Skip to content
Review Open access

Artificial Intelligence for Quality Assurance in Additive Manufacturing: A Systematic Review of Real-time Anomaly Detection Solutions through Non-Visual Signal Analysis

Oct 2026 · The International Journal of Advanced Manufacturing Technology · 0 citations · 95 references

Abstract

Additive Manufacturing (AM) represents an industrial paradigm shift, surpassing conventional methods by building complex three-dimensional products layer by layer, optimizing weight, customization, and sustainability. However, its widespread industrial adoption is hindered by critical quality challenges such as porosity, excessive surface roughness, and insufficient interlayer adhesion. Within this Industry 4.0 context, the convergence of AM with Big Data and Artificial Intelligence (AI) emerges as a promising solution, enabling real-time defect detection, parameter optimization, and mechanical property prediction through sensor-generated data analysis. This study conducts a Systematic Literature Review (SLR) to identify, evaluate, and synthesize AI-based data analysis solutions applied to automated AM processes, focusing on quality assurance and real-time anomaly detection through non-visual signal analysis. Employing PRISMA guidelines, the methodology utilized Scopus and Web of Science databases, culminating in the selection of 91 articles after rigorous screening. Content analysis revealed a predominance of research in metallic materials (60 occurrences), with porosity being the most studied defect (44 occurrences). In polymers (31 occurrences), dimensional deviations (12 occurrences) and nozzle failures (7 occurrences) were most explored. Diverse non-visual sensor technologies were identified, including photodiodes and ultrasound for metals, and accelerometers and temperature sensors for polymers. AI methods, notably Convolutional Neural Networks (CNNs), demonstrated high effectiveness, with reported efficiencies reaching up to 99.30%, in interpreting these complex signals. This work consolidates current knowledge on multi-sensor and AI-driven strategies, essential for enhancing reliability, reducing waste, and accelerating industrial adoption of AM, while outlining future directions for real-time quality assurance advancement.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.