This conference paper was submitted for presentation at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

The rising demand for high-performance medical devices, including hypodermic syringes, has created an urgent need for more efficient design and validation methods. Traditional approaches that rely heavily on physical prototyping are often time-consuming and expensive, which significantly limits innovation. However, advancements in machine learning (ML) and simulation-based techniques offer opportunities for optimizing these processes, enabling faster evaluation and improved accuracy in predicting device performance. In this study, we introduce a physics-driven machine learning (ML) model integrated with unified modeling and simulation to evaluate the performance of hypodermic syringes. Our objective is to minimize reliance on physical testing by effectively leveraging virtual prototypes and advanced predictive analytics. We first create a high-fidelity virtual twin of the syringe that complies ISO7886 standard for sustaining force, ensuring an accurate reflection of its physical behavior during operation. Subsequently, we generate comprehensive datasets through various sampling methods as part of a structured design of experiments. These datasets are utilized to rigorously train and test the ML model, which demonstrates impressive accuracy in predicting the sustaining force for new syringe designs, validated by several key performance metrics. Our systematic approach provides a flexible framework to streamline both the design and virtual validation processes for syringes, enabling manufacturers to efficiently adapt to evolving market demands. Additionally, this methodology allows designers to rapidly explore different design options and gain valuable insights through a template-based methodology. Ultimately, this research highlights the immense potential of integrating machine learning with physics-based modeling to transform the syringe design process, paving the way for innovative and efficient solutions in medical device manufacturing. This advancement not only enhances product development timelines but also contributes to improved patient outcomes through the reliable and safe use of hypodermic syringes in clinical settings, illustrating the importance of advanced modeling techniques in modern medical device design and development.