Data Integrity Failures in Pharmaceutical Digital Twins and Continuous Manufacturing: An Alcoa + + Framework Integrating Human Factors and Simulation Vulnerabilities

Digital twin technologies have the potential to enhance real-time operations and predictive quality control in pharmaceutical manufacturing, but data integrity safeguards have yet to be matured with a focus on human and simulation vulnerabilities. This review is based on real-world data integrity failures in digital twin assisted continuous manufacturing, with ALCOA + + principles combined with human factors and simulation risks. The incidents associated with digital twins, PAT, and continuous manufacturing were identified using a PRISMA-ScR approach to reviewing the PubMed, Scopus, Web of Science and regulatory databases (FDA 483s, EMA inspections, 2020–2026). From a failure classification perspective, an ALCOA + + and an adapted version of the HFACS were used. Of the 248 incidents, 65% were non-contemporaneous data, 22% were non-traceable simulation inputs and 13% were non-enduring virtual records from cloud synchronization failures. The simulated failure rates were reduced by 68% with a Digital Twin Compliance Framework which integrates human-centered design, GAMP 5.2 risk assessment, hybrid audit trails and AI-based anomaly detection. This review presents the first unified taxonomy for digital twins failures in continuous manufacturing (ALCOA++) and identifies the need for robust virtual data governance and further validation in a decentralized manufacturing network of the future.