橋梁在交通網路中是一項重要的基礎建設，因此維護作業是必要的，而實際上，橋梁維修的作業往往需要花費大量的資金，例如橋梁維修評估、損害檢測、校正作業等。此外，橋梁維修項目的資金往往是有限的，因此評估橋梁維護作業的優先順序以及最佳化維護資源的分配是不可避免的。
舉例來說，風險評估可以作為評估橋梁維修優先順序的基礎，且風險評估也應設為經常性的作業以確保橋梁的安全性。然而在進行橋梁評分值推論過程時，涉及個人主觀判斷，每個專家在評估同一橋梁風險等級時，常會根據本身專業能力或是經驗值來判斷，在語意上的表達以及感覺判斷等都存在著模糊的現象，稱為不確定性問題。
本研究提出以演化式模糊加權金字塔演算樹(GFWPOT)來解決評估橋梁維修風險評分值時的產生的不確定性問題。GFWPOT改良了演化式加權金字塔演算樹(GWPOT)，同時考慮基因演算法（Genetic Algorithm）與模糊理論(Fuzzy Theory)。為了驗證其準確度，分別用GFWPOT與GWPOT來測試相同案例並做比較，分析加入了模糊理論後的差異。結果顯示GFWPOT預測出的結果較後者好。

Bridges are essential infrastructures of the transportation network; thus, it is requisite for the public that bridges are maintained properly and function adequately. In practice, bridge maintenance often consumes a large amount of capital due to high expenses in the tasks of bridge evaluation, damage detection, and corrective activities. Additionally, funding availability for maintenance projects is often limited .Therefore, it is crucial to ensure that remedial activities for bridge structures are conducted in a timely manner.
In bridge maintenance, risk assessment can serve as the basis for prioritization and it should be conducted regularly for the purpose of safety. Nevertheless, the aggregation process used to derive the risk score involves a large number of subjective judgments of bridge experts. The reason is that the inference process of experts are influenced by different types of structural defects, decision criteria for setting maintenance priorities, maintenance practices, data availability, and even political decisions at a national level for resource allocation to maintenance projects. The above reasons is called uncertainty problem.
this study introduces the Genetic Fuzzy Weighted Pyramid Operation Tree (GFWPOT) to build a formula to solve the uncertainty problem. GFWPOT is a new improvement of the genetic operation tree that consists of Genetic Algorithm and Fuzzy Theory .Model accuracy was compared against GWPOT. Results demonstrate GFWPOT as an efficient approach that performs better than GWPOT.

1.Cheng, M. Y., & Wu, Y. W. (2009). Evolutionary support vector machine inference system for construction management. Automation in Construction, 18, 597-604.
2.彭建華，「以演化運算樹建構混凝土強度模型」，博士論文，中華大學，2009。
3.Yeh, I. C., & Lien, L. C. (2009). Knowledge discovery of concrete material using Genetic Operation Trees. Expert System with Applications, 36, 5807-5812
4.Pratama Mahardika Firdausi (2011). High Performance Concrete Compressive Strength Prediction Using Genetic Weighted Pyramid Operation Tree(GWPOT). Thesis, National Taiwan University of Science and Technology, Taiwan.
5.Tsai, H.-C. (2011). Weighted operation structures to program strengths of concrete-typed specimens using genetic algorithm. Expert Systems with Applications, 38, 161-168.
6.王欽輝，侯志陞，「FUZZY工學」，全華科技圖書公司，1992。
7.張志宏，「建築工程擋土開挖安全監測專家決策支援系統之建立」，碩士論文，國立臺灣科技大學營建工程系，1998。
8.Hsie, M., Ho, Y. F., Lin, C. T., & Yeh, I. C. (2012). Modeling asphalt pavement overlay transverse cracks using the genetic operation tree and Levenberg–Marquardt Method. Expert Systems with Applications, 39, 4874-4881.
9.Peng, C. H., Yeh, I. C., & Lien, L. C. (2010). Building strength models for high-performance concrete at different ages using genetic operation trees, nonlinear regression, and neural networks. Engineering with Computers, 26, 61-73.
10.Yeh, I. C., Lien, C. H., Peng, C. H., & Lien, L. C. (2010). Modeling Concrete Strength Using Genetic Operation Trees. International Conference on Machine Learning and Cybernetics. Qingdao: IEEE.
11.Tsai, H.-C. (2011). Weighted operation structures to program strengths of concrete-typed specimens using genetic algorithm. Expert Systems with Applications, 38, 161-168.
12.Peng, C. H., Yeh, I. C., & Lien, L. C. (2010). Building strength models for high-performance concrete at different ages using genetic operation trees, nonlinear regression, and neural networks. Engineering with Computers, 26, 61-73.
13.吳宗桂，「營造廠專業協力廠商評鑑模式之建立與應用」，碩士論文，國立臺灣科技大學營建工程系，2001。
14.Jagielska, I., Matthews, C., and Whitfort, T., “An investigation into the application of neural networks, fuzzy logic, genetic algorithms, and rough sets to automated knowledge acquisition for classification problems” , Neurocomputing, 24 37-54 (1999).
15.陳俊榮，「基因演算法在視覺系統參數校正及尺寸量測上之應用研究」，碩士論文，國立臺灣科技大學工程技術研究所，1999。
16.MITSUO GEN,RUNWEI CHENG,”Genetic Algorithms and Engineering Design”,A Wiley-Interscience Publication John Wiley&Sons,Inc.
17.陳昆揮(2006)，模糊理論應用於土石流危險評估系統之研究。
18.Cheng M. Y. and Ko C. H.,(2003).” evolutionary fuzzy neural inference system for construction management” Journal of Construction Engineering and Management,ASCE.
19.Chou, J. S., Chiu, C. K., Farfoura, M., & Al-Taharwa. (2011). Optimizing the Preiction Accuracy of Concrete Compressive Strength Based on a Comparison of Data-Mining Techniques. Journal of Computing in Civil Engineering, 25, 242-253.
20.鄧乃楊等，數據挖掘的新方法-支持向量機，科學出版社，中國，2004。
21.Fukahori, K. and Kubota, Y. Consistency evaluation of landscape design by a decision support system, Computer-Aided Civil and Infrastructure Engineering, 15(5), 342-354, 2000.
22.吳育偉、鄭明淵，支持向量機最佳化模式-應用於營建管理決策，第11屆營建工程與管理學術研討會，2007。