An efficient midpoint-radius representation format to deal with symmetric fuzzy numbers

This paper proposes a novel representation for symmetric fuzzy numbers that uses the midpoint-radius approach instead of the conventional lower-upper representation. A theoretical analysis based on the α-cut concept shows that the proposed format requires half the amount of operations and memory than the traditional one. Also, a novel technique involving radius increments is introduced, to mitigate floating-point rounding errors when using the proposed representation. We describe the implementation of all these features into a fuzzy arithmetic library, specifically tuned to run on Graphic Processing Units (GPU). The results of a series of tests using compute-bound and memory-bound benchmarks, show that the proposed format provides a performance gain of two to twenty over the traditional one. Finally, several implementation issues regarding GPU are discussed in light of these results