This work presents the development of an Artificial Neural Network (ANN) to predict biogas production from two-twin fabricated bioreactors against temperature, pH and chemical oxygen demand (COD) variables. A database designed with the experimental data was divided into two groups of training (70%) testing and validation (30%) and models obtained using MATLAB 9.3 R2017b toolbox. The ANN optimization of experimental factors achieved an 80% biogas yield. The ANN predicted the biogas production based on operational parameters and validated with a correlation coefficient of 0.98098. A back propagation (BP)-(ANN) algorithm showed a good regression between actual values and prediction values for the biogas yield. The model was tested by mean square error and regression value which were 0.99998 and 0.99722. This indicated that the 99.722% confidence agreed with the selected distributions of sigmoid. Log-sigmoid showed that the best prediction performance was observed in the model with network structure 4-5-1 corresponding to 4 input neurons, 5 hidden neurons and 1 output neuron. The error margin revealed minimally insignificant error in the overall data output between the test results and modelled results. Hence the ANN model predicted biogas production for the substrates at the variables and can be improved with large-scale operations.