https://doi.org/10.1016/j.apsb.2025.06.001
This new article publication from Acta Pharmaceutica Sinica B describes an AI integrated quality prediction and diagnostic framework that can convert small-scale pharmaceutical manufacturing from experience-driven practices to data-driven accuracy.
The pharmaceutical industry faces challenges, particularly in the quality digitization of complex multi-stage processes, in small-scale systems. In this article, how the Intelligent Quality Prediction and Diagnostics (IQPD) framework was developed and applied to Tong Ren Tang's Niuhuang Qingxin Pills, and utilizes four years of data collected from four production units to cover the entire process from raw materials to complete products. This framework proposes a new path-enhanced double ensemble quality prediction model (PEDGAT) that combines graph attention networks with path information to encode long distances and sequential dependencies between units. Furthermore, double ensemble strategies increase the stability of small sample models. Compared to the global traditional model, PedGat achieves cutting-edge results with an average improvement of 13.18% and 87.67% in predictive accuracy and stability for the three indicators. Furthermore, more detailed diagnostic models that utilize grey correlation analysis and expertise reduce reliance on large samples, provide a panoramic view of unit-wide attribute relationships, and increase process transparency. Finally, the IQPD framework is integrated into human and physical physical systems, enabling decision-making and real-time quality adjustments for Tong Ren Tang's Niuhuang Qingxin Pills. This facilitates the transition from empirically driven to data-driven manufacturing.
Keywords: smart manufacturing, artificial intelligence, intelligent quality prediction and diagnosis, small sample multi-unit manufacturing, data-driven manufacturing, real world tongrentan niuhuang Qingxin Pills
Graphical Abstract: Available at https://ars.els-cdn.com/content/image/1-s2.0-s2211383525003806-ga1_lrg.jpg
An AI integrated IQPD framework for intelligent quality prediction and diagnosis in the manufacture of small samples multi-unit pharmaceuticals.
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
