Strong and weak conditions of regularity and optimality

Abstract

Nondegenerate optimality conditions for Pareto and weak Pareto optimal solutions to multiobjective optimization problems with inequality and multi-equality constraints determined by Fréchet differentiable functions are established. First, weak and strong regularity conditions are derived, in order to determine weak Karush–Kuhn–Tucker (positivity of at least one Lagrange multiplier associated with objective functions) and strong Karush–Kuhn–Tucker (positivity of all the Lagrange multipliers associated with objective functions) conditions. Subsequently, the class of problems for which every weak (resp. strong) Karush–Kuhn–Tucker point is weak (resp. strong) Pareto solution is characterized. In addition examples that illustrate our results are presented.