Process reliability and quality output are critical indicators for the upscaling potential of a fabrication process on an industrial level. Fused filament fabrication (FFF) is a versatile additive manufacturing (AM) technology that provides viable and cost-effective solutions for prototyping applications and low-volume manufacturing of high-performance functional parts, yet is defect-prone due to the inherent aspect of parametrization. A systematic yet parametric workflow for quality inspection is therefore required. The work presented describes a versatile and reliable framework for automatic defect detection during the FFF process, enabled by artificial intelligence-based computer vision. Specifically, state-of-the-art deep learning models are developed for in-line inspection of individual thermoplastic strands’ surface morphology and weld quality, thus defining acceptable limits for FFF process parameter values. We examine the capabilities of an NVIDIA Jetson Nano, a low-power, high-performance computer with an integrated graphical processing unit (GPU). The developed deep learning models used in this analysis use a pre-trained model combined with manual configurations in order to efficiently identify the thermoplastic strands’ surface morphology. The proposed methodology aims to facilitate process parameter selection and the early identification of critical defects, toward an overall improvement in process reliability with reduced operator intervention.

Additive manufacturing (AM) technologies enable the production of customized and personalized medical devices that facilitate users’ comfort and rehabilitation requirements according to their individual conditions. The concept of a tailor-made orthopedic device addresses the accelerated recovery and comfort of the patient through the utilization of personalized rehabilitation equipment. Direct modeling, with an increasing number of approaches and prototypes, has provided many successful results until now. The modeling procedure for 3D-printed orthoses has emerged as the execution of steady and continuous tasks with several design selection criteria, such as cutting, thickening the surface, and engraving the shell of the orthosis. This publication takes into consideration the aforementioned criteria and proposes the creation of a holistic methodology and automated computational design process for the customization of orthotic assistive devices, considering aspects such as material properties, manufacturing limitations, recycling, and patients’ requirements. This proposal leads to the designing and manufacturing of a wrist orthopedic device based on reverse engineering, Design for AM (DfAM), and Design for Recycling (DfR) principles. The proposed methodology can be adjusted for different limbs. A dual-material approach was attained utilizing rigid, mechanically enhanced feedstock material and soft elastic material with reduced skin irritation risks to achieve both mechanical requirements and adequate cushioning for user comfort during rehabilitation. Recyclable thermoplastic matrices were selected, which also allow for the option to create washable devices for product life extension. Then, 3D scanning procedures were implemented to acquire the initial anatomic measurements for the design of the WHO and ensure and assess the dimensional accuracy of the final product. Physical mechanical testing was implemented to evaluate the WHO’s mechanical behavior and verify its functionality during basic wrist movements. The extracted dimensional data for the two main orthosis components that indicated approximately 50% and 25% of the tolerance values, respectively, were within the range (−0.1 mm, 0.1 mm).

Acoustic materials are widely used for improving interior acoustics based on their sound absorptive or sound diffusive properties. However, common acoustic materials only offer limited options for customizable geometrical features, performance, and aesthetics. This paper focuses on the sound absorption performance of highly customizable 3D-printed Hybrid Acoustic Materials (HAMs) by means of parametric stepped thickness, which is used for sound absorption and diffusion. HAMs were parametrically designed and produced using computational design, 3D-printing technology, and feedstock material with adjustable porosity, allowing for the advanced control of acoustic performance through geometry-related sound absorbing/diffusing strategies. The proposed design methodology paves the way to a customizable large-scale cumulative acoustic performance by varying the parametric stepped thickness. The present study explores the challenges posed by the testing of the sound absorption performance of HAMs in an impedance tube. The representativeness of the test samples (i.e., cylindrical sections) with respect to the original (i.e., rectangular) panel samples is contextually limited by the respective impedance tube’s geometrical features (i.e., cylindrical cross-section) and dimensional requirements (i.e., diameter size). To this aim, an interlaboratory comparison was carried out by testing the normal incidence sound absorption of ten samples in two independent laboratories with two different impedance tubes. The results obtained demonstrate a good level of agreement, with HAMs performing better at lower frequencies than expected and behaving like Helmholtz absorbers, as well as demonstrating a frequency shift pattern related to superficial geometric features.