Rédiger un PREreview

The pharmaceutical supply chain suffers from a persistent and costly inefficiency: critical batch data generated digitally by Contract Development and Manufacturing Organisations (CDMOs) is transmitted to drug product manufacturers as static PDF certificates or Excel spreadsheets, forcing skilled personnel to manually re-enter data into Quality Management Systems (QMS). This introduces transcription errors, delays batch release, undermines data integrity, and creates compliance risk under Good Practice (GxP) regulations. This paper presents a Design Science Research (DSR) artefactthe AAS Open Exchange Portalan interactive platform demonstrating how the Asset Administration Shell (AAS) standard, developed by the Industrial Digital Twin Association (IDTA), can eliminate this bottleneck through automated, machineto-machine batch data exchange between CDMOs and brand owners. The artefact was initially developed during the ISPE Emerging Leader Hackathon (April 2026, Copenhagen), sponsored by Roche, and subsequently extended into a universal MVP. The platform demonstrates bidirectional data exchange, Digital Calibration Certificate (DCC) integration, role-based access control using Attribute-Based Access Control (ABAC), AI-driven demand forecasting, a GxP-compliant audit trail, and a manual bridge mode for organisations not yet fully integrated. The AAS exchange is implemented as a browser-based simulation using client-side JavaScript; no live AAS server or production API was deployed, consistent with the scope of a DSR feasibility study. The paper evaluates the artefact against ALCOA+ data integrity principles, EU GMP Annex 11 requirements, and WHO traceability mandates, demonstrating that the AAS standard is not only technically feasible for pharmaceutical supply chain data exchange but natively satisfies regulatory requirements that current PDF-based workflows actively violate.