This study represents a novel method to accelerate history matching using artificial intelligence. Artificial intelligence is becoming popular in the oil and gas industry. The scope of this study is to provide guidelines to develop and train an artificial neural network (ANN) coupled with a genetic algorithm (GA) to optimize networks that can give an improved history match when given as input to a reservoir simulation model. For this work the concept of nominal decline rate (D) is used. For training the neural network, the difference in nominal decline rates between the varied numerical simulations and the base case field performance (Delta D) is used. A neural network model was developed to predict the differences between nominal decline rates (Delta D). Then a genetic algorithm used the trained neural network prediction model to determine the optimized parameter values. The feed-forward network with back propagation and the hyperbolic tangent sigmoid function (tansig) in the hidden layers of the network is used for the training/learning process. Results of the study showed that the NN-GA system considerably reduces the time and number of simulation runs required to achieve a good history match. Using the decline curve parameter for target data decreases the complexity and difficulty proxy and requires a smaller amount of training data to train the network.