應用灰關聯多屬性決策分析於風格典範評估之研究——以現代圈椅設計為例

The Application of Grey Multiple Attribute Decision-Making Method on Style Typicality Evaluation-Using Modern Loop Chair Designs as Examples

泛國際化與現代化的設計趨勢，造成產品普同化現象，使產品成為失去意義的造形。設計師應從自身文化中尋求原創本質，建立屬於自己的文化語言與符號，將傳統文化視為創作的基礎進行傳統風格的現代化。因此，本研究以圈椅設計為探討對象，應用灰關聯多屬性決策法於現代中國風格典範的評估，藉由原型的評選，進而探求營造現代圈椅風格的造形特徵與形式規則。研究結果指出，對於產品風格典範性，不同造形屬性具有不同的影響權重，而圈背、背靠板造形與造形比例具有較高的權重值，是形成風格辨識的主要特徵，負載最多的文化訊息。同時，造形特徵與規則的變形程度影響其風格典範績效。最後，藉由研究提出之灰關聯多屬性評估模式應用於現代圈椅典範的評估，其評選結果與使用者評價結果有相同之評估趨勢，以Spearman等級相關檢定，相關係數高達0.928，證明模式應用之有效性，可實際應用風格意象的評價與輔助文化產品設計之工具。

The international design and Modernism causing the product forms become homogenized and meaningless. Hence, designers should create the unique design languages and rules by mining from their own culture and tradition. And by modernizing traditional style to solve the problem of meaning communicating between products and users. In this study modern loop-chairs are used as examples to investigate the critical factors constituting the Neo-Chinese-image of products. We applied a Grey multiple attribute decision-making method on style typicality evaluation, and finding out the prototype of modern loop-chairs design and deducing the features and rules of style forming. The results indicate that （1）. Using AHP determined the weights of the attributes for style typicality evaluation. And these features-armrest, back-plank and scale, got higher weight value, are the critical features for perceiving and recognizing style image; （2）. Identifying the style typicality effect in feature attributes, and different degrees of attribute transformation mean to operate the style of products; （3）. Applying a gray MADM style typicality evaluation model to evaluate modern loop-chairs designs. And compared the evaluation performances between the model and user-survey, knowing the selecting orders are similar. Furthermore, using Spearmam correalation analysis to test the relativity. The Spearmam correalation coefficient is 0.928 high, indicating the selecting orders are very relative between user-survey and the model's performances. According to the results, proving this model is an efficient method to evaluate style typicality and aid to conduct the style of products for designers.