Scientists have developed a groundbreaking approach to accelerate the design of liquid formulations, such as shampoo, using machine learning (ML) technology. Liquid formulations are prevalent in various industries, including cosmetics, food, pharmaceuticals, and agrochemicals, but the complex interactions between ingredients pose challenges in customizing formulations to meet specific targets.
To address this issue, researchers have created an open experimental dataset with eighteen diverse formulation ingredients to train ML models for rinse-off formulations development. This dataset, consisting of 812 formulations, includes stable samples covering the entire design space. Phase stability, turbidity, and rheology measurements were conducted using a semi-automated, ML-driven workflow, with sample-specific uncertainty measurements to train predictive surrogate models.
The study focuses on liquid formulations, particularly shampoo, to explore the use of ML models in accelerating the design process. By selecting a range of surfactants, conditioning polymers, and thickeners commonly used in rinse-off products, researchers aim to empower ML models with the ability to predict formulation properties accurately. With over 800 formulations generated within eight months, the dataset provides valuable insights for training ML models for formulations properties prediction.
The research team adopted a high-throughput workflow utilizing automation and robotics to streamline the formulation design process efficiently. Combining ML-guided design of experiments with innovative liquid handling robots and analytical instruments, the workflow enabled the rapid generation of a chemically diverse set of formulations for comprehensive characterization.
Overall, the study showcases the potential of ML-driven approaches in formulating liquid products, with a focus on shampoo formulations. By leveraging advanced technologies and experimental datasets, researchers are paving the way for faster, more cost-effective formulation design processes with a strong emphasis on sustainability and eco-friendly ingredients.
