The Karush-Kuhn-Tucker optimality conditions for the optimization problem with fuzzy-valued objective function (Q2472184): Difference between revisions

The author considers the following minimization problem with fuzzy-valued objective function (P) \(\min \widetilde{f}(x)=\widetilde{f}( x_{1},\dots ,x_{n}) \) subject to \(g_{j}( x) \leq 0\), \(j=1,\dots ,m\) where the real-valued constraint functions \(g_{j}:\mathbb{R} ^{n}\rightarrow \mathbb{R}\) are convex on \(\mathbb{R}^{n}\) for \(j=1,\dots ,m.\) First, some basic properties and arithmetics of fuzzy numbers are introduced (Sect. 2). The author uses the well-known Hausdorff metric to define the distance between any two fuzzy numbers and uses this metric to considers the continuity and limit of a fuzzy-valued function (Sect. 3). He uses the concept of Hukuhara difference between two fuzzy numbers, to propose the differentiation of fuzyy-valued function. (Sect. 4). Then, the author provides a solution concept for Problem (P) (Sect. 5) and derives Karush-Kuhn-Tucker optimality conditions for problem (P) by introducing the Lagrange function multipliers (Sect. 6). Finally some examples are provided to illustrate the applications.