Evaluations of corporate sustainability indicators based on fuzzy similarity graphs

Evaluations of corporate sustainability indicators based on fuzzy similarity graphs

WoS Article

The paper deals with inconsistencies of composite sustainability indicators and their different subsets (economic, environmental, social, and corporate governance). Corporate sustainability performance is usually highly nonlinear, vague, partially inconsistent and multidimensional. The resulting models are often oversimplified. The key reason is an information shortage which eliminates the unsophisticated applications of classical statistical methods. Numbers are accurate and information intensive. Verbal quantifications are less accurate and therefore not that information intensive. Fuzzy sets and fuzzy reasoning are used to make verbal quantifiers suitable for computer applications. A fuzzy similarity graph is defined. A team of experts identified 17 relevant variables (e.g. Environmental costs, Occupational diseases, Number of complaints received from stakeholders) and 12 company data sets are available. Each company is presented as a fuzzy conditional statement. A set of fuzzy pairwise similarities is generated and used to evaluate five similarity graphs: a Total Graph (based on all 17 variables) and graphs based on relevant specific subsets of variables, Economic, Environmental, Social and Corporate Governance graphs. The topologies of these graphs are significantly different. No prior knowledge of fuzzy reasoning is required.

The paper deals with inconsistencies of composite sustainability indicators and their different subsets (economic, environmental, social, and corporate governance). Corporate sustainability performance is usually highly nonlinear, vague, partially inconsistent and multidimensional. The resulting models are often oversimplified. The key reason is an information shortage which eliminates the unsophisticated applications of classical statistical methods. Numbers are accurate and information intensive. Verbal quantifications are less accurate and therefore not that information intensive. Fuzzy sets and fuzzy reasoning are used to make verbal quantifiers suitable for computer applications. A fuzzy similarity graph is defined. A team of experts identified 17 relevant variables (e.g. Environmental costs, Occupational diseases, Number of complaints received from stakeholders) and 12 company data sets are available. Each company is presented as a fuzzy conditional statement. A set of fuzzy pairwise similarities is generated and used to evaluate five similarity graphs: a Total Graph (based on all 17 variables) and graphs based on relevant specific subsets of variables, Economic, Environmental, Social and Corporate Governance graphs. The topologies of these graphs are significantly different. No prior knowledge of fuzzy reasoning is required.

Corporate sustainability; Fuzzy; Indicators; Similarity graph

Corporate sustainability; Fuzzy; Indicators; Similarity graph

PAVLÁKOVÁ DOČEKALOVÁ, M.; DOUBRAVSKÝ, K.; DOHNAL, M.; KOCMANOVÁ, A.

2018

19.03.2017

Elsevier

Nizozemsko

1470-160X

Ecological Indicators

2017

78

Kingdom of the Netherlands

108

114

7

https://www.sciencedirect.com/science/article/abs/pii/S1470160X17301012?via%3Dihub

http://hdl.handle.net/