The focus of the present paper is the application of a novel optimization algorithm called an imperialist competitive algorithm (ICA). In this algorithm, the cost function, which is a function of the input parameters, has to be optimized. In the present research, our cost function is the average free convection heat transfer from a horizontal isothermal cylinder located underneath an adiabatic ceiling. The input parameters are the ratio of the cylinder spacing from the adiabatic ceiling to its diameter (L/D), varying from 0.1 to 2.4, and the Rayleigh number, varying from 5500 to 40000. After data reduction, the regression equation of the average Nusselt number of the cylinder surface was obtained as a function of the Rayleigh number and the cylinder spacing ratio (L/D). The cost function was then optimized using the ICA. The results show that the proposed algorithm is powerful enough to be used for optimizing the cost function. According to the results obtained, the minimum Nusselt number corresponds to the optimum distance between the adiabatic wall and the cylinder.