Towards mass customising respiratory protective equipment via additive manufacture: a scalable and automated design process

Abstract

Respiratory Protective Equipment (RPE) such as disposable N95/FFP3 masks used in the healthcare industry, or re-usable elastomeric respirator masks used in construction, oil & gas, firefighting industries require an effective seal to ensure best performance. However, problems such as poor compliance and skin injuries have been commonly observed among users as existing commercial RPE fit poorly on individuals’ faces. Additive Manufacture (AM) has been proposed as a promising alternative to produce custom-fitted RPE for individuals. However, the cost of AM remains prohibitive for adoption at scale. While research studies have been mainly focused on the development of better and more efficient AM technologies, little has been done to make the AM design process more cost-efficient and scalable. Of the existing design methods, most have been semi-automated processes which still involve considerable design time and manual efforts.

This thesis aims to develop a scalable and automated AM design process, and therefore making it more feasible to adopt AM for the mass customisation of RPE. Specifically, four data acquisition methods were evaluated to identify an appropriate one for obtaining geometrically accurate 3-Dimensoinal facial data with minimal resources needed. More importantly, an automated design pipeline based on the use of a template facial mesh was developed to bring facial meshes into dense correspondence and create customised RPE Computer-Aided Design models in large quantities. The template facial mesh was subsequently optimised considering geometric accuracy, surface quality of the fitted mesh and computational time. Next, the universality of the pipeline has been validated against a database of 205 facial scans obtained from people of different demographic backgrounds. Finally, the pipeline was further validated through conducting physical experiments to compare the performance of the 3D printed custom-fitted RPE against commercial RPE.