透天住宅正立面因子與美觀性價比之探討

To Interpret the Facade Factors and Aesthetics Price-Performance Ratio of Street Houses

趙世詠(SHIH-YUNG CHAO)

指導教授 ： 陳清山

朝陽科技大學/設計學院/建築系建築及都市設計碩士班/碩士(2020年)

摘要

根據營建署統計，全國有人居住的729萬戶住宅中，透天住宅比例達49.2%，可見透天住宅在台灣住宅類型中佔有相當比例。而人類是視覺動物，建築外觀給人的第一印象，對消費者決定是否購買，有決定性的影響力，立面甚至讓消費者產生對個案性價比的評定，因此本研究將探討透天住宅正立面設計與造價間的關係，以便瞭解如何在美觀和成本中取得平衡。

本研究蒐集252個台灣地區各時期透天住宅正立面案例，經由建築專家評分後建立資料庫，藉由模糊類神經網路學習專家的知識與經驗，建立推論模型，然而模糊類神經網路可以精確的推論，卻無法運算出各因子與正立面美觀之關聯程度，此時可以利用灰色理論中的灰關聯度計算，分析各因子對於透天住宅正立面美觀性價比的影響。

影響透天住宅正立面美觀性之因子類型多樣且不易探討，故以主成份分析法將40項美觀影響因子，歸納為15項主成份，並運用灰色理論分別探討獨棟、雙併及連棟透天住宅主成份對美觀評分、單位造價、美觀性價比三項參考序列之關聯性，從研究結果可知，不同類型透天住宅之主成份，於三項參考序列中排序不盡相同，但由單一參考序列之平均值角度觀察，可以發現不同類型透天住宅與參考序列關聯性較高的主成份大致相同。

本研究亦運用模糊類神經網路與多元迴歸，推論獨棟、雙併、連棟透天住宅正立面之美觀評分、單位造價及美觀性價比。從分析推論結果可知，模糊類神經網路較多元迴歸適合用於推論透天住宅正立面之美觀評分、單位造價及美觀性價比。研究成果可供設計者與後續學術研究之參考。

關鍵字

透天住宅正立面 ； 主成份分析 ； 灰色理論 ； 美觀性價比 ； 模糊類神經網路 ； 多元迴歸

並列摘要

According to the statistics of Construction and Planning Agency, MOI,There are 7.29 million inhabited dwellings in Taiwan, the proportion of street house is 49.2%. From this statistics, it can be seen that street houses occupy a considerable proportion in the types of housing in Taiwan. Humans are visual animals, the first impression of the building's appearance has a decisive influence on consumers decision to buy the house. The building facade even promotes consumers to evaluate the Price-Performance Ratio of the case. Therefore, this study will explore the relationship between the design of building facade and the cost of street houses in order to understand how to strike a balance between building beauty and cost.

This study collected the facade data of street houses in various periods in Taiwan, and established a database after scoring by construction experts. Learning the knowledge and experience of experts by fuzzy neural networks, and establishing inference models. However, fuzzy neural networks can accurately infer, but they cannot calculate the degree of correlation between each factor and the facade of the building. At this time, the gray correlation coefficient in the gray theory can be used to analyze the impact of various factors on the aesthetic and the price-performance ratio of the facade of the street houses.

As the variety of influential factors of aesthetics. It is difficult to evaluate the price-performance ratio.The principal components analysis were taken to distinguish 40 elements into 15 principal components. These 15 principal components could be used to the grey theory and discussed the relevance between 15 principal components of street houses and the aesthetic score, unit cost, and aesthetic cost-effectiveness. From the results, it can be seen that the main components are ranked differently in the three reference sequences, but from the perspective of the average value of a single reference sequence, it can be found that the principal components are roughly the same.

This research also applied multiple regression analysis and fuzzy neural networks, inferring the aesthetic score, unit cost, and aesthetic cost-effectiveness. The conclusion can be found that fuzzy neural networks are more applicable than the multiple regression analysis to evaluate the price-performance ratio of street houses. The research results can be used by designers for reference, and the research methods are also can provide academics for references.

並列關鍵字

Facade of Street House ； Principal Component Analysis ； Grey Theory ； Price-Performance Ratio of Aesthetics ； Multiple Regression Analysis ； Fuzzy Neural Networks

參考文獻

中文參考文獻

(1) 耿相曾，1976，《盲人的適應》，泰成出版社，台灣。

(2) 吳玉成譯，Rob Krier著，1990，《建築元素》，胡氏圖書公司，台灣。

(3) 傅立，1992，《灰色系統理論及其應用》，科學技術出版社，北京。

(4) 陳淑貞譯，Mary Guzowski著，2001，《日光照明與永續設計》，六合出版社，台灣。