Data Envelopment Analysis with Z-Numbers - An Application to Project Selection

Data Envelopment Analysis in its fuzzy version is used to rank projects on the basis of their, in the general case, fuzzy inputs and outputs. The possibility of a biased evaluation of the inputs and outputs is taken into account. To eliminate bias influence, Z-numbers are applied. Although there exist attempts to integrate them in Data Envelopment Analysis, the known approaches are too simplistic, as they do not distinguish between biases concerning project inputs from those concerning project outputs. In this paper we propose a modified (addressing the above issue) Data Envelopment Analysis model with project inputs and outputs represented by Z-numbers and show its usefulness in the area of project portfolio management. The approach is illustrated by means of an example.