L-Lysine is one of the important amino acid required for humans and animals. It has a high commercial market. Large scale production of this amino acid is essential to meet the commercial demands. Typically, L-lysine is produced by batch fermentation. In the present study, the important process, as well as nutrient parameters such as glucose concentration (g/L), rpm, incubation temperature (°C), pH and incubation time for L-lysine production by Corynebacterium glutamicum ATCC13032, were optimized by a combined approach of response surface methodology (RSM) with artificial neural network (ANN) method. Initially, 32 runs face central composite design was employed. In the first step, the data was analyzed by the RSM and the optimum conditions for L-lysine production were determined. In the second step, the same data was used to train the neural network. A feed-forward neural network with error backpropagation was used. The best network was obtained by optimizing the no of neurons in the hidden layer. From the best network, the optimized weights and predicted responses were used to optimize the conditions of the selected parameters by genetic algorithm (GA). Overall with the combination of RSM-ANN-GA onefold of L-lysine production from Corynebacterium glutamicum ATCC 13032 was improved.