This study proposes an artificial intelligence-driven model that can enhance productivity and reduce manufacturing costs in the brewery industry of Nigeria. The research initiated with a critical literature review on the factors of productivity in the knowledge-intensive industries, choosing thereupon the brewery sector based on expert advice. In total, three predictive models were developed, namely Artificial Neural Network, Machine Learning, and a hybrid Artificial Neural Network-Machine Learning model, for predicting productivity. The Mean Squared Error was 0.001399 for the Artificial Neural Network model, Root Mean Squared Error was 0.037407, and Mean Absolute Error was 0.037283, while the Machine Learning had Mean Squared Error of 0.040378, Root Mean Squared Error of 0.200943, and Mean Absolute Error of 0.183000, the hybrid having Mean Squared Error of 0.013982, Root Mean Squared Error of 0.118247, and Mean Absolute Error of 0.110141. It also proved the fact that the Machine Learning model is able to predict productivity based on maintenance, Mean Time Before Failure, and Mean Time to Repair indicators since the obtained values for this type of model had lower errors than all the others: Mean Absolute Error = 0.08508, Mean Squared Error = 0.19275, Root Mean Squared Error = 0.43903.