A new geometric approach for sensitivity analysis in linear programming

Abstract

In this paper, we present a new geometric approach for sensitivity analysis in linear programming that is computationally practical for a decision-maker to study the behavior of the optimal solution of the linear programming problem under changes in program data. First, we fix the feasible domain (fix the linear constraints). Then, we geometrically formulate a linear programming problem. Next, we give a new equivalent geometric formulation of the sensitivity analysis problem using notions of affine geometry. We write the coefficient vector of the objective function in polar coordinates and we determine all the angles for which the solution remains unchanged. Finally, the approach is presented in detail and illustrated with a numerical example.