RELATIONSHIPS BETWEEN DATA ENVELOPMENT ANALYSIS AND MULTICRITERIA DECISION-ANALYSIS

We contrast the different approaches of Data Envelopment Analysis (DEA ) and Multiple Criteria Decision Making (MCDM) to superficially simila r problems. The concepts of efficiency and Pareto optimality in DEA an d MCDM are compared, and a link is demonstrated between the ratio effi ciency definition in DEA and a distance measure in input-output space based on linear value functions. The problem of weight sensitivity is discussed in terms of value measurement theory, highlighting the assum ptions needed during model formulation in order to justify the use of value judgements to constrain weight flexibility in DEA. Finally, we p ropose a stochastic approach, in which a probability distribution on e fficiencies can be derived for each decision making unit, as a basis f or comparison.