Many apparel companies are currently faced with the demands of a mature marketplace while trying to satisfy customers who expect apparel that is highly customized. However, making a customized garment for each customer is typically not feasible. Thermal profiling solves this problem for thermal regulatory apparel by grouping consumers into specific thermal families. These thermal families define micro-consumer groups for which mass-customized functional apparel can be created. This study identified whether individuals can be grouped into thermal families, then examined which demographic and anthropometric variables were most associated with thermal profiling. Data mining procedures were used to extract individuals from a pre-existing database with a total sample size of 796. First, utilizing a hierarchical cluster analysis, each individual, by gender, was placed into one of three thermal families (uniform, v-shaped, and abdominal trough). These families were determined by clusters of spatial temperature distributions. Next, ten separate statistical goodness-of-fit measurements were employed to determine the "fit" of the individual to their assigned thermal family. Lastly, the thermal families were correlated with selected demographic and anthropometric variables. Results indicated that there were at least three distinct thermal families, and that individuals could be categorized by thermal profile. Neither age nor exercise frequency was correlated with a thermal pattern and therefore could not be used as a means of categorizing individuals into thermal families. In contrast, results indicated that several anthropometric measures could be used as a proxy to predict an individual's thermal family. In summary, for apparel companies seeking to gain a competitive edge, thermal profiling and the resultant thermal families may be a tactical advantage in a mature, competitive market.
|Journal||Journal of Textile and Apparel, Technology and Management|
|State||Published - 2013|
- Thermal imagining
- Thermal profiling 3D mapping data mining