Advancing Dutch Circular Economy through Additive Manufacturing: Strategies for Repair and Remanufacturing using AM

Our society faces major challenges: resource scarcity, CO₂ emissions, and growing waste. Many products are discarded even though they could technically be repaired. Add-reAM explores how 3D printing can enable local, fast, and customized production of parts. For example, worn-out train components or turbine blades can be restored instead of replaced, which will lead to saving materials, energy, and costs. A critical enabler for successful remanufacturing is reliable condition assessment. Our task develops data-centric approaches for predicting Remaining Useful Life (RUL) and quality assessment of both operational and failed components. While existing methods focus on sophisticated algorithms, we address a fundamental gap: most industrial settings lack high-quality data due to noise, missing values, and limited datasets. Our approach prioritizes robust machine learning methods that work with imperfect real-world data, integrating RUL predictions with optimization algorithms to support cost-effective decisions on repair, reuse, or disposal.

Add‑reAM develops robust data‑centric methods to improve RUL predictions using imperfect industrial data, enabling sustainable, local, and cost‑effective repair, remanufacturing, and reuse.