[0001] Optimization problems associated with constraints are referred to as constrained optimization problems. Some examples include power flow problems, image reconstruction problems, pattern recognition problems, data processing problems, network flow problems, and optimal control problems. A constrained optimization problem generally includes an objective and at least one constraint. Conventionally, to solve a constrained optimization problem, the problem is first represented in a mathematical formulation, which is generally in a form of an analytical function of at least one parameter. A solution to the mathematical formulation is to find the parameter value that minimizes or maximizes the mathematical formulation while satisfying the constraint associated with the optimization problem. For example, the mathematical formulation of a constrained optimization problem could be . One solution is to find the parameter x of a certain value bound by the constraint associated with the problem that results in an extreme value of f(x). However, on many occasions, a constrained optimization problem cannot be simply characterized by an analytical function in terms of the parameters of interest. Moreover, since both objective and constraint from a real or simulated system are often corrupted by random noises, deriving an efficient and convergent computational procedure to perform simulation optimization for solving a constrained optimization problem is generally challenging.