Controlled dual perturbations for posynomial programs (Q1112725)

With geometric programs getting larger and larger, the dual based solution methods which optimize a less complicated problem than the original problem become more important. However there are two major difficulties related to the development of a dual method. One is the non- differentiability of the dual objective function over its feasible region and the other one is the conversion of a dual solution to a primal solution. In this paper the authors introduce a well-controlled method for posynomial programs. Given any arbitrarily small number \(\epsilon >0\), they select a proper perturbation vector \(\ell\) to construct a perturbed dual program with a differentiable objective function over linear constraints. Solving this perturbed dual program, they find an \(\epsilon\)-optimal dual solution. Then they use this dual solution to construct a linear program. An \(\epsilon\)-optimal solution to the original geometric program is provided by the dual solution of this linear program.