How to process the random part of RFVs: comparison of available methods and new proposal

In the recent years, fuzzy variables (FVs) and random-fuzzy variables (RFVs) have been proposed to represent the measurement results with their associated uncertainty. However, up to now, the different authors do not yet agree in the mathematical way FVs should be composed together, so different approaches have been proposed. This paper compares these approaches to find their advantages and disadvantages and shows a new proposal that is supposed to hopefully overcome the disadvantages of the original approaches.