@article{xu2026aigoverned,
  title = {AI-Governed Knowledge Infrastructure for Predictive Compliance and Supply Resilience in Class II Medical Devices: A Case Study},
  author = {Kenneth XU and Nick VYAS and Markus MAU},
  year = 2026,
  url = {https://ibimapublishing.com/articles/JIBBP/2026/218352/},
  journal = {Journal of Innovation and Business Best Practice},
  volume = 2026,
  pages = 12,
  doi = 10.5171/2026.218352,
  abstract = {Small and mid-sized Class II medical device manufacturers face a scale-up challenge when regulatory, quality, and supply chain information grows faster than the organization’s ability to manage it. This paper examines whether governed artificial intelligence can convert fragmented records into decision-ready knowledge without weakening compliance, traceability, or human accountability. Using an applied case-study design, the research evaluates a six-month pilot at New Horizon Biotech, an anonymized Class II diabetes device manufacturer. A five-FTE AI Business Unit connected regulatory evidence, supplier records, quality documentation, and operational signals through a dual-loop model. The Regulatory Readiness Loop supported evidence retrieval, predicate comparison, submission drafting, and EU MDR GSPR mapping. The Supply Adaptation Loop supported supplier early warning, alternate-source readiness, change impact review, and landed-cost analysis. Pilot results showed a 32.3% reduction in drafting cycle time, 67.8% reduction in defect density, 88.0% reduction in evidence retrieval time, 38.8% faster lead-time recovery, zero critical stockouts, a Month 6 groundedness score of 0.94, and an 82.4% six-month ROI. The findings suggest that governed AI can support regulatory throughput and supply resilience when it is implemented as controlled knowledge infrastructure rather than autonomous decision-making.},
  keywords = {Governed AI; knowledge management; supply chain resilience; Class II medical devices},
  note = Article ID: 218352
}