Application of model‐based design of experiments for process development of solid oral dosage forms

Abstract Design of experiments (DoE) has been used extensively for strategic experimentation and process development in the pharmaceutical industry. Conventional DoE approaches, while foundational, often require extensive resources and do not fully utilize existing system knowledge. In this work, we demonstrate the use of a continuous effort‐driven, discrete model‐based DoE approach that calculates multiple locally optimal experiments from discretized control variable ranges by leveraging existing process knowledge. This workflow enables efficient experimental space exploration and parallel experimentation, reducing development times and material costs significantly and elucidating input–output correlations that may not be obvious without prior knowledge of the process model. Our work establishes that traditional statistical DoE constructs are neither superior nor necessary in advancing process development when an initial process model (prior knowledge) is available. The regulatory expectation that a DoE must resemble a fractional factorial is misled and only driven by legacy practices of empirical process development approaches.