Introduction This study focuses on the development and implementation of an Exceptional Events Management (EEM) framework that detects, diagnoses, and mitigates exceptional events inherent to particulate processes that are prevalent in the pharmaceutical industry. Methods The EEM framework combines several methods, including signed directed graph (SDG), qualitative trend analysis (QTA), and fast Fourier transform analysis. Results We discuss the efficacy of the EEM framework in detecting and diagnosing abnormal events, and demonstrate its application on roller compaction. We demonstrate that various commonly occurring exceptional events such as no powder entering the roll region (which includes jamming in the nip region), caking of powder on rolls and varying moisture in powder can be detected and diagnosed using the EEM framework. In addition, we show that the SDG-QTA approach is useful in determining the type and location of sensors that would enable better monitoring thereby improving the detection and diagnostic capabilities of our EEM framework. Conclusion The developed EEM framework shows potential for facilitating the pharmaceutical transition from batch to continuous manufacturing.

• Purdue University

• Oral solid dose
• Process Control

1. Detection and diagnosis
2. Fast Fourier transform
3. Mitigation
4. oral solid dose
5. process control
6. Qualitative trend analysis
7. Research Article
8. Roller compactor
9. Signed directed graph