(438g) In silico Process Characterization of Drug Product Filling Operations: Predicting Quality Attributes from First Principles | AIChE

(438g) In silico Process Characterization of Drug Product Filling Operations: Predicting Quality Attributes from First Principles

Process characterization of drug product filling operations is typically a time- and labor-intensive process requiring substantial manufacturing line down-time due to the large number of process inputs and lack of mechanistic understanding of the fluid dynamics and filler systems’ sensitivities to drug product properties (e.g. viscosity, density, surface tension), or variations in site, equipment type, settings, and ambient conditions.

As an alternative to empirical process characterization, we have developed a computational model which predicts process quality attributes – for instance flow rate, fill volume/weight, cavitation, droplet formation, and product concentration – from first principles of product dynamic properties, fluid dynamics, structural mechanics and control theory. Formulation of the model couples finite element and finite volume solutions to structural deformation and fluid dynamics in pinch valves, and leverages analytical solutions to ordinary differential and algebraic equations accounting for orifice and pipe flow as well as equipment control laws.

We present an application of the model to a Time/Pressure filling operation, including formulation, deployment as a software application, and development of a workflow for in-silico process characterization utilizing design space response studies, local sensitivity studies, and uncertainty analysis.