隨著住宅公寓大廈休閒化的趨勢，建築師於公寓大廈社區內設計許多公共設施來增加生活的機能，然而設施的增加也增加了營運成本，亦常有因為維營運成本過高導致公共設施無法繼續營運下去的問題。原因為建築師於規劃設計階段沒有預測營運成本的工具，來評估公共設施的適切性與住戶的負擔能力，造成公寓大廈於使用階段發生設施無法繼續營運的問題。過去的研究發現規劃設計決策決定了後續80%的營運成本花費；預測成本的方法有迴歸分析與類神經網路。所以本研究的主題為於規劃設計階段建立一個可以在規劃設計階段預測公寓大廈使用階段營運成本的預測模型，研究目的有：分別以迴歸分析與類神經網路建立模型，與比較兩個模型的預測能力。
本研究經由文獻回顧的方式歸納出可能影響日後營運成本的設計決策因子（屋齡、戶數規模、總樓層數、住宅平均單價、總樓地板面積、公共設施面積）。
並以田野調查的方式從65棟公寓大廈案例調查變數因子的資訊；透過55個案例分別以迴歸分析與類神經網路兩方法建立預測模型；最後以10個案例檢測與比較兩個模型之預測準確率。研究成果提出預測營運成本的迴歸分析與類神經網路模型，並發現使用類神經網路方法建立模型預測營運成本的準確率高於迴歸分析模型。本研究之貢獻在於提供建築師預測公寓大廈共用部分營運成本的工具，藉以評估公共設施規劃設計之適切性，及營運成本之合理性。

The lack of an predicting operation and maintenance cost method of apartment building for architect at preliminary design stage. This research intended to establish the operation and maintenance cost model. The objectives of this research are:
1. To establish the operation and maintenance cost model using regression method.
2. To establish the operation and maintenance cost model using ANN method.
3. To compare operation and maintenance costs of the apartment building predicted by regression model and neural network model.
This research used literature review to generalize that to influence operation and maintenance cost method variable at preliminary design stage: building age; total number of household; total number of floors; the average price; total floor area; total areas of public facility. This research used field study method to collect operate and maintain cost data of 65 apartment buildings. This research used 55 apartment buildings to part establish two models and check models ability with another 10 apartment buildings.
The study revealed that the ANN prediction model produces better predictions results as compared to the regression model. The results show that the ANN model outperforms the regression model used for predicting operation and maintenance cost of the apartment building.

1. 林惠玲，1999。應用統計學。雙葉書廊有限公司。
2. 高英欽，1996。中高層集合住宅日常管理費構成之研究-以台中市為例。國立成功大學碩士論文。
3. 徐春福，2002。公寓大廈共用部分機電設備維護費用支出之研究-以台中市為例。國私立中華大學碩士論文。
4. 陳信夫，1993。建築工程成本估算法之比較研究-與類神經網路估算法之研究。國立台灣大學碩士論文。
5. 陳覺惠，1999。公寓大廈共用部份適宜性與使用管理之調查研究-以台中市為例，內政部建築研究所專題研究計畫成果報告。
6. 游晉程，1996。集合住宅屬性與設施管理之關係研究。國立成功大學碩士論文。
7. 郭炳煌，2002。以統計方法與類神經網路模式預估工程直接成本之研究。國立高雄第一科技大學碩士論文。
8. 黃東寅，2001。休閒住宅之「用後評估」以休閒設施為例。私立中華大學碩士論文。
9. 葉怡成，1997。應用類神經網路。儒林圖書有限公司。
10. 賴榮俊，1992。公寓大廈住宅共用設施之研究-以生活及休閒設施為例。國立台灣大學碩士論文。
11. Bogenstätter, U. （2000）. Prediction and optimization of life-cycle costs in early design. Building Research and Information, 28 (5), p. 376-286.
12. Creesr, R. C. & Moore, L. T. （1990）. Cost modeling for concurrent engineering. Cost Engineering, 32(6), p.23-27.

13. Emsley, M. W., Lowe, D. J., Duff, A. R., Harding, A. & Hickson, A. （2002）. Data modeling and the application of a neural network approach to the prediction of total construction costs. Construction Management and Economics, 20(6), p.465-472.
14. Günayd1n, H. M. & Dogan, S. Z. （2004）. A neural network approach for early cost estimation of structural systems of buildings. International Journal of Project Management, 22(7), p.595-602.
15. Jiang, R., Zhang, W. J. & Ji, P.（2003）. Required characteristics of statistical distribution models for life cycle estimation. International Journal of Production Economics,83(2), p.185-194.
16. Kim, G. H., An, S. H. & Kang, K. I.（2004）. Comparison of construction cost estimating models based on regression analysis, neural networks, and case-based reasoning. Building and Environment, 39(10), p. 1235-1242.
17. Lin, Y. Y. & Liu, M.（2004）.Using neural network to predict performance of design-build projects in Singapore. Building and Environment, 39(10), p. 1263-1274.
18. Masters, T. （1993）. Practical Neural Network Recipes in C++. Boston: Academic Press.
19. Moore, D. S. （1991）. Statistics: concepts and controversies. Freeman and Company, New York.
20. Murray, S. （1993）. Neural Networks for Statistical Modeling. NY: Van Nostrand Reinhold.

21. Seo, K. K., Park, J. H., Jang, D. S. & Wallace, D. （2002）. Prediction of the life cycle cost using statistical and artificial neural network methods in conceptual product design. International Journal of Computer Integrated Manufacturing, 15 (6), p. 541-554.
22. Taylor, J. G.（1996）.Neural networks and their applications. John Wiley and Sons Ltd.
23. Thompson, B. & Borrello, G. M.（1985）.The importance of structure coefficients in regression research. Educational and Psychological Measurement, 45(3), p. 203-209.
24. Zurada, J. M. (1992). Introduction to Artificial Neural Systems. West Publishing Company, St. Paul, MN.