A technique to eliminate the bounds in nonlinear programming problems

Abstract EN

In this paper, we consider the problem of minimization of an objective function having continuous first and second partial derivatives, subject to nonncgativity restrictions or upper and lower bounds on the variables of a nonlinear programming problem.

An appropriate transformation of its variables has been made to convert such a constrained optimization problem into an unconstrained optimization problem.

Some conditions are developed to guarantee that every local minimum of the unconstrained problem satisfies the Kuhn-Tuckcr's first order necessary conditions for a local minimum of the constrained problem.

There are some conditions for which the transformed objective function maintains the convexity of the original objective function in a neighbourhood of the solution.