Multiple-criteria performance ranking based on profile distributions: An application to university research evaluations

Kunsch, Pierre Louis; Ishizaka, Alessio

doi:10.1016/j.matcom.2018.05.021

Cited by 17 publications

(7 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For each surrogate Table 1 , where a bar on top of a variable indicates its average value; this property means that the simplex-centroid vector is the average weight vector of the simplex. This is wellknown for EW, and it has been first recognised for the ROC weights in (Barron 1992); the RS centroid-property as being the average weight vector for the uniform distribution in the corresponding surrogate simplex has been established later in (Jia & al., 1998) based on orderstatistics arguments (Arnold & al., 1992); (Kunsch & Ishizaka, 2018); (Kunsch & Brans, 2019) but it is not mentioned by later authors. The centroid vector is located at the barycentre or centre of gravity (COG) of each simplex.…”

Section: Comparison Experimentsmentioning

confidence: 96%

A note on using centroid weights in additive multi-criteria decision analysis

Kunsch

Ishizaka

2019

European Journal of Operational Research

Self Cite

View full text Add to dashboard Cite

Section: Comparison Experimentsmentioning

confidence: 96%

A note on using centroid weights in additive multi-criteria decision analysis

Kunsch

Ishizaka

2019

European Journal of Operational Research

Self Cite

View full text Add to dashboard Cite

“…However,

c_{1}

has a lower variance than

c_{2}

, indicating a higher level of mutual consent. To take it into account, we define a profile distribution table (Kunsch, 2018):…”

Section: A Novel Criteria Preprocessing Frameworkmentioning

confidence: 99%

Criteria preprocessing in multi‐actor multi‐criteria analysis

Huang

Canoy

Brusselaers

et al. 2023

Multi Criteria Decision Anal

View full text Add to dashboard Cite

Multi‐actor multi‐criteria analysis is a group decision‐making framework that allows multiple stakeholder groups to be involved in the decision‐making process, facilitating the understanding of the points of consensus and conflict among the stakeholder groups. Carefully selecting suitable criteria is important as they illustrate the possibly divergent priorities of the respective stakeholder group, and overlooking important criteria can lead to erroneous outcomes. Furthermore, the number of criteria needs specific consideration, as a too large number poses problems for human cognition, but a too small number inaccurately represents the stakeholder's interest. In stakeholder groups with many members, such as those representing citizens, defining a criteria set is likely to be even more complicated. Currently, there is no formal guideline to assist facilitators in defining these criteria sets. In this paper, we propose a novel framework for criteria preprocessing with stakeholder involvement that includes a guideline for firstly selecting criteria into a tentative list and secondly selecting the final criteria set. It provides a procedure on how to determine criteria considering the priorities of stakeholder groups with regard to the context. As a final step, we propose a mathematical model for selecting a number of criteria that are both cognitively manageable and representative for the participants' priorities. Based on the principles of the Pareto analysis, as well as the cognitive judgment theory “magic number seven plus or minus two”, a recommendation list of the criteria is generated. It prevents key criteria from being omitted while at the same time limiting the overall number of criteria. This framework is applied to a social decision‐making case for construction logistics, and the results are compared with the conventional approach of criteria definition.

show abstract

“…In the survey, the decision-maker can also ask questions on their socio-economic profiles for later research. The Profile Ranking with Order Statistics Evaluations (PROSE) is applied for the evaluation [20]. This approach combines MCDA, voting theory.…”

Section: Mass-participation Decision-making In Mamcamentioning

confidence: 99%

“…This method applies a weighted sum approach based on order statistics to combine the individual profile distribution. It is well suitable for mass-participation evaluation, as it does not considers only the mean distribution values, but also standard deviations [20].…”

Section: Weight Allocation and Alternative Evaluationmentioning

confidence: 99%

The Multi-Actor Multi-Criteria Analysis (MAMCA) for Mass-Participation Decision Making

Huang

Mommens

Lebeau

et al. 2021

Lecture Notes in Business Information Processing

View full text Add to dashboard Cite

The Multi-Actor Multi-Criteria Analysis is a methodology that allows for the involvement of multiple stakeholders within a decision-making process. It reveals the consensus and conflicts between the different groups of people that are involved in the evaluation but hold different interests. Nowadays, the concept of the "stakeholder" in MAMCA gradually shifts to the "stakeholder group", and there is a need for involving more than one evaluator in the stakeholder group to make sure all the voices from the group will be heard instead of being represented by one. Especially when a stakeholder group contains a large variation in interests, concerns and socio-economic characteristics. Additionally, one group can have subgroups that might be hard to reach, and therefore are not or un-der-represented in the analysis. This is typically the case for the 'citizens' stakeholder group. In order to fulfill the needs of the involvement of many different stakeholders within stakeholder groups, the mass-participation function was developed in MAMCA and the MAMCA survey tool is designed. This tool allows the decision-maker to design the dedicated survey for the stakeholder group which needs the mass-participation function. The easy-to-understand evaluation process is used to avoid time-consuming elicitation. It is possible to check the homogeneity and heterogeneity of the stakeholders within the stakeholder group based on the socio-economic profiles collected in the survey.

show abstract

Multiple-criteria performance ranking based on profile distributions: An application to university research evaluations

Cited by 17 publications

References 45 publications

A note on using centroid weights in additive multi-criteria decision analysis

A note on using centroid weights in additive multi-criteria decision analysis

Criteria preprocessing in multi‐actor multi‐criteria analysis

The Multi-Actor Multi-Criteria Analysis (MAMCA) for Mass-Participation Decision Making

Contact Info

Product

Resources

About