(322f) Web Deployment of Mechanistic Models for Pharmaceutical Applications: Case Studies in Spray Drying and Distillation

In recent years, mechanistic models have been applied to a broad range of applications in pharmaceutical R&D and Engineering, improving efficiency through reduced data requirements, facilitating scale-up and tech transfer, and enabling virtual design space exploration. Despite their successes in these areas, mechanistic models have typically been utilized by a small group of modeling experts, who are not necessarily the key stakeholders in the answers that the models provide. As a result, scientists and engineers who are not themselves modelers have limited access to these tools and are less likely to take advantage of their full potential. Some common barriers to adoption include a lack of training, support and know-how, a high amount of effort required to establish a fit-for-purpose model-based solution, and a limited access to software for installation. Some recent efforts in the modeling community have aimed at “democratizing” models – reducing the barriers to adoption for those without modeling expertise. Web-based deployment of simple modeling tools is proposed as a potential solution to this problem.

In this work, model deployment for pharmaceutical applications to scientists and engineers will be discussed in the context of two case studies performed at Eli Lilly: a spray drying operation and a batch distillation and collection process. The modelling workflow requirements of scientists and engineers without prior modeling experience will be compared with those of modeling experts. For each application, a fit-for-purpose model was developed and deployed as a simple web interface to answer targeted questions of interest to scientists and engineers. Outcomes of the deployment exercise will be presented, and future perspectives on democratization of mechanistic models for pharmaceutical applications will be discussed.