Investigation of Distributed Additive Manufacturing for 3D Printed Headband Component During Global Supply Chain Disruptions

Abstract During the COVID-19 global pandemic, a shortage of personal protective equipment (PPE) occurred. Manufacturers of these critical products were unable to keep up with the growing demands as the situation evolved. Most face shields, a common type of PPE, include a headband component to provide the appropriate fit for the user and curvature for the clear face shield attachment. To mitigate this shortage in the local region, the Portsmouth Naval Shipyard (PNSY) generated a computer-aided design (CAD) file for the headband component and asked the University of New Hampshire (UNH) to manufacture it with all available 3D printers, including printers housed off-campus. Despite using the same part file, the headbands exhibited inconsistencies, particularly in part quality and performance. These variations in manufactured products can be attributed to the different machines, locations, materials, printing parameters, and personnel utilized in their production. This disparity in part quality resulted in the scrapping of approximately 5% of the 2000+ headbands printed at UNH, an unacceptable manufacturing metric. To address quality concerns, the 3D printer operators manually altered the original part file’s cross-sectional area in CAD software through a trial-and-error approach until the printed parts from each printer met the quality specification. This process of manual adjustment hindered the fast setup times and low costs attributed to distributed manufacturing resulted in essential PPE being scrapped. Ideally, using smart manufacturing, the part file geometry could be updated automatically, without further user inputs, based on the material properties used in the 3D printing. This could be accomplished by the implementation of an analytical model into a script for the CAD software that would calculate and update the part dimensions. In this paper, a case study to support distributed additive manufacturing for a simplified PPE headband during a global supply chain crisis, e.g., pandemic, is presented. Headbands were printed using four different material filaments and the original part file from PNSY to collect preliminary data. A finite element model was created to predict the force-displacement curves exhibited by headband parts produced using the different materials. An analytical model was derived to determine the geometry modifications required to achieve the desired part quality, i.e., force-displacement curve. Experimental validation was conducted and revealed good agreement with the analytical model predictions. In future work, these updates to the part geometry will be completed autonomously, using cyber manufacturing, by implementing the analytical model into CAD software via a script. This analytical model allows for distributed manufacturing to get past the hindrances of 3D printing manufacturing variances, advancing towards Industry 4.0.