Artificial neural networks (ANN) have become very significant tools in many areas including agricultural system. ANN was used in this research for modeling and simulation of plantain growth prediction. Plantain production is concentrated in the rain-forest belt of West and Central Africa where it constitutes an important staple food of the local population. Using ANN to stimulate the growth of plantain will make plantain farmers to plan their planting. They would already know the outcome of the production and can forecast the financial expenses before and after planting. If they know the height of the plant sucker, (NEL); Leaf width, (LW(cm)); and the Leaf length, (LL(cm)) to predict into future the values of Ht , G50, NEL, LW, LL, BW in kg (Bunch weight), NHB (Number of hands in the bunch) and NFB (Number of fingers in the bunch) of plantain plant based on the planting conditions for dataset used for training in this research. Elman time series neural network, a back-propagation algorithm, with 1 input neuron and 1 output neuron with varying number of neuron in the hidden layer was employed to train the parameters. The network result of Ht, G50, NEL, LW and LL, are used to train the network of BW, NHB, and NFB. Their corresponding coefficient of determination (R2) and the Root Mean Squared Error (RMSE) were recorded to determine the acceptance of each network architecture. After all these parameters have been trained, the network predicted into future the value of plantain height, the girth at 50 cm above the soil level, number of emitted leaves, leaf width, leaf length, bunch weight of plantain, number of hands in the plantain bunch and number of fingers in the plantain bunch given the known value ( from the plantain sucker, an experimental value) of Ht, G50, NEL, LW, and LL to the network. It was evidenced that the neural network was able to predict future values of plantain plant and yield at harvest.