道路鋪面的基本功能為提供使用者一個安全舒適的道路品質，因此道路使用者的觀感為維持道路服務性之重要評估因子，其通常以路面平坦程度決定道路鋪面之舒適度；此外，鋪面會因交通量、雨量及氣溫等因素產生多種不同的破壞型式，然國內目前對於鋪面狀況的評量與維護工作，仍存有缺乏長期路面資料、資料紀錄不良及未有準確評估模式等諸多問題，故鋪面服務性之維護與預測作業尚未臻完善；綜上所述，探討鋪面破壞型式與道路使用者觀感間的關係及維持良好的服務性指標，甚而預測未來鋪面服務性為相當值得深究的議題。
以往較常用來評估及預測鋪面的模式為統計迴歸方法，如美國工程師Carey及Irick於1960年利用迴歸模式，建立主觀資料PSR與儀器量測資料間的關係之PSI迴歸式濫觴(Chang et al, 2003 ) ；國內吳政隆與江煌基於1995年分別以台北地區及高速公路為案例建構現況服務能力指標PSI迴歸式。惟PSR指標值及其他破壞等級之判定存在著模糊語意性質，且傳統迴歸並不能處理帶有人為不確定或語意性的檢測結果；故使用模糊迴歸方法可將具有模糊或語意性質的資料提供有效的量化處理，進而發展PSI迴歸式。
本研究旨在探討鋪面服務能力指標與破壞型式的關係，並應用Pan and Hwang於2008年提出適用於本研究特性之可處裡模糊因變數及混合型自變數的模糊迴歸式。此外，實測個案為國立成功大學週邊道路(前鋒路、勝利路)進行市區道路檢測、道路現況服務力評分PSR(Present Serviceability Rating, PSR)、目視破壞項目檢測、車轍試驗及平坦度等試驗進行數據蒐集與分析。
本研究應用模糊迴歸模式建立評估鋪面服務性指標之PSI迴歸式，且依據現場試驗與模式分析結果發現，平坦度項目影響鋪面服務能力指標最鉅，其次則為人孔數分佈的多寡；再者，透過模糊判定係數R2檢驗發現，本研究所建置之模糊迴歸模式對於PSR與各破壞型式間的關係有70%左右的解釋能力，故可提供鋪面管理者有效的評量鋪面服務性及制定鋪面維修和養護策略之依據。

The main objective of pavement construction is to provide the safe and comfortable quality of the road for users. The roughness of pavement is an important effect to the users’ feeling of the road. However, the pavement condition deteriorates because of traffic, rainfalls, temperature, etc. The estimation and maintanence of pavement condition domestically has some problems, for example lack of long-term pavement data, incompleted records, unprecise models etc. In conclusion, considering the relationship between deteriorated pavement condition and the users’ feeling of road, maintaining good pavement serviceability and predicting pavement condition are worthy subjects to research.
In the past, conventional regression model often had been used in estimating pavement condition. For example, Carey and Irick in 1960 established PSI(Present Serviceability Index) regression model to deal with the relation between PSR (Present Serviceability Rating) and objective measurement(Chang et al, 2003). But PSR and other deteriorated pavement condition is evaluated by the subjective judgment. The conventional regression model cannot deal with uncertainty and fuzziness data ,so fuzzy regression model is used to establish the PSI regression model and handle this kind of problems.
This study aims at disscussing the relationship between PSR and the deteriorated pavement and applying Pan and Hwang’s fuzzy regression model in 2008 to develop fuzzy PSI(Present Serviceability Index) regression model. This model can deal with the fuzzy dependent parameters and fuzzy/crisp independent parameters.
Based on the result of this study, roughness affect pavement serviceability mostly and manholes affect secondly. Therefore, this fuzzy regression model in this study has 70% explained the ability in relationship between PSR and the deteriorated pavement condition through R-square. It can help pavement managers to estimate pavement serviceability effectively and order suitable pavement maintenance strategies.

參考文獻
一、外文部分(A~Z)
1. ASTM D6433-99, “Standard Practice of Roads and Parking Lots Pavement Condition Index Surveys”, 1999.
2. Ahmed, K., Abu-Lebdeh G. and Lyles, D.W., “Prediction of pavement distress index with limited data on causal factors：An auto-regression approach”, International Journal of Pavement Engineering, 2007.
3. Butt, Abbas A., Mohamed Y. Kieran J. Feighan, and Samuel Carpenter, H., “Pavement Performance Prediction Model using the Markov process.”, Transportation Research Record 1123, Transportation Research Board, 1985.
4. Cook, W.D and A. Kazakov. Pavement Performance Prediction and Risk Modeling in Rehabilitation Budget Planning: A Markovian Approach. Proceedings, Second North American Conference on Managing Pavements, Vol. 2, Toronto, Ontario, Canada, pp2.63~2.75, 1987.
5. Canadian Good Roads Association, “Pavement Evaluation Studies in Canada,” Proc. First International Conference on Structural Design of Asphalt Pavement ,University of Michigan, 1967.
6. Canadian Good Roads Association, “Field Performance Studies of Flexible Pavement in Canada,” Proc. First International Conference on Structural Design of Asphalt Pavement ,University of Michigan, 1971.
7. Chang, Jia-Ruey, Tzeng, Gwo-Hsiung, Hung, Ching-Tsung, and Lin, Hsin-Hwa,
”Non additive Fuzzy Regression Applied to Establish Flexible Pavement Present Serviceability Index”, The IEEE International Conference on Fuzzy Systems, 2003.
8. Chu, Chih-Yuan., Pablo, L., and Durango-Cohen., “Estimation of infrastructure performance models using state-space specifications of time series models”, Transportation Research Part C, pp17~32, 2007.
9. Chen, S.J. and Hwang, C.L., “Fuzzy Multiple Attribute Decision Making, Springer-Verlag”, New York. , 1992.
10. Dubois, D. and Prade, H., “Ranking fuzzy number in the setting of possibility theory”, Information Science, pp183~224, 1983.
11. Federal Highway Administration, “DISTRESS IDENTIFICATION MANUAL for the Long-Term Pavement Performance Program”, 2003.

12. Finn, F., “Pavement Management Systems ?Past, Present, and Future”, National Workshop on Pavement Management in New Orleans, July 20, 1997.
13. Fwa, T.F. and Gan, K.T., “Bus-Ride Panel Rating of Pavement Serviceability”, ASCE Journal of Transportation Engineering, Vol. 115, No.2, 1989.
14. Federal Highway Administration. “1999 Status of the Nation’s Highways, Bridges, and Transit：Conditions and Performance Report.” Federal Highway Administration. Report FHWA-PL-08-017, Washington D.C. 2001.
15. Finn, F., ”Pavement Management System？Past, Present, and Future.”, Public Roads, Vol 62, July/August, 1998.
16. Faraway, J.J., “Linear Models with R”, Compman and Hall/CRC Company, New York, 2005.
17. Gunaratne, M., Chameau, J.L., and Altschaeffl, A.G., “Introduction to Fuzzy Sets in Pavement Evaluation”, Transportation Research Record 985, pp22~24.
18. Gendreau, Michel and Soriano, Patrick, “Airport Pavement Management System：An Appraisal of Existing Methodologies”. Transpn Res.-A, Vol.32, No.3, pp.197~214,1998.
19. Haas, R. and Hudson, W.R., “Pavement Management Systems, ” Robert E. Krieger Publishing Company, 1986.
20. Janoff, M.S., “Methodlogy for Computing Pavement Ride Quality From Pavement Roughness Measurements”, Transportation Research Record 1084, 1986.
21. Klir, G. J. and Folger, T.A. “Fuzzy sets, uncertainty and information”, NJ：Prentice-Hall, 1988.
22. Lou, Z., Gunaratne, M., Lu, J.J., and Dietrich, B., Members, “Application of Neural Network Model to Forecast Short-term Pavement Crack Condition：Florida Case Study”,Journal of infrastructure systems, December 2001.
23. Luo, Zairen, and Eddie Chou, Y., “Pavement Condition Prediction Using Clusterwise Regression”, TRB 2006 Annual CD-ROM, 2006.
24. Moore, R.J., Clark, G.N. and Plumb, G.N., “Present Serviceability-Roughness Correlation Using Rating Panel Data”, Transportation Research Record 1117, 1987.
25. Neter, J., Kuntner, M.H., Nachtsheim, C.J. and Wasserman, W. “Applied Linear Regression Models”, McGraw-Hill, New York, 1999.
26. Nick, J.B. and Janoff, M.S., “Evaluation of Panel Rating Methods for Assessing Pavement Ride Quality”, Transportation Research Record 946, 1983.
27. Pan, N.F. and Hwang, K.C., " A linear regression analysis with fuzzy-inputs and fuzzy-output model", Fuzzy sets and Systems (Accepted) , 2008.
28. Pan, N.-F., Yang, M.-D., Hsu, K-C, "Estimates of bridge girder conditions based on fuzzy inspection data", Proceedings of the 5th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD'08), IEEE, 2008, Jinan , China (EI indexed).
29. Shekharan Raja, A., “Effect of Noisy Data on Pavement Performance Prediction by Artificial Neural Networks”, Transportation Research Record 1643,pp7~13, 1998.
30. Shekharan Raja, A., “Assessment of Relative Contribution of Variables to Pavement Performance Prediction by Artificial Neural Networks”, Transportation Research Record 1655,pp35~41, 1999.
31. Shoukry Samir, N., Martinelli David, R., and Reigle Jennifer, A., “Universal Pavement Distress Evaluator Based on Fuzzy Set”, Transportation Research Record 1592, pp180~186, 1997.
32. Samuel, Owusu-Ababio., “Modeling Skid Resistance for Flexible Pavements：A Comparison Between Regression and Neural Network Models.” Transportation Research Record 1501, p60~71, 1996.
33. Smith, R.E., “Structuring a Microcomputer Based Pavement Management System for Local Agencies, ” Ph.D. dissertation , University of Illinois at Urbana-Champaign, IL,1986.
34. Shahin, M. Y., “Airfield pavement distress measurements and use in pavement management”, Transportation Research Record 893, pp.59-63, 1982.
35. Shahin, M.Y., Darter, M.I. and Kohn, S. D., “Evaluation of airfield pavement condition and determination of rehabilitation needs.” , Transportation Research Record 700, pp.1-10, 1997.
36. Shahin, M. Y. and Becker, J. M., “Development of performance prediction models for airfield pavements. ” , Transportation Research Record 985, 1984.
37. Shahin, M. Y., “Pavement Management for Airports, Roads, and Parking Lots” , Chapman & Hall, New York, London, 1994.
38. Saraf, C.L., “Pavement Condition Rating System Review of PCR Methodology”, Report OH-99/004, FHWA, Ohio Department of Transportation, 1998.
39. The American Society for Civil Engineers, Report card for America ’s infrastructure, 2007, available at http://www.asce.org/reportcard/2005/page.cfm?id=22, accessed 12/14/2007 .
40. Tanaka, H.,Uejima, S. and Asai, K., “Fuzzy linear regression analysis for fuzzy input-output data”,IEEE Transactions on System, Man and Cybernetics, Vol.12, No.6, pp 903-907, 1982.
41. Ullidtz, P. and Larsen, B. K., “Mathematical model for predicting pavement performance.” Transportation Research Record 949, p45~55, 1983.
42. Wong, Wing-gun, and He, Guiping, “Gray Evaluation Method of Concrete Pavement Comprehensive Condition”, Journal of Transportation Engineering, November/December 1999.
43. Wang, Kelvin and Liu, Fengxian, “Fuzzy Set-based and Performance-Oriented Pavement Network Optimization System.” Journal of infrastructure systems, December/1997.
44. WSDOT Pavement Management and Condition Rating Systems, available at http://training.ce.washington.edu/WSDOT/Modules/09_pavement_evaluation/09-6_body.htm.
45. Yu, Jianiong, Eddie Chou, Y.J., and Luo, Zairen, “Development of Linear Mixed Effects Models for Predicting Individual Pavement Conditions”, Journal of Transportation Engineering, June 2007.
46. Yang, M.S. and Ko, C.H., “On a class of fuzzy c-numbers clustering procedures for fuzzy data”, Fuzzy Sets and Systems, Vol. 84, pp49-60, 1996.

二、中文部分
47. 周家蓓，「台灣地區一般公路鋪面養護管理系統建立之研究－第二期」，交通部運輸研究所，1993。
48. 吳政隆，「柔性鋪面現況服務力指標之研究」，碩士論文，國立台灣大學，1995。
49. 江煌基，「高速公路鋪面服務能力指標建構之研究」，碩士論文，淡江大學，1995。
50. 周少凡，「模糊集理論於鋪面表面狀況評估之應用」，碩士論文，國立成功大學土木工程所，1997。
51. 葉怡成，「應用類神經網路」，儒林圖書公司，1997.
52. 蘇木春、張孝德，「機械學習：類神經網路、模糊系統以及基因演算法則」，全華科技圖書股份有限公司，1999.
53. 曹銳勤，「模糊迴歸模式之建構與分析」，博士論文，國立交通大學，公管所， 1999。
54. 邱垂德，「路面平坦度驗收規範之檢討研究」，交通部國道新建工程局，2000。
55. 洪境聰，「柔性鋪面現況服務力指標與預測模式建立之研究」，碩士論文，國立中央大學土木工程所，2000。
56. 鄭昇旭，「柔性鋪面評分之探討--以高雄市中鋼路為例」，碩士論文，國立屏東科技大學土木工程所，2001。
57. 林勝傑，「類神經網路與馬可夫鏈理論於鋪面狀況評估之應用」，碩士論文，國立成功大學土木工程研究所，2001。
58. 交通部台灣區國道高速公路局，「中山高速公路路面檢視研究及現況評估」，期末報告，2001。
59. 林志棟，姚志廷，「國際糙度指標與鋪面破損之相關性分析」，土木水利第28卷第三期，2001。
60. 張家瑞，「建立台灣地區瀝青路面網級養護管理系統－以公路局中壢工務段為例」，博士論文，國立中央大學土木工程所，2001。
61. 林志棟，曾國雄，張家瑞，洪境聰，「以模糊迴歸方法建立柔性路面現況服務能指標」，第十一屆鋪面工程學術研討會，2001。
62. 「市區道路管理維護與技術規範手冊研究」，營建署，2002。
63. 鄭有良，「類神經網路於鋪面評估之應用」，碩士論文，國立成功大學土木工程所，2003。
64. 交通部，「公路養護手冊」，交通技術標準規範公路類公路工程部，2003
65. 宋宗勳，「柔性鋪面狀況指標檢測之研究」，碩士論文，國立中央大學土木工程所，2004。
66. 林元生，「智慧型鋪面檢測車應用於鋪面平坦度之研究」，碩士論文，國立中央大學土木工程所，2004。
67. 溫建龍，「維生管線孔蓋對道路平坦度影響之研究」，碩士論文，國立中央大學土木所， 2005。
68. 吳宜叡，「智慧型鋪面檢測車認證及鋪面平坦度之研究」，碩士論文，國立中央大學土木工程所，2005。