Purpose <p>The rising demand for sustainable solutions in 3D-printed construction necessitates advanced methods for evaluating and optimizing material performance across environmental, economic, and functional dimensions in the process. Most of the conventional material selection methods and LCAs are transparent, scalable, lack functionality for handling uncertainty, and are ineffective in fast-changing fields like 3D-printed construction. It contributes to this effort by designing an AI-based multi-criteria decision-making framework that is integrated with life cycle sustainability assessment tools for the optimization of 3D-printed building materials. Sustainable construction materials are increasingly critical due to the accelerated adoption of 3D printing, which introduces complex trade-offs among emissions, energy use, durability, and recyclability that conventional assessment tools are ill-equipped to handle in the process.</p> Methods <p>This study develops a new AI-based framework for sustainable construction materials in 3D printing by integrating life cycle sustainability assessment with advanced machine learning and optimization techniques. The framework incorporates Random Forest for material ranking, SHapley Additive exPlanations (SHAP) for interpretability, Gaussian Process Regression (GPR) for uncertainty-aware life cycle inventory prediction, and Proximal Policy Optimization (PPO) for real-time multi-objective optimizations.</p> Results <p>The results showed significant improvements in the sustainability performance across several metrics. It ranked material A with the highest sustainability grade of 85. SHAP values further indicating that carbon footprint, with 0.35, and energy consumption, with 0.25, were the two most influential factors in the choice of material. GPR predicted life cycle inventory parameters with high precision and gave reliable estimates. PPO was used to optimize material formulations and process parameters. The result shows the framework in its robustness and effectiveness in 3D-printed materials’ optimization towards sustainability. The quantitative gains of the developed system amount to 15% less carbon emissions, 8% of construction costs saved, and improved durability of materials by 20%. It provides a model for decision-making that is explainable, adaptive, and scalable, and thus a transparent alternative to the standard sustainability assessment procedures. Integration with explainable AI and reinforcement learning empowers data-driven, interpretable decision-making in sustainable construction with 3D printing technology.</p> Conclusion <p>The paper proposed an end-to-end AI-driven framework of life cycle sustainability assessment in 3D-printed building materials that integrated RF with SHAP for explainability, GPR for LCI, and PPO for sustainability-driven optimization. This proposed model could mitigate some of the important shortcomings in the traditional methods of material selection and life cycle assessment by using new approaches that brought more transparency, modeled uncertainty, and finally optimized multiple sustainability objectives together. These techniques are being put together to provide informed decision-making in a rather complex domain for sustainable constructions.</p>