Abstract The thermal methods commonly applied in developing fairly homogeneous heavy-oil reservoirs may not be viable options in channeled heavy-oil reservoirs mainly because the injected fluid (steam) would bypass the heavy oil by flowing through the high permeability channels. Experimentally, it was found that the key to success in developing such reservoirs is to have optimum heat conduction which overcomes the reservoir heterogeneity. Hence, steam flooding, if properly optimized, could overcome this problem. Injection rate and steam quality are the main operational parameters that controls the economic heat propagation in the reservoirs. Additionally, placing the wells in more conducive areas to ensure that the regions with low-to-medium permeability get enough heat, is crucial. In this work, we propose the optimization of steam quality using CMA-ES. A three-dimensional channeled reservoir model undergoing steam flooding was built and CMA-ES was used to estimate the optimal values of three variables of steam flooding. The Net Present Value (NPV) of the project was used as the objective function and three cases of optimization were studied. In Case 1, only the injection and production rates were optimized. In Case 2, well placement was optimized along with the rates. In Case 3, steam quality was included in the list of optimized variables. Results indicated that Case 2 gave higher NPV than Cases 1 and 3 signifying the importance of optimum selection of well placement. Also, it is understood that steam quality is an important parameter which controls the overall steam flooding process and a large number of function evaluations is required to obtain optimum values of steam quality.