In this paper a polynomial regression model where the constituents of are of arbitrary order is proposed. A genetic algorithm is used to find the optimal terms to be included in the so-called optimal polynomial regression model . The objective for the genetic algorithm is to minimize the sum of squared errors of the predicted responses. In practice the genetic algorithm generates an optimal set of exponents of the design variables in a polynomial regression model. Several example problems are presented to show the performance and accuracy of the optimal polynomial regression model. Results show a greatly improved performance for optimal polynomial regression models compared to traditional regression models.