Additive manufacturing (AM) has emerged as a transformative technology across multiple industries, from aerospace to biomedical applications. The integration of artificial intelligence (AI) and machine learning (ML) into AM processes represents a paradigm shift toward intelligent, autonomous manufacturing systems. This comprehensive review synthesizes recent advances in AI/ML applications across the entire AM life cycle—from design optimization and process planning through in situ monitoring, closed-loop control, and post-process qualification. The analysis is organized by ISO/ASTM AM process families, including powder bed fusion (PBF), directed energy deposition (DED), material extrusion (MEX), vat photopolymerization (VP), binder jetting (BJ), material jetting (MJT), and sheet lamination (SL). For each process family, the review examines the specific AI/ML techniques employed, the data modalities utilized (thermal imaging, acoustic signals, in situ cameras, CT/NDE data), and the current state of deployment from research prototypes to industrial implementation. The analysis reveals that while significant progress has been made in single-stage ML applications such as defect detection and parameter optimization, truly integrated life cycle-wide AI-driven AM workflows remain largely aspirational. Key challenges are identified including data scarcity, model generalization across machines and materials, real-time control constraints, and certification requirements. Finally, future research directions are outlined toward autonomous AM systems enabled by physics-informed ML, digital twins, and hierarchical AI architectures.

Non-insulated heat exchangers in gas-to-gas service lose substantial energy to the surroundings. This study evaluates Al₂O₃ and ZnO ceramic coatings (200 $\mathsf{\mu }$m) as passive thermal retention layers on the inner surface of the outer tube in a copper double-pipe heat exchanger, using 3D CFD simulations verified for internal consistency against Log Mean Heat Transfer Rate analytical solutions. Six cases were modelled: three coating conditions across parallel-flow and counter-flow configurations under laminar conditions ($R{e}_i\approx 525$, $R{e}_o\approx 192$) with air as the working fluid. The coating elevates the outer tube inner wall temperature $T_3$, increasing the convective driving force to the cold fluid while suppressing ambient dissipation. In parallel flow, Al₂O₃ increases the net inter-fluid heat transfer rate by 35.7% and reduces ambient losses by 81.4%; ZnO achieves 30.9% and 70.4%, respectively. In counter-flow, Al₂O₃ yields a 26.6% enhancement and 73.2% loss reduction. The coated parallel-flow configuration outperforms the uncoated counter-flow baseline. Thermal circuit analysis shows that Al₂O₃ superiority arises from its higher conductivity (40 vs. 19 W m⁻¹ K⁻¹), which sustains a higher equilibrium $T_3$ and a heat partition ratio of 11.84 versus 7.17 for ZnO. These results show that a single ceramic coating layer can recover a large fraction of the thermal energy lost through non-insulated walls, offering a low-cost, retrofit-compatible pathway to improve the energy efficiency of gas-to-gas heat exchangers in HVAC, building energy recovery, and industrial process heat applications.