Combining DRSA decision-rules with FCA-based DANP evaluation for financial performance improvements
Combining DRSA decision-rules with FCA-based DANP evaluation for financial performance improvements
Author(s): Kao-Yi Shen, Gwo-Hshiung TzengSubject(s): Economy, Socio-Economic Research
Published by: Vilnius Gediminas Technical University
Keywords: dominance-based rough set approach (DRSA); formal concept analysis (FCA); decision making trial and evaluation laboratory (DEMATEL); multiple criteria decision making (MCDM); DEMATEL-based ANP (DANP);
Summary/Abstract: This study proposes a combined method to integrate soft computing techniques and multiple criteria decision making (MCDM) methods to guide semiconductor companies to improve financial performance (FP) – based on logical reasoning. The complex and imprecise patterns of FP changes are explored by dominance-based rough set approach (DRSA) to find decision rules associated with FP changes. Companies may identify its underperformed criterion (gap) to conduct formal concept analysis (FCA) – by implication rules – to explore the source criteria regarding the underperformed gap. The source criteria are analysed by decision making trial and evaluation laboratory (DEMATEL) technique to explore the cause-effect relationship among the source criteria for guiding improvements; in the next, DEMATEL-based analytical network process (DANP) can provide the influential weights to form an evaluation model, to select or rank improvement plans. To illustrate the proposed method, the financial data of a real semiconductor company is used as an example to show the involved processes: from performance gaps identification to the selection of five assumed improvement plans. Moreover, the obtained implication rules can integrate with DEMATEL analysis to explore directional influences among the critical criteria, which may provide rich insights and managerial implications in practice.
Journal: Technological and Economic Development of Economy
- Issue Year: 22/2016
- Issue No: 5
- Page Range: 685-714
- Page Count: 30
- Language: English