Nonparametric Multivariate Ranking Methods for Global Performance Indexes

The proper definition of a global performance index is a challenging topic, especially in the field of New Product Development and Education. Very often in this context, the research aim is focused on evaluating the performances of treatments (products, services, etc.) from a multivariate point of view, that is, in connection with more than one aspect and/or under several conditions. Therefore, the main goal of statistical data analysis consists in the calculation of a proper index to obtain a global performance evaluation of the treatments under investigation. The purpose of this work is to present an innovative nonparametric method for ranking of treatments, with reference to the analysis of variance layout, using a suitable global performance index, and to critically compare two challenging indexes that can be used in complex situations whereas parametric procedures can not be reliably employed. The goal of this paper is to find which procedure is more reliable. In particular, the procedures are tested by varying experimental conditions such as the number of variable and the distributions of random errors.