Features selection based on fuzzy entropy for data envelopment analysis applied to transport systems

Recently, great attention has been paid to the data envelopment analysis (DEA) for the analysis of efficiency of transportation systems. In real world applications, the data of production processes cannot be precisely measured or can be affected by ambiguity. This is particularly worrying when assessing efficiency with frontier-type models, such as Data Envelopment Analysis (DEA) models, since they are very sensitive to possible data errors. Many research works have faced the problem of using DEA models when the inputs and outputs are uncertain. Fuzzy Theory based methods are one of the approaches that have been recently proposed even without a determined (or unique) framework. In this work we have defined a fuzzy version of the classical DEA models, and, in particular, a feature selection analysis has been developed to investigate the effects of uncertainty on the efficiency of the considered transport services. The feature selection method developed in this paper is based on fuzzy entropy measures and it can be applied to DMUs (Decision Making Units) on the entire frontier. Having identified the efficient and inefficient DMUs in fuzzy DEA analysis, the focus is on the stability of classification of DMUs into efficient and inefficient performers. A numerical example is then presented, considering as DMUs a set of international container ports with given number of inputs and outputs properly modified.