422303 Implementation of Advanced Multilayer Plant-Wide Control Architecture into a Direct Compaction Continuous Pharmaceutical Manufacturing Process

Monday, November 9, 2015: 4:18 PM
Ballroom B (Salt Palace Convention Center)
Ravendra Singh, Fernando J. Muzzio, Marianthi Ierapetritou and Rohit Ramachandran, Chemical and Biochemical Engineering, Rutgers University, Piscataway, NJ

Continuous pharmaceutical manufacturing provides an appropriate platform where feed-forward and feedback control strategies, including necessary alarms, can be implemented to improve the product quality, minimizing product rejection and operating expenses. The precise control of the quality of the pharmaceutical product requires corrective actions on the process/raw material variability before they can influence the product quality. It is therefore highly desirable to implement a combined feed-forward/feed-back control system into a pharmaceutical manufacturing plant [1] where multivariate measurements of raw and intermediate materials, finished products, and process parameters are used to control quality. The feed-forward controller takes into account the effect of process disturbances proactively while the feedback control system ensures the end product quality consistently. The integrated feed-forward/feedback control system ensures minimum variability in the final product quality irrespective of process and raw material variations.

In this work, efficient multilayer plant-wide control architecture has been designed and implemented into a direct compaction continuous tablet manufacturing pilot-plant. The proposed control architecture consists of four layers. In the first layer, necessary sensors including PAT tools have been implemented for real time process monitoring. In second layer, feedback control loops have been implemented to ensure the critical quality attributes. In third layer, feed-forward control loops have been integrated with feed-back control loops to proactively mitigate the effects of process and raw materials variability and thereby to achieve more precise control of end product quality with minimum rejections. In final layer of the control architecture, the necessary alarms and signals have been implemented to control the plant operation more efficiently and reject the out of specs powder blends and tablets automatically.

The critical process parameters and quality attributes are monitored at each unit operation via a centralized control platform using advanced monitoring tools. For example, feeder flow rate and feeder screw rotational speed have been monitored at feeding operation. Powder blend composition, uniformity (RSD) and bulk density are monitored at the blending operation. Tablet weight, thickness, hardness, compression forces, ejection force, and operating parameters (e.g. fill depth, feed frame speed) are monitored at the tableting operation. NIR is applied first time to measure the powder bulk density in real time. As part of second layer of control architecture, supervisory feedback loops to control powder blend composition [2], main compression force, tablet weight and hardness are implemented. The tablet weight and hardness control loops have been decoupled so that both critical quality attributes can be controlled simultaneously. Main compression force and tablet weight are controlled through a cascade control arrangement. The necessary, feed-forward control loops are implemented so that the upstream measurements can be used to take downstream mitigation actions in real time. For example, measured powder bulk density was feed-forwarded in real time to the tablet press to proactively mitigate the effect of powder bulk density on tablet weight and hardness via changing the fill depth. Finally, the necessary alarms are implemented for unexpected events management including sending the warning/alarms signals to the operator, stopping the production if needed and rejecting powder blends and tablets out of specification. The four layers of control architecture have been implemented using industry standard control platform (e.g. DeltaV), PAT data management tool (synTQ), OPC communication protocol, fieldbus devises, chemometric tools and real time monitoring tools. The proposed control strategy is found to be better than implementations using only feedback control or without control strategy and therefore demonstrates potential to further improve pharmaceutical tablet manufacturing operations.

The objective of this presentation is two-fold. First to highlight the combined feed-forward/feedback based four layers of control architecture, and second to demonstrate its performance to control the direct compaction continuous tablet manufacturing process.


  1. Singh, R., Muzzio, F., Ierapetritou, M., Ramachandran, R. (2015). A combined feed-forward/feed-back control system for a QbD based continuous tablet manufacturing process. PROCESSES Journal, 3, 339-356.
  2. Singh, R., Sahay, A., Karry, K. M., Muzzio, F., Ierapetritou, M., Ramachandran, R. (2014). Implementation of a hybrid MPC-PID control strategy using PAT tools into a direct compaction continuous pharmaceutical tablet manufacturing pilot-plant, International Journal of Pharmaceutics, 473, 38–54.

Extended Abstract: File Not Uploaded