運用資料探勘建立國軍服裝尺碼系統

Using Data Mining to Establish a Sizing System for Military Uniforms

鍾宜樺(Yi-Hua Chung) ； 鄭志展(Chih-Chan Cheng) ； 溫志皓(Chih-Hao Wen)

《人因工程學刊》 22卷1期 (2020/12) Pp. 45-59

https://doi.org/10.6273/JES.202012_22(1).0004

摘要

本研究運用二階段集群分析，將國軍人員進行體型分群，進而建立國軍最適尺碼系統，並運用總損失數作為尺碼系統配適性的衡量指標。研究結果如下：（1）新的尺碼系統具有高樣本涵蓋率及尺碼組數少等優點；（2）計算總損失數均低於理想值範圍3.59cm（男性：3.13cm；女性：3.12cm），顯示該尺碼系統配適性較符合國軍人員之體型。研究結果不僅提升服裝套量便利性，還可提供補給單位服裝生產之參考依據，進而減輕生產與庫存成本。最後，本研究製作專屬APP，除能讓使用者能快速獲得所需要的尺碼，還能提供補給單位未來在新、舊尺碼銜接時，縮短尺碼轉換時間，增進籌補效能。

關鍵字

制服 ； 3D人體掃描儀 ； 人體計測 ； 尺碼系統 ； 集群分析

並列摘要

This study uses a two-stage cluster analysis to clustering the national military personnel into groups, and then establish the optimal sizing system for military uniforms., and the index of aggregate loss was used as a measure of the fitness of the sizing system. The research results are as follows: (1) the new sizing system has the advantages of high sample coverage and small number of size groups; (2) The calculated aggregate losses are lower than the ideal value range 3.59cm (male: 3.13cm; female: 3.12cm), showing that the fitness of the sizing system is more suitable for the national military personnel. The research results not only improve the convenience of measure clothes, but also provide a reference basis for future supply unit clothing production, thereby reducing the pressure on production and inventory costs. Finally, this research makes an exclusive APP, which not only allows users to quickly obtain the required size, but also provides the supply unit with the ability to shorten the size conversion time and increase the replenishment efficiency when new and old sizes are connected in the future.

並列關鍵字

Uniform ； 3D Body Scanner ； Anthropometric ； Size System ； Cluster Analysis