The present study is a hybrid approach of machine learning (ML) based prediction and explainable machine learning (XML). Three different ML approaches like Linear Regression (LR), Decision Tree (DT), and Random Forest (RF), were used. An array of statistical and graphical methods was used for model evaluation and comparison. RF-based prediction model offered superior results, with an R2 score of 0.9650 for training data and 0.8525 for testing data. For the RF model, the mean squared error (MSE) values were 0.0181 and 0.0677, which means that the predictions were accurate and the models were able to generalize well. The combined error distribution plots illustrated a Gaussian shape and most of the data points centred around zero, suggesting the model was resilient and sans any systemic deviation. Parity plots showed that most of the predicted points are within 10% of the experimental values, which shows that the model is very accurate. The residuals were spread out at random, with no trend or bias, which proved that there was no underfitting or overfitting. SHapley Additive exPlanations (SHAP) analysis was used to make the model more transparent by finding the most important process factors that affect syngas yield and allowing for feature-level interpretability. The study found that the carbon content and steam-to-biomass ratio were the most important factors for higher syngas yield. The results show that explainable ML-based frameworks have a lot of potential for helping the switch to renewable energy by using strong and understandable modelling methods.