Abstract
We discuss methods suitable for finding approximate solutions to fuzzy relational equations defined in finite universe of discourse and taking the form of either a system of equations
X
k
·
R =
Y
k
,
k = 1, 2,…,
K, (with
R being unknown fuzzy relation) or a single equation
X ·
R =
Y (with
X being looked for). The key idea of the proposed algorithms is to incorporate some statistics concerning the solution of a single equation and utilize it in the final process of approximation of the unknown fuzzy relation (set). Numerical considerations support the established algorithms.