Optimisation of Topical and Transdermal Pharmaceutical Formulations: Methods, Challenges and Opportunities

ABSTRACT Topical and transdermal pharmaceutical formulations are a large and diverse class of products, yet their optimisation remains a complex, resource‐intensive challenge. Statistical design‐of‐experiments approaches have long provided the methodological backbone for this process and have shown broad effectiveness across topical and transdermal formulation systems, but face growing demands as formulation complexity and the number of variables increase. Classical machine learning offers more flexible modelling of nonlinear formulation–performance relationships; adaptive sequential design adds uncertainty‐aware experiment selection; high‐throughput screening platforms address the data‐generation bottleneck and self‐driving laboratories integrate automated execution with model‐guided decision‐making. However, their application in topical and transdermal formulation remains uneven, and their advantages over established methods depend strongly on data quality, validation design and experimental context. This review examines optimisation approaches applied to topical and transdermal formulation development, from classical factorial designs and response surface methodology through machine learning and adaptive sequential design to high‐throughput screening and self‐driving laboratories.