A process systems approach to streamline the development of pharmaceutical manufacturing processes

In recent decades, the pharmaceutical industry has played a crucial role in the enhancement of the human well-being, and has significantly contributed to positive economic growth. Historically, the pharmaceutical sector has relied on resource intensive and empirical trial-and-error methods for process development and decision-making. However, conservative approaches are not capable of optimizing returns on Research and Development investments; rather, drug shortages and difficulties in launching new products are often experienced. Adopting model-based strategies in process development and scale-up is essential for streamlining pharmaceutical development and enhancing manufacturing efficiency. Despite the growing focus on mathematical modeling for pharmaceutical development, several issues must be addressed to establish a comprehensive modeling framework for the practical application of regulatory guidelines. This Dissertation addresses some key challenges in the systematic adoption of model-based approaches within pharmaceutical industry, with focus on regulatory compliance and process optimization. Specifically, the objectives of this work are threefold: (i) to evaluate model prediction fidelity as a result of uncertainty in model parameter estimation, (ii) to develop computationally efficient methodologies for identifying the process Design Space – i.e., the combination of input conditions within which the process is safely operated – while accounting for variability in critical input factors entering the system, and (iii) to improve the predictive capabilities of mechanistic models by leveraging available process data.