Among topics in manufacturing and design of industrial parts, quality and roughness are very important for the producers. Large quantities of industrial parts are cylinders and cylindrical parts. These parts are polished by techniques such as internal grinding and honing depending on the type, diameter, length, number and quality levels expected. In this study, the quality of the surface of the honing as the output factor was investigated and some of the process parameters comprised of lubricant type and temperature, machining speed, initial surface roughness, degree of abrasive tools and sweep cycles were selected for the experimental purposes. According to the results obtained, the gasoline liquid at 20 ° C was the best candidate for reducing the surface roughness and therefore lubricant type and temperature parameters were considered constant during throughout the experiments. The DOE achieved based on full factorial. Diagram results show that the number of cycles, initial surface roughness and degree of abrasive tools are the most important factors among the factors studied. However, due to the fact that the deceleration time increased the polish time, one can use the high speeds for this purpose. The GeneXproTools software was used for the modeling process. The proposed modeling correlates the honing parameters with the surface roughness using a mathematical formula. The mathematical formula ensures that if the honing parameters are known, the surface roughness is predictable. An optimization process is then conducted to optimize the honing parameters in order to maximize the surface roughness predicted by the mathematical formula. The results show that the combination of modeling and optimization processes leads to the prediction and maximization of surface roughness, which in turn brings about significant benefits including saving the energy and time consumed as well as reducing the costs of producing samples with the best surface. At the end, a few case studies were performed for evaluating the results obtained by the modeling and optimizing algorithms. The result showed a good agreement between the experimental data and the simulated ones. Key words : Honing, Surface Roughness, Gray Cast Iron, Silicon Carbide, Modeling, Optimization.