Digital Twins in STEM Education: A Systematic Review

This is the authors' abstract. We don't add key points for this paper.

Digital Twins (DT) have emerged as a transformative technology in STEM higher education, offering unique opportunities to overcome laboratory costs, safety risks, and access limitations. However, current literature lacks a systematic synthesis that integrates both technical implementation and measurable pedagogical effectiveness. This systematic literature review addresses four research questions concerning architectural models, pedagogical benefits, real‐time synchronization strategies, and usability challenges. Analysis of representative studies reveals a predominance of three‐component (36.4%) and five‐layer (27.3%) architectures, with 63.6% of implementations utilizing bidirectional data flow. Key pedagogical findings indicate significant improvements in academic performance ( n = 6) and student motivation ( n = 5), although only 27.3% of the studies provide rigorous statistical significance ( p < 0.05). Technical synchronization is primarily achieved through protocols like MQTT, AMQP, and serial communication integrated with engines such as Unity3D. Usability challenges highlight a persistent lack of social interaction and high development complexity. While DTs show promise for enhancing engineering education, critical gaps remain, including a lack of longitudinal studies, absence of interoperability standards, and missing cost‐benefit analyses. This review provides a strategic roadmap for researchers and educators aiming to integrate high‐fidelity digital replicas into engineering curricula.