A focused simulation-based optimization of print time and material usage with respect to orientation, layer height and support settings for multi-pathological anatomical models in inverted vat photopolymerization 3D printing

Abstract

Background3D printing of anatomical models requires multi-factorial decision making for optimal model manufacturing. Due to the complex nature of the printing process, there are frequently multiple potentialities based on the desired end goal. The task of identifying the most optimal combination of print control variables is inherently subjective and rests on sound operator intuition. This study investigates the effect of orientation, layer and support settings on print time and material usage. This study also presents a quantitative optimization framework to jointly optimize print time and material usage as a function of those settings for multi-pathological anatomical models.MethodsSeven anatomical models representing different anatomical regions (cardiovascular, abdominal, neurological and maxillofacial) were selected for this study. A reference cube was also included in the simulations. Using PreForm print preparation software the print time and material usage was simulated for each model across 4 orientations, 2 layer heights, 2 support densities and 2 support tip sizes. A 90–10 weighted optimization was performed to identify the 5 most optimal treatment combinations that resulted in the lowest print time (90% weight) and material usage (10% weight) for each model.ResultsThe 0.1 mm layer height was uniformly the most optimal setting across all models. Layer height had the largest effect on print time. Orientation had a complex effect on both print time and material usage in certain models. The support density and the support tip size settings were found to have a relatively minor effect on both print time and material usage. Hollow models had a larger support volume fraction compared to solid models.ConclusionsThe quantitative optimization framework identified the 5 most optimal treatment combinations for each model using a 90–10 weighting for print time and material usage. The presented optimization framework could be adapted based on the individual circumstance of each 3D printing lab and/or to potentially incorporate additional response variables of interest.