應用聯合分析法於商標設計之最佳化研究——以羅昇企業設計個案為例

The Application of Conjoint Analysis in Optimum Logo Design-A Design Project as an Example

平面設計師接受公司或組織委託執行企業識別體系的規劃、設計與建立時，如何正確決定構成企業商標的基本要素－－標誌、標準字及標準色，是影響企業的視覺識別能否成功傳達的重要關鍵。而公司決策者往往對於設計師所提出的商標設計提案未能及時達成共識而影響設計案的順利進行，更導致企業識別體系的實施受到延宕。因此本研究以進行中的商標設計案為例，選擇基本要素組合系統中最為常見的商標編排方式，應用聯合分析法尋求最佳的設計方案。由於商標設計屬自主性極高的創意行業，概念發展階段是不易解析聯合分析法所需界定之屬性和水準，因此尚未發現學界應用此方法於商標設計相關議題的研究。羅昇企業決策階層首先藉由群體討論的方式，選定設計提案中每個屬性（亦即標誌、標準字及標準色）層屬下的各個可行水準，以整體輪廓法組成12個設計選擇方案，同時進行視覺複雜性和偏好的李克量表問卷評比。調查結果需經由ANOVA變異數分析檢定三屬性間的交互作用是否具有統計上的顯著性，以決定方案評估的總體效用函數應採用之模式，進而求取各個準則的成分效用值和屬性間的相對重要性。研究結果顯示屬性間的相對重要性分別為：標誌82.21%，標準字9.56%及標準色8.29%；而此貢獻值也清楚說明個別屬性是如何地影響商標的整體呈現。同時利用問卷調查所得之數據進一步探討商標視覺複雜性與偏好的相關性。最後，並針對此次羅昇企業委託的商標設計案最終選擇結果及其決策過程，分析討論此模式在應用上的可行與限制。希望本研究成果有助於平面設計師從事商標設計時對不同組合系統的編排方式建立準則和參考依據，且應用此模式作為設計方案評價與選擇的有效工具。

A successful visual identity, especially in logo design, ensures the corporate identity system （CIS） performed and communicated efficiently and effectively. Three basic logo elements, including symbol, logotype, and color（s）, play the most influential roles in the initial stage of logo development. Many graphic designers of logos or trademarks often rely on their subjective emotion, esthetic perception, as well as experience, to generate design proposals. Unfortunately, the corporate identity generated with subjectivity rarely meet customers' image requirements because of their customers' perception may differ from that of the designers. Thus, this paper proposes a new approach to help designers and decision maker（s） of business in selecting an optimum logo design using conjoint analysis by means of the merit of its representation of relative importance. Thirty subjects participated to investigate the effects of a proceeding design project for symbol （outline, solid, and stripe appearances）, logotype （Chungi and Lishu fonts）, and color （PMS185 and PMS286） in visual complexity and preference tasks with twelve profiles （i.e. different combinations of attribute levels）. The results showed that conjoint analysis provided quantified preference functions for multi-attribute logo design alternatives, including the contributions of each attribute and attribute level to the overall logo utility and the relative importance of logo attributes: Symbol has a relative importance of 82.21%, the most preferred factor in logo design, followed by logotype （9.56%）. The least preferred attribute is color, being as similar as logotype, with an importance value of 8.29%. Simultaneously, an initial exploration of the correlation between visual complexity and preference is presented for future further studies. Finally, the proposed model constructed with analysis of variance （ANOVA） and conjoint analysis can make logo design easier and more successful.