EN
Multidimensional analysis uses various measures for assessing economic effects. However, no single synthetic measure, regardless how popular, can give a satisfactory solution to the above problem. In general, various approaches of combining measures can lead to stable outcomes. Nevertheless, when combining "weak" classifiers one can obtain inevitably poorer classification. We propose here a new approach to construct doubly synthetic measures. The main goal of this work is to analyse the influence these new synthetic measures on the ranking of multidimensional objects.