Entrepreneurship

Abstract: The high volume of plastic waste and the extremely low recycling rate have created a serious challenge worldwide. Local distributed recycling and additive manufacturing (DRAM) offers a solution by economically incentivizing local recycling. One DRAM technology capable of processing large quantities of plastic waste is fused granular fabrication, where solid shredded plastic waste can be reused directly as 3D printing feedstock. This study presents an experimental assessment of multi-material recycling printability using two of the most common thermoplastics in the beverage industry, polyethylene terephthalate (PET) and high-density polyethylene (HDPE), and the feasibility of mixing PET and HDPE to be used as a feedstock material for large-scale 3-D printing. After the material collection, shredding, and cleaning, the characterization and optimization of parameters for 3D printing were performed. Results showed the feasibility of printing a large object from rPET/rHDPE flakes, reducing production costs by up to 88%.

Abstract: The application of computer vision and machine learning methods for semantic segmentation of the structural elements of 3D-printed products in the field of additive manufacturing (AM) can improve real-time failure analysis systems and potentially reduce the number of defects by providing additional tools for in situ corrections. This work demonstrates the possibilities of using physics-based rendering for labeled image dataset generation, as well as image-to-image style transfer capabilities to improve the accuracy of real image segmentation for AM systems. Multi-class semantic segmentation experiments were carried out based on the U-Net model and the cycle generative adversarial network. The test results demonstrated the capacity of this method to detect such structural elements of 3D-printed parts as a top (last printed) layer, infill, shell, and support. A basis for further segmentation system enhancement by utilizing image-to-image style transfer and domain adaptation technologies was also considered. The results indicate that using style transfer as a precursor to domain adaptation can improve real 3D printing image segmentation in situations where a model trained on synthetic data is the only tool available. The mean intersection over union (mIoU) scores for synthetic test datasets included 94.90% for the entire 3D-printed part, 73.33% for the top layer, 78.93% for the infill, 55.31% for the shell, and 69.45% for supports.

Abstract: Open-source design of medical devices, following the concept of frugal engineering, provides unrestricted descriptions of technical details, allowing the low-cost and local fabrication of devices to reduce global inequities in healthcare.