In this paper, we present the theory and design of interval type-2 fuzzy lo
gic systems (FLSs), We propose an efficient and simplified method to comput
e the input and antecedent operations for interval type-2 FLSs; one that is
based on a general inference formula for them. We introduce the concept of
upper and lower membership functions (MFs) and illustrate our efficient in
ference method for the case of Gaussian primary MFs, We also propose a meth
od for designing an interval type-2 FLS in which we tune its parameters, Fi
nally, we design type-2 FLSs to perform time-series forecasting when a nons
tationary time-series is corrupted by additive mise where SNR is uncertain
and demonstrate improved performance over type-1 FLSs.