Semiparametric sufficient efficiency conditions and duality models for multiobjective fractional subset programming problems with generalized \((\mathcal F,\rho, \theta)\)-convex functions (Q699943)

We shall present a fairly large number of global semiparametric sufficient efficiency conditions and duality results under various generalized \(({\mathcal F},p,\theta)\)-convexity hypotheses for the following multiobjective fractional subset programming problem: \[ \text{Minimize} \left(\frac {F_1(S)} {G_1(S)},\frac {F_2(S)}{G_2(S)},\dots, \frac {F_p(S)}{G_p(S)} \right)\text{ subject to }H_j(S)\leq 0,\;j\in\underline q,\;S\in\mathbb{A}^n,\tag{P} \] where \(\mathbb{A}^n\) is the \(n\)-fold product of the \(\sigma\)-algebra \(\mathbb{A}\) of subsets of a given set \(X,F_i,G_i\), \(i\in \underline p\equiv\{1,2,\dots,p\}\), and \(H_j\), \(j\in\underline q\), are real-valued functions defined on \(\mathbb{A}^n\), and for each \(i\in\underline p\), \(G_i(S)>0\) for all \(S\in\mathbb{A}^n\) such that \(H_j(S)\leq 0\), \(j\in \underline q\). In Section 2 we recall the definitions of differentiability, convexity, and certain types of generalized convexity for \(n\)-set functions, which will be used frequently throughout the sequel. We begin our discussion of sufficient efficiency criteria for (P) in Section 3 where we state and prove a number of sufficiency results. More general sets of sufficiency conditions are formulated and discussed in Section 4 with the help of two partitioning schemes. The first of these schemes was originally used by B. Mond and T. Weier for constructing generalized dual problems for nonlinear programs with point-functions, whereas the second, which is a variant of the first, was recently utilized in by X. Yang for stating a general dual problem for a multiobjective fractional program with point-functions. In Section 5 we turn to an investigation of the notion of semiparametric duality for (P). Here we consider a simple dual problem and prove weak, strong, and strict converse duality theorems. In Section 6 we formulate another dual problem with a relatively more flexible structure that allows for a greater variety of generalized \(({\mathcal F},\rho,\theta)\)-convexity assumptions under which duality can be established. Finally, in Sections 7 and 8 we state and discuss two general duality models which are, in fact, two families of dual problems whose members can readily be identified by appropriate choices of certain sets and functions.