1st BILSEL INTERNATIONAL ASPENDOS SCIENTIFIC RESEARCH CONGRESS, Antalya, Türkiye, 24 Şubat 2024
Arithmetic
with fuzzy numbers is widely applied in
many different directions such as in engineering and decision making. This high
potential of applications of fuzzy arithmetic in different disciplines is
attracting a huge amount of interest of researchers. Most of these applications
use the minimum t-norm as the join in the well-known Zadeh's extension formula.
For this type of arithmetic the alpha-level method is easy to employ because of
the monotonicity of arithmetic operators. Nevertheless, arithmetic with other
types of t-norms still plays an important role in applications. However, fuzzy
arithmetic with different type of t-norms, which actually is a generalization
of Zadeh's extension by replacing the minimum with a general t-norm, is facing
much complexity in practical applications. These difficulties with
product-arithmetic have to be overcome since there is a necessity for it in
certain fields of applications. For example it is proven that the product
t-norm is the only t-norm for which fuzzy constrained optimization problems are
scale-invariant. Another example could be the successful employment of product
t-norms in fuzzy controllers. On the other hand we observe that in the
literature the majority of papers deal with normal fuzzy numbers. Arithmetic
taking into account subnormal fuzzy numbers is however needed for more
appropriate modeling of real life problems. One comes across subnormal fuzzy
numbers in applications like fuzzy risk analysis, fuzzy decision making and
fuzzy linear programming.
The objective of this work is to provide a
particular response to the requirements mentioned above. Namely, the paper
deals with the product t-norm arithmetic of subnormal triangular fuzzy numbers
which will be called generalized triangular fuzzy numbers hereafter. The work
expands the authors former study about product arithmetic with normal fuzzy
numbers, so in the same way the extension principle is followed.