy sets, and intuitionistic fuzzy sets. Aggregation formulae are proposed to combine expert criteria evaluations and find a collective preference. A consensus degree is calculated for measuring the agreement between the experts. An illustrative example is presented to clarify the steps of the methodology, and validity of the approach is assured through comparative analysis with the existing methods.