Fuzzy multiattribute grey related analysis using DEA

  • Authors:
  • Desheng Dash Wu;David L. Olson

  • Affiliations:
  • RiskLab, University of Toronto, 105 St. George Street, Toronto, Canada M5S 3E6;Department of Management, University of Nebraska, Lincoln, NE 68588-0491, United States

  • Venue:
  • Computers & Mathematics with Applications
  • Year:
  • 2010

Quantified Score

Hi-index 0.09

Visualization

Abstract

Qualitative input is encountered in many real decision-making domains. Often such domains include tradeoffs among multiple attributes, and estimates of parameters are often expressed with some degree of uncertainty. Grey related analysis has been proposed as a means to use interval fuzzy representation of data. When dealing with multiple criteria data, model parameters that can involve uncertainty include both performances of alternatives on attributes as well as attribute weights. DEA is proposed as an objective way to derive weights. This paper presents a grey-related fuzzy set methodology incorporating data envelopment analysis as a way to more objectively rank alternatives. The purpose is to demonstrate the method. The focus is on identifying alternatives performing the most efficiently with respect to the decision maker's preference. The method is demonstrated on a multiattribute siting problem. Simulation is applied to validate model efficiency.