(342b) Using First-Principles and Data-Driven Models to Guide Inference of Pharmaceutical Reaction Data

Mathematical modeling of chemical reaction kinetics has been proven to aid the development of new reactions and processes. Chemical kinetic modeling is well-established principle in chemical engineering that uses fundamental knowledge of the reaction mechanism to predict conversion data. In the pharmaceutical drug development, elementary-type kinetics are hardly common, due to the nature of the complex organic reaction mixtures and low-level impurities. Thus, statistical modeling plays an important role in understanding the relationship between parameters and reaction profiles. Emerging new technologies, such as high-throughput platforms and auto-samplers, enable greater data collection to enrich our understanding of chemical reactions. As a result, statistical analysis has shifted from conventional end-point analysis to modeling the entire reaction profile using more advanced statistical models, such as the dynamic response surface model and Gaussian process regression model.

In this presentation, a rigorous and general modeling workflow is described on the application of kinetic models and statistical models to the same set of dynamic reaction data. In particular, a semi-parametric empirical model was chosen. An industrial case study is presented, as showcase of the performance and robustness of the two modeling approaches and their impacts, side by side, on parameter effect estimation, reaction robustness range finding, and reaction optimization and operation window prediction. Emphasis is placed on the impact of data and sampling on the model inference. New and innovative visualization techniques are shared in this article for efficient data and model result interpretation.