Palm oil biomass-based gasification has become a potential technology to overcome anthropogenic environmental challenges. While physical experimentation is time-consuming and expensive, it can be avoided when determining the best settings for a particular gasifier and the behaviour of palm oil biomass. An Artificial Neural Network (ANN) model was developed to estimate syngas composition (CO and H 2 ) over a wide range of palm oil biomass characteristics and gasifier operating conditions. A vast amount of secondary data comprising both categorical and numerical was gathered for the development of the proposed ANN model. To improve the model’s performance, uncorrelated input data were removed using International Business Machines Statistical Package for the Social Sciences (IBM SPSS) Statistics software by utilizing Spearman's Correlation Coefficient (SCC) matrix. Feed-Forward Back Propagation (FFBP) and Levenberg-Marquardt (LM) learning algorithms with one and two hidden layers, as well as a range number of neurons and transfer functions were used to train the network using the ANN toolbox, available in Simulink, MATLAB software. The best-performing network structure was identified based on the lowest Mean-Squared Error (MSE) and highest Regression value, subjected to numbers of network topologies. The developed ANN model is able to accurately predict the output of syngas composition (MSE = 0.1 and R 2 > 0.8). The results indicated that the ANN model shows excellent model prediction which can aid in the effective operation of biomass gasification under various operating conditions.