應用品質工程模式於產品感性設計

Abstract

本研究以品質導向、多品質/準則決策、二維品質Kano model與使用者滿意度之概念，取代傳統感性工學之單一品質/準則分析與品質與滿意度為線性的假設，系統性地探討產品造形參數、感性品質與使用者滿意度之關聯。研究針對感性品質與使用者滿意度提昇過程所面對的相關課題：重要造形因子的確定、感性品質度量與掌握模式之建立、使用者需求的定位與規劃、使用者需求的具體化，與使用者對需求達成滿意度的衡量等，系列性地進行探討。相關研究方法與結果如下： (1).首先，提出一整合產品形態特徵與構成關係之形態分析模式，進行水壺為例之感性工學探討，確認重要造形因子。研究結果驗證此整合分析模式，確能比單純考慮形態特徵之分析模式更有效地掌握使用者的感性與偏好。 (2).接著整合田口方法與灰關聯分析，以手機為例，進行美感品質最佳化模式的推導，以達到整體品質最大化與最小變異的穩健設計目標。結果證明了此灰色田口方法確能有效達成感性品質度量與穩健設計之目標。 (3). 為解決使用者需求定位與需求的具體化目標，整合田口與Kano model進行美感設計最佳化。研究先藉由權重方法取得各感性品質之權重，再應用「Kano品質雙向問卷調查法」對各感性品質予以分類，並根據品質判定結果進行權重的調整，取得使用者滿意度提昇所須的品質優先排序，有效地解決多品質權衡取捨，並進一步推論手機最佳化造形設計，有效提升使用者美感滿意度。 (4).為有效釐清使用者需求，應用Kano迴歸分析方法，探討產品感性品質績效與滿意度之間的不同相關性，以為感性品質優先排序判斷之依據。同時，提出一Extended Kano model，在原有之魅力的(A)、一元的(O)、必要的(M)、無差異的(I)與反向的(R)品質之外，提出一望目的(NB)品質分類，並將反向的品質細分為-A、-O、-M與-NB，更完整解釋使用者對於感性品質的需求特性。並進一步提出一資源分配模式，根據品質績效之優劣勢分析與品質之Kano分類結果，求得數量化之資源分配比例，提供設計者作多品質設計決策之最適化資源規劃。 (5).最後，比較Kano「迴歸分析」推論模式(多屬性效益函數)、「Kano滿意係數」推論模式與FAHP權重模式的使用者感性滿意度推論，並藉由驗證實驗，證明以「使用者需求」導向，「二維品質」Kano model為基礎之Kano迴歸滿意度推論模式，優於其他兩種模式。同時，利用數量化一類分析，建立造形因子/水準與不同Kano品質分類之數值關係。並整合分析結果與Kano model之二維品質觀點，提出設計造形類目選擇之判定準則，與「參數效益值分析」方法，協助設計者了解產品各造形因子對於使用者滿意度的影響性，以作為設計者手機造形設計最佳化與品質改善之設計參數選擇的權衡依據。 本研究藉由實際設計案例(水壺與手機)的分析與驗證，證明Kano品質模式、最佳化參數設計、多品質改善率推論與使用者滿意度推論模式、參數效益值分析方法等的應用，確能協助設計者解決產品感性設計之多屬性、多參數權衡，達到滿意度有效提升之目標。而研究所提出之相關方法與模式，亦可推廣至其他產品(品質)設計之應用。

In this study, we systematically explore the relationship between product forms, affective attributes, and affective user satisfaction with product design for resolving the problem of Kansei engineering on design optimization. Kansei engineering uses a uni-dimensional construction to deduce the ideal product design for achieving the desired performance for a specific attribute, while without considering the relationships between affective attributes and user satisfaction. To overcome these shortcomings, the robust design approach and two-dimensional Kano model are applied for resolving the difficult trade-off of multiple-attribute decision making for effective enhancing user satisfaction. The issues related with user satisfaction are systematically discussed in this study, including: identifying the user needs and important design parameters, constructing the quality model of affective attributes of product design, constructing the optimal design model for affective user satisfaction, and building the affective user satisfaction inference model. The results of the research are described as following: (1).Firstly, we propose a modified kansei engineering system based on the form element features and feature composition to effectively explain the users’ affective evaluations on product design. Using the kettle designs as examples, a series of statistical analyses was conducted to compare the difference between the modified model and the traditional model, a form element feature-oriented model. The result indicates that the modified kansei system is more valid on explaining users’ perception of affective attributes. (2).For enhancing aesthetic user satisfaction with considering multiple criteria and reducing the variation of users’ evaluations simultaneously, the robust design approach which combining grey relational analysis and Taguchi method is applied to optimize the aesthetic quality of mobile phone designs. The result of confirmation experiment concludes that the optimized design of multiple aesthetic criteria characteristics has a superior performance on quality maximized and variation minimized in overall performance. It verifies that the robust design approach can be applied to measure and manage not only objective, but also subjective quality. (3). A robust design approach incorporating the Kano model is proposed to obtain the optimal combination of design form elements. The Kano model is used to better understand the relationship between performance criteria and user satisfaction. A weight adjustment process based on Kano model analysis is proposed to magnify the weights of higher-return criteria in increasing user satisfaction. This process guides the prioritizing of multiple criteria, resolving the trade-off situation of multiple criteria optimization and leading to higher user satisfaction. The proposed approach then is applied to infer the optimal mobile phone design of aesthetic quality with multiple criteria characteristics. The results show that the generated optimal design can effectively enhance overall aesthetic performance and user satisfaction. (4).The regression method of Kano classification is used to help designers explore the different effects on user satisfaction, prioritize attributes and allocate resource in this study. According to the Kano model, attributes should be classified into different categories, must-be(M), one-dimensional(O), attractive(A), indifferent(I) and reversal qualities(R), depending on their ability to create user satisfaction or cause dissatisfaction. Considering the difference of user perception between subjective and physical quality attributes, an extended Kano model comprising one additional quality types beyond the conventional Kano model, i.e. nominal-the-better (NB) and further distinguishing the reversal qualities(R) into four sub-types (-A, -O, -M and -NB), is proposed and discussed. Result shows that the proposed extended Kano model, can appropriately classify quality attributes into suitable quality categories in a systematic and comprehensive manner. While, a formula combining a decision matrix with the Kano classification results was further proposed to infer the proper proportions which should be allocated to each attribute for improvement. It can help designers resolve the trade-offs dilemma in multiple-attribute optimization of user satisfaction and direct the improvement efforts of attributes in different competitive situation in the development of a new product design. (5).Finally, the model integrated with the additive multi-attribute utility function and Kano model is proposed to infer affective user satisfaction with mobile phone designs. The regression method of Kano classification is applied to explain the quantitative relationship between attributes performance and user satisfaction in this model. To verify the validity of the proposed model, the inference models of the additive weighting method, FAHP and Kano’s coefficient weighting method, are also constructed for comparisons. The result of verification shows that the proposed integrative model based on the Kano model can more accurately infer the affective user satisfaction than the other models. While, Quantitative Theory Type Ι analysis was used to explore the effect of each form level belonging to each form item on different Kano quality classifications. A parameter utility method based on Kano model is also proposed. The proposed approach then is applied to infer the optimal mobile phone design of affective user satisfaction with multiple criteria characteristics. The results show that it can help designers resolve the trade-offs dilemma of choosing the optimal combination of product form design in multiple-attribute optimization of user satisfaction. Results of this study verify that the application of Kano model and these proposed models can help designers to better understand user affective requirements, to resolve the trade-off dilemma in multiple criteria decision making and to prioritize actions and resource allocation to guide ideal product conception for enhancing maximum affective user satisfaction. Although these proposed approaches in this study use mobile phone designs and kettle designs as examples, with some adjustment, they may be equally applicable to other product (quality) design with multiple criteria characteristics.