A tractable approximation of expectation-based stochastic posynomial programs

In this paper, we consider a specific type of stochastic posynomial programming, which minimizes the expectation of a stochastically perturbed posynomial subject to inequalities of expectations of stochastically perturbed posynomials. Our goal is to build a computationally tractable approximation of this (typically intractable) optimization problem. We show that, under the assumption that the random perturbations are normal and narrow, these expectation-based stochastic posynomial programs can be approximated as conventional (deterministic) posynomial programs, that interior-point methods can solve efficiently