NZDF Uniform Size Prediction Model and Application Development: Technical Report to the New Zealand Defence Force (NZDF)
Kolose, S; Stewart, T; Hume, P
MetadataShow full metadata
BACKGROUND Project scope As per the contract for work, this project involved the refinement of a previously created uniform size prediction model and the development of a functional computer application. The application (BodyClothApp) automatically converts the Human Solutions Ltd VITUS XXL 3D body scanner output into a NZ Army Multi Camouflaged Uniform (MCU) size prescription (both shirt and trousers) in real‐time. The purpose of the BodyClothApp was to simplify and increase the accuracy of the Army uniform sizing process. This project included the development of the BodyClothApp and trialling its usability at a NZ Army sizing activity in July 2019. PURPOSE To conduct a validation trial as part of the Army recruit sizing activity (TAD 395) in Waiouru on 30–31 July 2019. The trial tested whether the 3D body scan data and clothing size prediction model (embedded within the BodyClothApp) produced the same NZ Army MCU shirt and trouser size as that determined by the Workwear Group tailors. METHODS Design A cross sectional cohort study. Participants New NZ Army recruits part of the July intake undergoing basic training at the Waiouru military camp were invited to participate during their routine medical/dental assessment day. Four platoons, consisting of approximately 40 individuals each, were included in the study. The trial had ethics approval from the Auckland University of Technology Ethics Committee (AUTEC #14/126 NZDF anthropometry Survey: Variations in kinanthropometry and implications for the New Zealand Defence Force). The aims of the study, what to wear during the scan, and how to stand during the scanning process were explained to each participant. Written informed consent was provided by each participant, confirming they understood the study requirements before they participated. Participants could withdraw from the study at any time. Procedure and Measures BodyClothApp: The 3D body scan clothing prediction application The BodyClothApp is written in the R programming language. It is presented as a user‐friendly dashboard interface that runs alongside the Anthroscan (body scanner) software. The BodyClothApp can make individual predictions of MCU shirt and trouser size and display these immediately after an individual is scanned. The BodyClothApp also provides visual inspection of individual body measurements, group‐level results, and generates a PDF report with a summary of key results for each participant.