关键词: 服装设计, 马面裙, 结构设计参数, 下坠感, 坠感指数, 多元线性回归分析

Abstract:

Objective Current research on the horse-face skirts has primarily focused on craftsmanship, style, and fabrics, with limited attention to its wearing comfort. In order to provide a more intuitive understanding of the comfort associated with the horse-face skirts and to quantify the drape sensation caused by its complex structure and heavy fabric, this study introduces the concept of the "drape index" (defined as a metric reflecting the intensity of downward sagging during wearing). By establishing a mathematical model, the research further investigates the relationship between the design parameters of the horse-face skirts and the drape index, analyzing how the skirt's structure influences its drape sensation. This provides theoretical guidance for optimizing the design and improving the wearing comfort of the horse-face skirts.

Method The definition and calculation formula of the drape index were analyzed, and the definitions of the design parameters of the horse-face skirts and their regression relationships were listed. The area calculation formula for the horse-face skirts was proposed. A total of 900 experimental data sets on the structures of horse-face skirts were collected. Correlation analysis and multiple regression analysis were employed to establish a mathematical model between the structural design parameters of the horse-face skirt and the drape index, thereby analyzing the influence of the skirt structure on the drape sensation.

Results The drape index of the horse-face skirt was determined by its total area. Through the correlation analysis of the structural design parameters and the drape index, it was found that the Pearson correlation coefficient between skirt width and the drape index was 0.257, with a significance level of 0.183. This indicates that the change in skirt width has a minimal effect on the drape index, and no significant correlation exists between the two. This result may be related to the characteristics of the skirt width as a planar structural component of the skirt, which occupies a relatively small area and has a more uniform force distribution, thus having a limited impact on the overall drape sensation. The remaining four variables (skirt length, waist size, number of pleats, and pleat overlap) showed significant correlation with the drape index, which were used as independent variables to construct a linear model. According to the multiple linear regression analysis, the adjusted coefficient of determination for the model incorporating skirt length, waist size, number of pleats and pleat overlap reached 0.970, indicating these parameters collectively explain 97% of the variance in the drape index. Additionally, the Durbin-Watson statistic was 1.963, suggesting that the sample parameters of the horse-face skirt are independent and that there is no autocorrelation among them. The contribution rates of skirt length, waist size, number of pleats, and pleat volume to the drape index were 18.4%, 8.5%, 65.4% and 68.9%, respectively. Both pleat volume (68.9%) and number of pleats (65.4%) demonstrated particularly significant effects, while skirt length (18.4%) and waist size (8.5%) showed relatively minor impacts. This phenomenon may be attributed to how pleat volume directly affects the density of skirt folds and fabric stacking degree, thereby increasing the skirt's weight and drape sensation. The established mathematical formula confirmed that all four variables positively influence the drape sensation of the horse-face skirt, with pleat volume having the greatest effect. The regression coefficients provide quantitative guidance for adjusting structural design parameters, enabling an optimal balance between wearing comfort and aesthetic appeal.

Conclusion The proposed quantitative method for evaluating the weighted drape sensation of horse-face skirts offers theoretical guidance for the design of more comfortable versions of skirt and enriches the research on horse-face skirts. However, it only considers the influence of design parameters on the drape sensation. Future research is needed to explore the relationship between different styles of horse-face skirt structures and their drape sensations, as well as the influences of various fabrics and manufacturing techniques on the drape sensation.

Key words: fashion design, horse-face skirt, structural design parameter, drape sensation, drape index, multiple regression analysis