Cluster size prediction for military clothing using 3D body scan data
Stephven Kolose1, Tom Stewart2, Patria Hume1
1Sport Performance Research Institute New Zealand, Auckland University of Technology, Auckland, New Zealand.
Applied Ergonomics
|June 10, 2021
Summary
Anthropometric data from the New Zealand Defence Force revealed distinct body shape clusters for males and females. This research supports the development of tailored uniform sizing systems for improved fit and function.
Area of Science:
- Anthropometry
- Biomechanical engineering
- Garment science
Background:
- Developing effective uniform sizing systems requires understanding population-specific anthropometric variations.
- Previous sizing systems may not adequately represent the diverse body shapes within the New Zealand Defence Force (NZDF).
Purpose of the Study:
- To identify and characterize anthropometric clusters within the NZDF population.
- To provide data-driven insights for the creation of new NZDF uniform sizing systems.
Main Methods:
- Utilized anthropometric data (84 variables) from 1,003 NZDF participants (NZDFAS).
- Applied Principal Component Analysis (PCA) to determine key variables for clustering.
- Employed a combination of two-step and k-means clustering, stratified by gender, for analysis.
Main Results:
- PCA identified optimal variables for shirt and trouser sizing based on gender.
- Clustering resulted in 6 distinct clusters for female clothing and 10 for male clothing.
- Female clusters exhibited greater intra- and inter-cluster variability compared to male clusters.
Conclusions:
- Anthropometric clustering effectively partitions individuals into distinct groups.
- Identified anthropometric dimensions can guide the garment industry in developing specialized NZDF sizing systems.
