Premium
Sensitivity analyses and measurements for group decisions using weighted scoring rules
Author(s) -
Zhao Yong,
Wu Fan,
Chen Yang
Publication year - 2021
Publication title -
international transactions in operational research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.032
H-Index - 52
eISSN - 1475-3995
pISSN - 0969-6016
DOI - 10.1111/itor.12588
Subject(s) - sensitivity (control systems) , group (periodic table) , invariant (physics) , group decision making , ordinal scale , scale (ratio) , computer science , mathematics , artificial intelligence , econometrics , statistics , social psychology , psychology , geography , engineering , cartography , chemistry , organic chemistry , electronic engineering , mathematical physics
Weighted scoring rules (WSRs) are usually used for solving problems of group decisions based on individual ordinal preferences. Generally, individuals are influenced by some subjective and objective factors, which lead to the mutability of individual preferences in group decisions. Therefore, while getting optimal solutions, decision makers often need to make qualitative or quantitative sensitivity analyses of their decisions or choices, and to evaluate reliabilities of the solutions. First, by using inclined angles of score vectors, this paper proposes a sensitivity measuring method and some related concepts of group decision using WSRs. Then, a calculation method is presented, and a size of a minimum disturbance group is discussed in depth because this size is also a useful index to evaluate sensitivities or reliabilities of the decisions. Next, we investigate and discuss the sensitivities of WSRs via their weights, and obtain some suggestions about designing WSRs. Finally, we use the sensitivity degree to analyze a controversy of the National Basketball Association Most Valuable Player evaluation. Our measuring methods are: well visualized, scale invariant, commensurable, and calculability.