Accelerating the design of flame-retardant polypropylene composites: An approach driven by explainable ensemble machine learning and Bayesian optimization

Key Points

• Enhancing the design of flame-retardant polypropylene composites may lead to improved safety standards.
• Machine learning techniques, particularly ensemble methods, played a critical role in optimizing composite properties.
• Bayesian optimization was employed to fine-tune the design process for these materials, maximizing their effectiveness.
• Findings indicate a promising approach that can be applied to other composite materials for better performance.