在面對國道高速公路鋪面的面積與壽齡的逐年增加，而整修經費成長的幅度有限的挑戰下，為能有效管理公路鋪面，在提供用路人最高用路品質的同時有效管理所需之成本，目前國道高速公路局積極推動科學化與數位化鋪面管理工作，而其中最重要的工具即是鋪面管理系統，而鋪面管理系統中的一個重要核心功能為路網層級的鋪面績效預測模型。一般來說，鋪面績效預測模型的建立需要長期的鋪面績效(車轍、開裂、IRI等)資料，然而目前國道路網缺乏車轍與開裂資料，因此如何在缺乏長期鋪面績效資料下建立適合國道養護單位使用的鋪面績效預測模型實為目前迫切的課題。本研究的目的在建立路網層級國道鋪面績效預測模型架構，研究中利用AASHTO MEPEG模擬鋪面車轍、開裂、及IRI績效，並結合國道養護工程資料來建立模型參數。研究中拜訪高公局北、中、南分局，收集四個案例路段之工程資料，利用AASHTO MEPDG模擬在改變交通載重、材料特性、及鋪面結構等條件產生下的960種情境，對鋪面車轍、開裂、及IRI績效之影響，進而利用此些結果發展路網層級鋪面績效預測模型。此外，研究將會利用工程參數與車轍級開裂之間關係建立車轍及開裂鋪面績效預測模型，再利用所預測之車轍與開裂值建立IRI績效進行預測，所建立之車轍及開裂模型為包含各工程參數之多變數績效預測模型，IRI模型則為包含車轍、開裂及非結構性破壞項次之多變數績效預測模型，其模型之準確度高(均方根誤差皆為0.05m/km以下)，而精確度(R平方為0.55至0.9)相對有落差，但仍為可接受之範圍，由於模型之準確度相當高以及精確度仍為可接受之範圍，因此後續依照本研究所提出之模型與校正架構，持續收集現地鋪面績效數據優化模型，方可建立本土化模型參數。

A well-functioning and high-quality infrastructure system plays an important role in a country’s economy. In Taiwan, National Freeway Bureau (NFB) invests lots of resources every year to maintain driving quality and safety for publics in Taiwan. However, with aging infrastructure and slowing increasing budget, it needs a robust pavement management system (PMS) tool to facilitate activities scheduling and budget allocation of roadway maintenance and rehabilitation. The core of the PMS is the pavement performance prediction model. Therefore, the purpose of this study is to establish the network-level pavement performance prediction model framework for NFB network with little historical pavement condition data (rutting, cracking, IRI). Four sections of pavement were selected from three NFB maintenance district. The real engineering, traffic and climatic data of these four pavement sections were used as input for an uncalibrated AASHTO MEPDG model to obtain pseudo pavement performance data (rutting, cracking, IRI). These data were used to develop a semi-empirical multivariable prediction model formation that considered the engineering data availability of NFB. The results show that the model achieve RMSE and R-squared within an acceptable range at 0.05m/km and 0.55-0.9 respectively. Since the accuracy of the model is good and the precision is still within an acceptable range, in accordance with the model and calibration framework proposed by this research, the localized model parameters can be established by continuously collecting the performance data.

英文文獻
1. Abdullah Nur, M., Khattak, M. J., & Bhuyan, M. R. U. K. (2013). Rutting Model for HMA Overlay Treatment of Composite Pavements. International Scholarly Research Notices.
2. American Association of State Highway and Transportation Officials. (2010). Guide for the Local Calibration of the Mechanistic-Empirical Pavement Design Guide.
3. American Association of State Highway and Transportation Officials. (2020). Mechanistic-empirical pavement design guide: a manual of practice. AASHTO.
4. ARA. (2004). Guide for Mechanistic-Empirical Design Guide of New and Rehabilitated Pavement Structures.
5. Asphalt Institute. (1981). Thickness Design Manual (MS-1), 9th ed. In The Asphalt Institute, College Park, MD.
6. Beach, Adam Charles, Baladi, Gilbert Y, Dawson, Tyler A, Dean, Christopher M, Haider, PE, Waqar, Syed, & Chatti, Karim. (2012). Strategy Analysis Using Treatment Transition Matrices of In-service Pavement Condition Data. Paper presented at the Transportation Research Board 91st Annual Meeting.
7. Bonnaure, F., Gravois, A., & Udron, J. (1980). A New Method of Predicting the Fatigue Life of Bituminous Mixes. Journal of the Association of Asphalt Paving Technologists., Vol. 49, 499–529.
8. Delgadillo, R., Wahr, C., & Alarcón, J. P. (2011). Toward Implementation of the Mechanistic–Empirical Pavement Design Guide in Latin America: Preliminary Work in Chile. Transportation Research Record, 2226(1), 142-148.
9. El-Ashwah, A. S., El-Badawy, S. M., & Gabr, A. R. (2021). A Simplified Mechanistic-Empirical Flexible Pavement Design Method for Moderate to Hot Climate Regions. Sustainability, 13(19), 10760.
10. Francisco Dalla Rosa, Litao Liu, M.ASCE, Nasir G. Gharaibeh, M.ASCE.(2017). IRI Prediction Model for Use in Network-LevelPavement Management Systems.
11. Francken, L., & De Lurdes Antunes, M. (2000). Advanced Models for Analytical Design of European Pavement Structures. Progress report of COST333 and AMADEUS Projects.
12. Georgene M. Geary (2018). User Review of the AASHTO ‘Guide for the Local Calibration of the Mechanistic-Empirical Pavement Design Guide’
13. Huang, Y. H. (2003). Pavement analysis and design II ed.
14. Jain, S. S., Aggarwal, S., & Parida, M. (2005). HDM-4 pavement deterioration models for Indian national highway network. Journal of Transportation Engineering, 131(8), 623-631.
15. Kharoufeh, J. P., Solo, C. J., & Ulukus, M. Y. (2010). Semi-Markov models for degradation-based reliability. IIE Transactions, 42(8), 599-612.
16. Kim, SH., Kim, N. (2020). Development of performance prediction models in flexible pavement using regression analysis method.
17. Le, A. T., Lee, H. J., Park, H. M., & Kim, T. W. (2011). Development of Korean pavement design guide for asphalt pavements based on the mechanistic-empirical design principle. The Baltic Journal of Road and Bridge Engineering, 6(3), 169-176.
18. Li N, Xie W-C, Haas R (1996) Reliability-based processing of Markov chains for modeling pavement network deterioration. Transportation Research Record, 1524:203–213.
19. Newcomb, D. & Timm, D. (2001). Mechanistic Pavement Design. Transportation Research Record.
20. Powell, W. D., Potter, J. F., Mayhew, H. C., & Nunn, M. E. (1984). The structural design of bituminous roads (No. LR 1132 Monograph).
21. Primusz, P., & Tóth, C. (2018). Aszfaltburkolatú útpályaszerkezetek egyszerűsített analitikus méretezése. KÖZLEKEDÉSTUDOMÁNYI SZEMLE, 68(5), 17-33.
22. Pulugurta, H., Shao, Q., & Chou, Y. J. (2009). Pavement condition prediction using Markov process. Journal of Statistics and Management Systems, 12(5), 853-871.
23. Saleh MF, Mamlouk MS, Owusu-Antwi EB (2000) Mechanistic roughness model based on vehicle-pavement interaction. Transp Res Rec 1699:114–120.
24. The AASHO Road Test: Report 1, History and Description of the Project. Special Report 61A. (1961) Highway Research Board, National Academy of Sciences.Washington, D.C.
25. Yang, J., Gunaratne, M., Lu, J. J., & Dietrich, B. (2005). Use of recurrent Markov chains for modeling the crack performance of flexible pavements. Journal of Transportation Engineering, 131(11), 861-872.
26. Yao, L., Dong, Q., Jiang, J., & Ni, F. (2019). Establishment of prediction models of asphalt pavement performance based on a novel data calibration method and neural network. Transportation Research Record, 2673(1), 66-82.
27. Younos M.A., Abd El-Hakim R.T.,El-Badawy S.M. et al. (2020). Multi-input performance prediction models for flexible pavements using LTPP database.
中文文獻
1. 林勝傑. (2001).類神經網路與馬可夫鍊理論於鋪面狀況評估之應用. (碩士), 國立成功大學,台南市.
2. 周少凡. (1998). 模糊集理論於鋪面表面狀況評估之應用. (碩士), 國立成功大學, 台南市.
3. 侯洸廷.(2002).灰預測理論於鋪面評估之應用. 成功大學土木工程學系學位論文, 1-75.
4. 黃山琿. (1998). 馬可夫鍊理論在鋪面管理系統中之應用. (碩士), 國立成功大學, 台南市.
5. 蔣子平. (1999) 類神經及模糊集理論於評估柔性鋪面裂縫破壞之應用