Systematic optimisation of carrier particle geometry in interactive powder mixtures via a parametric design approach using Bayesian optimisation
Year: 2026
Type: article
Abstract: In this study, a systematic approach for improving the carrier particle geometry in interactive mixtures was conducted. For the evaluation of the geometries, an in-silico model was used, which allowed the creation of a carrier particle loaded with drug particles and to investigate the crucial detachment of the latter in two different kinds of collision simulations. The carrier particle geometries were created with a parametric design tool that enabled continuous modification of the geometry. Data evaluation was performed with a Bayesian Optimisation, which could suggest potentially interesting new geometry modifications based on the evaluated results. The whole setup was implemented as a fully automated process. Particle geometry creation, procession of new geometries in the simulation software and evaluation of the results, including the setup of the following iterations, was performed by precise communication between the different software applications. As a result, it could be demonstrated that the systematic modification of the geometries improved performance in terms of drug detachment. Furthermore, characteristic geometry features responsible for the detachment could be identified.
Authors: Melvin Wostry, Regina Scherließ
Institution: Christian-Albrechts-Universität zu Kiel
Language: English
Subfield: Computational Mechanics
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,070
OpenAlex ID: W7124273446