Machine learning models for predicting the hardness of additively manufactured Polymethyl Methacrylate (PMMA) parts for biomedical applications

Additively fabricated Polymethyl Methacrylate (PMMA) is highly preferred for medical applications. Its biocompatibility makes it the preferred material for biomedical applications such as prosthetics and dentistry. It is crucial to understand the mechanical properties of such materials when used in biomedical applications. Determining the mechanical properties of parts is a time-consuming and costly process. Additive manufacturing involves various process parameters that directly influence the mechanical properties of the fabricated parts. Hence, relying on traditional methods to determine the mechanical properties of additively fabricated parts would be ineffective. With advancements in technology, optimizing the process parameters for additive manufacturing of components would simplify the manufacturing process. Similarly, to determine the mechanical properties of materials, technological advancements must be integrated with additive manufacturing. Although non-destructive tests exist, they have limited applications and are too costly. This investigation analyses the feasibility of using machine learning models to predict the hardness of additively fabricated polymethyl methacrylate. Layer height, infill density, infill pattern, and raster orientation or infill line direction strongly influence mechanical properties and are considered variable input process parameters. The influence of each parameter on the hardness of the material is described. The machine learning models are evaluated using metrics to determine the best-fit model. This investigation would help medical experts determine the most suitable process parameters for additively fabricating Polymethyl Methacrylate parts for biomedical applications. Additionally, this investigation would help experts vary the hardness of biomedical parts according to the requirements or specific applications.