The study analyzes the impact of artificial intelligence algorithms on composite manufacturing processes. It examines how deep learning methods and time series analysis improve the accuracy and efficiency of monitoring production parameters. Predictive maintenance algorithms are found to help minimize equipment downtime. Examines the impact of neural networks on product quality, achieving 98.3% accuracy for defect-free predictions. The effectiveness of time series in enabling timely parameter adjustments to prevent deviations is determined. It was found that limitations in defect data reduce model effectiveness in defect prediction. A proposal was formulated to expand the data set for more accurate prediction of rare events. The use of synthetic data to model rare failures and defects is proposed. A solution was developed that combines traditional and AI approaches to improve the reliability and resilience of production systems.