This paper is to deal with Mond-Weir duality and Wolfe duality for constrained set-valued optimization problems in terms of conedirected Clarke derivatives. Firstly, necessary and sufficient optimality conditions for constrained set-valued optimizations in terms of cone-directed Clarke derivatives for the cone-semilocally convex like maps are investigated. Then, the Mond-Weir duality and Wolfe duality for a constrained set-valued optimization and their weak duality, strong duality and converse duality are considered. TÓM TẮT Bài báo này khảo sát bài toán đối ngẫu dạng Mond-Weir và Wolfe cho bài toán tối ưu đa trị có ràng buộc sử dụng đạo hàm đa trịClarke theo hướng nón. Trước hết, điều kiện tối ưu cần và đủ cho bài toán tối ưu đa trị có ràng buộc sử dụng đạo hàm đa trị Clarke theo hướng nón cho lớp hàm tựa lồi nửa địa phương được khảo sát. Sau đó, bài toán đối ngẫu dạng Mond-Weir và Wolfe cho bài toán tối ưu đa trị có ràng buộc và các tính chất về đối ngẫu mạnh, đối ngẫu yếu và đối ngẫu ngược được trình bày.

No abstract

In the paper, we study the generalized differentiability in set-valued optimization, namely stydying the second-order composed radial derivative of a given set-valued mapping. Inspired by the adjacent cone and the higher-order radial con in Anh NLH et al. (2011), we introduce the second-order composed radial derivative.  Then, its basic properties are investigated and relationships between the second-order compsoed radial derivative of a given set-valued mapping and that of its profile are obtained. Finally, applications of this derivative to sensitivity analysis are studied. In detail, we work on a parametrized set-valued optimization problem concerning Pareto solutions.  Based on the above-mentioned results, we find out sensitivity analysis for Pareto solution mapping of the problem. More precisely, we establish the second-order composed radial derivative for the perturbation mapping (here, the perturbation means the Pareto solution mapping concerning some parameter). Some examples are given to illustrate our results. The obtained results are new and improve the existing ones in the literature.