M. J. Gisbert Francés, M. J. Cánovas Cánovas, J. Parra López, F. J. Toledo Melero
This talk is focused on the sensitivity analysis in the context of linear programming under perturbations of the right-hand-side (RHS) coefficients. In contrast with the classical approach (in which only one constraint is perturbed), in our framework simultaneous perturbations of all constraints are allowed. Specifically, we quantify the stability of the optimal value function by means of the study of the calmness property and the computation/estimation of calmness constants. Roughly speaking, a calmness constant in this setting provides an upper bound for the ratio between the variation of the optimal value and the perturbation of the RHS coefficients. We present our recent results about this property and try to connect them with previous results on the calmness property of the feasible and the optimal set mappings. Some illustrative examples are provided.
Palabras clave: Linear programming, optimal value, sensitivity analysis, calmness
Programado
L08.4 Optimización Lineal, Estocástica y Robusta
5 de septiembre de 2016 15:40
Aula 21.07
