Optimization over the efficient set (Q5896268)

Let \(X\subset {\mathbb{R}}^ n\) be a nonempty set and C a \(k\times n\) matrix. The author considers the vector maximization problem (V) VMAX: Cx subject to \(x\in X\) as a problem of finding all efficient solutions. Here a point \(x^ 0\) is said to be an efficient solution of (V) when \(x^ 0\in X\) and there is no \(x\in X\) such that \(Cx\geq Cx^ 0\) and \(Cx\neq Cx^ 0\). Let \(X_ E\) denote the set of all efficient solutions for the problem (V). Let \(d\in {\mathbb{R}}^ n\). Then the problem (P) of central concern in this paper is given by \(\phi(X_ E)=\sup<d,x>\). subject to \(x\in X_ E\). Necessary and sufficient conditions are given for the problem (P) to be unbounded and for efficient and for arbitrary solutions of problem (V) to be optimal solutions for problem (P). Some of these conditions are geometric in nature, others algebraic. Finally conditions are given under which the set of optimal solutions for the problem (P) possesses certain special properties.