加勁纖維材料具抗拉、耐久及質量輕等特性，應用於大地工程中，可加強土壤強度、提升加勁結構物(如加勁邊坡)的整體穩定性，亦有環境美化及結合生態之優點。過去在三軸與直剪試驗，已進行許多纖維加勁土壤(Fiber Reinforced Soil, FRS)的力學實驗研究，並提出加勁土壤剪力強度的理論或經驗預測公式。然而，因加勁纖維混合在土壤中可視為隨機分佈，具有不確定性，其剪力強度參數往往無法以理論或經驗公式準確推估。有鑒於此，本研究回顧1983-2015年文獻，擷取纖維加勁土壤屬性，建立一特徵資料庫，包含土壤參數(試體尺寸、土壤摩擦角及凝聚力)、纖維參數(纖維長徑比、體積含量、纖維與土壤界面摩擦角度等)及應力參數(圍壓或正應力)等。伺資料庫建置完成後，應用「資料探勘技術」，包含(1)分類與迴歸法：線性迴歸模型(LR)、分類迴歸樹(C&R Tree)、廣義線性模型(GENLIN)及卡方自動交叉檢驗(CHAID)；(2)機器學習法：人工類神經網路(ANN)、支援向量機(SVM)、迴歸機器學習(SVR)；(3)啟發式演化組合模型法(Meta Ensemble Models)：表決法(Voting)、重複採樣平均表決法(Bagging)、演化堆疊法(Stacking)、分類-迴歸層級法(Tiering)，建構纖維加勁土壤內摩擦角預測模式。分析結果發現主要影響纖維加勁土壤剪力強度的預測因子有(1)纖維含量、(2)纖維長徑比、(3)土壤摩擦角以及(4)圍壓與正應力。經由後續模型訓練與交叉驗證、測試，顯示啟發式模型組合法的分類-迴歸層級法Tiering SVM-(SVR/SVR)模型之預測值與文獻內記述實際值間的相關係數(correlation coefficient, R)高達0.9(1為完全相關)、平均絕對值誤差率(mean absolute percentage error, MAPE)小於4%、方根誤差(root mean square error, RMSE)小於2度、絕對誤差(mean absolute error, MAE)小於2度，各評估指標之改善效能優於傳統公式達9.31-79.50%，本研究之貢獻為提出一高效能預測FRS剪力強度參數之進階人工智慧模型。

Fiber-reinforced materials exhibit high tensile strength, durability, and a light mass. Such materials, when used in geotechnical engineering, can improve soil strength and the overall stability of reinforced structures such as side slopes. In addition, fiber-reinforced materials can be employed for landscaping and facilitate integral ecology. Scholars have studied the mechanics of fiber-reinforced soil (FRS) by using triaxial and direct shear tests, proposing theories and empirical models for predicting the shear strength of reinforced soils. However, mixing fibers into soil can be regarded as a random distribution, demonstrating uncertainty. The shear strength parameters cannot be accurately predicted using theories or empirical prediction equations. Therefore, this study established an FRS characteristic database comprising relevant literature from 1983 to 2015. The parameters collected and analyzed in this study included soil parameters (i.e., size of test piece, friction angle of the soil, and soil cohesion), fiber parameters (i.e., aspect ratio of fiber, volume content, and interface friction between fiber and soil), and stress parameters (i.e., confining pressure and normal stress). After datafication, data mining technologies were employed to identify factors influencing shear strength and to predict the friction angle of FRS. The analysis techniques included (1) classification and regression methods, such as linear regression analysis, classification and regression tree analysis, a generalized linear model, and chi-squared automatic interaction detection; (2) machine learners, such as an artificial neural network and support vector machine/regression; and (3) meta ensemble models, such as Voting, Bagging, Stacking and Tiering. The results indicated that fiber content, fiber aspect ratio, friction angle of soil and stress parameter were the primary factors influencing the shear strength of FRS. Through subsequent model training,cross-validation and testing, the optimal model obtained was the Tiering SVM-(SVR/SVR) model. The correlation coefficient of the prediction values with the actual values recorded in the literature was 0.9, indicating a high correlation. The mean absolute percentage error was < 4%, root mean square error was < 2°, and mean absolute error was < 2°. The overall improvement in performance measures compared with that demonstrated using conventional theory or empirical equation was 9.31%-79.50%. This study contributed to the domain knowledge by proposing an effective artificial intelligence model for predicting the friction angle of FRS.

[1]F. Ahmad, F. Bateni, M. Azmi, Performance evaluation of silty sand reinforced with fibres, Geotextiles and Geomembranes 28 (1) (2010) 93-99.
[2]A.-S. Al-Refeai, Dynamic and static characterization of polypropylene fiber-reinforced dune sand, Geosynthetics International 5 (5) (1998) 443-458.
[3]T.O. Al-Refeai, Behavior of granular soils reinforced with discrete randomly oriented inclusions, Geotextiles and Geomembranes 10 (4) (1991) 319-333.
[4]I. Alkroosh, H. Nikraz, Correlation of pile axial capacity and CPT data using gene expression programming, Geotechnical and Geological Engineering 29 (5) (2011) 725-748.
[5]I. Alkroosh, H. Nikraz, Simulating pile load-settlement behavior from CPT data using intelligent computing, Open Engineering 1 (3) (2011) 295-305.
[6]V. Anggraini, A. Asadi, B.B.K. Huat, H. Nahazanan, Effects of coir fibers on tensile and compressive strength of lime treated soft soil, Measurement 59 (0) (2015) 372-381.
[7] P.C. Austin, J.V. Tu, J.E. Ho, D. Levy, D.S. Lee, Using methods from the data-mining and machine-learning literature for disease classification and prediction: a case study examining classification of heart failure subtypes, Journal of Clinical Epidemiology 66 (4) (2013) 398-407.
[8]A.G. Benardos, D.C. Kaliampakos, Modelling TBM performance with artificial neural networks, Tunnelling and Underground Space Technology 19 (6) (2004) 597-605.
[9]L. Breiman, Bagging Predictors, Machine Learning 24 (2) (1996) 123-140.
[10]M.S. Chauhan, S. Mittal, B. Mohanty, Performance evaluation of silty sand subgrade reinforced with fly ash and fibre, Geotextiles and Geomembranes 26 (5) (2008) 429-435.
[11]C.-W. Chen, A constitutive model for fiber-reinforced soils, University of Missouri--Columbia, 2007.
[12]R.-H. Chen, P.-C. Chi, T.-C. Wu, C.-C. Ho, Shear Strength of Continuous-Filament Reinforced Sand, Journal of GeoEngineering 6 (2) (2011) 99-107.
[13]S.-C. Cheng, Soil Fiber Reinforcement, University of Minnesota Rochester 2014.
[14]J.-S. Chou, C.-F. Tsai, Concrete compressive strength analysis using a combined classification and regression technique, Automation in Construction 24 (0) (2012) 52-60.
[15]J.-S. Chou, K.-H. Yang, J.P. Pampang, A.-D. Pham, Evolutionary metaheuristic intelligence to simulate tensile loads in reinforcement for geosynthetic-reinforced soil structures, Computers and Geotechnics 66 (0) (2015) 1-15.
[16]N. Consoli, M. Casagrande, M. Coop, Effect of fiber  reinforcement on the isotropic compression behavior of a sand, Journal of Geotechnical and Geoenvironmental Engineering 131 (11) (2005) 1434-1436.
[17]N. Consoli, M. Casagrande, M. Coop, Performance of a fibre-reinforced sand at large shear strains, Géotechnique 57 (9) (2007) 751-756.
[18]N. Consoli, L. Festugato, K. Heineck, Strain-hardening behaviour of fibre-reinforced sand in view of filament geometry, Geosynthetics International 16 (2) (2009) 109-115.
[19]N. Consoli, J. Montardo, M. Donato, P. Prietto, Effect of material properties on the behaviour of sand—cement—fibre composites, Proceedings of the ICE-Ground Improvement 8 (2) (2004) 77-90.
[20]N. Consoli, P. Prietto, L. Ulbrich, Influence of fiber and cement addition on behavior of sandy soil, Journal of Geotechnical and Geoenvironmental Engineering 124 (12) (1998) 1211-1214.
[21]N.C. Consoli, M.A. Vendruscolo, A. Fonini, F.D. Rosa, Fiber reinforcement effects on sand considering a wide cementation range, Geotextiles and Geomembranes 27 (3) (2009) 196-203.
[22]S.K. Das, Artificial neural networks in geotechnical engineering: modeling and application issues, Metaheuristics in Water, Geotechnical and Transport Engineering (2012) 231.
[23]D. Freitag, Soil randomly reinforced with fibers, Journal of Geotechnical Engineering 112 (8) (1986) 823-826.
[24]A. Goh, Seismic Liquefaction Potential Assessed by Neural Networks, Journal of Geotechnical Engineering 120 (9) (1994) 1467-1480.
[25]A.T. Goh, F.H. Kulhawy, C. Chua, Bayesian Neural Network Analysis of Undrained Side Resistance of Drilled Shafts, Journal of Geotechnical and Geoenvironmental Engineering 131 (1) (2005) 84-93.
[26]D. Gray, T. Al‐Refeai, Behavior of fabric‐versus fiber‐reinforced sand, Journal of Geotechnical Engineering 112 (8) (1986) 804-820.
[27]D. Gray, H. Ohashi, Mechanics of fiber reinforcement in sand, Journal of Geotechnical Engineering 109 (3) (1983) 335-353.
[28]G.H. Gregory, Shear strength, creep and stability of fiber-reinforced soil slopes, Oklahoma State University, 2006.
[29]S.M. Hejazi, M. Sheikhzadeh, S.M. Abtahi, A. Zadhoush, A simple review of soil reinforcement by using natural and synthetic fibers, Construction and Building Materials 30 (0) (2012) 100-116.
[30]IBM, Clementine 12.0 algorithms guide, Chicago, Integral Solutions Limited, 2009.
[31]E. Ibraim, S. Fourmont, Behaviour of sand reinforced with fibres, Soil Stress-Strain Behavior: Measurement, Modeling and Analysis 146 (2007) 807-818.
[32]D. Jahed Armaghani, M. Hajihassani, B. Yazdani Bejarbaneh, A. Marto, E. Tonnizam Mohamad, Indirect measure of shale shear strength parameters by means of rock index tests through an optimized artificial neural network, Measurement 55 (0) (2014) 487-498.
[33]S. Kaniraj, V. Havanagi, Behavior of cement-stabilized fiber-reinforced fly ash-soil mixtures, Journal of Geotechnical and Geoenvironmental Engineering 127 (7) (2001) 574-584.
[34]G.V. Kass, An exploratory technique for investigating large quantities of categorical data, Applied statistics 29 (2) (1980) 119-127.
[35]R. Kohavi, A study of cross-validation and bootstrap for accuracy estimation and model selection, Stanford University 1995.
[36]S. Kumar, E. Tabor, Strength characteristics of silty clay reinforced with randomly oriented nylon fibers, Electronic Journal of Geotechnical Engineering 8 (2) (2003) 10.
[37]C. Li, J. Zornberg, Mobilization of reinforcement forces in fiber-reinforced soil, Journal of Geotechnical and Geoenvironmental Engineering 139 (1) (2013) 107-115.
[38]J. Li, C. Tang, D. Wang, X. Pei, B. Shi, Effect of discrete fibre reinforcement on soil tensile strength, Journal of Rock Mechanics and Geotechnical Engineering 6 (2) (2014) 133-137.
[39]J. Loehr, R. Romero, E. Ang, Development of a strain-based model to predict strength of geosynthetic fiber-reinforced soil, Geosynthetics Research and Development in Progress, 2005, pp. 1-7.
[40]M. Maher, D. Gray, Static response of sands reinforced with randomly distributed fibers, Journal of Geotechnical Engineering 116 (11) (1990) 1661-1677.
[41]R. Michalowski, J. Čermák, Triaxial compression of sand reinforced with fibers, Journal of Geotechnical and Geoenvironmental Engineering 129 (2) (2003) 125-136.
[42]R. Michalowski, A. Zhao, Failure of fiber-reinforced granular soils, Journal of Geotechnical Engineering 122 (3) (1996) 226-234.
[43]S. Mortazavian, A. Fatemi, Fatigue behavior and modeling of short fiber reinforced polymer composites: A literature review, International Journal of Fatigue 70 (0) (2015) 297-321.
[44]S. Najjar, S. Sadek, A. Alcovero, Quantification of model uncertainty in shear strength predictions for fiber-reinforced sand, Journal of Geotechnical and Geoenvironmental Engineering 139 (1) (2013) 116-133.
[45]M.S. Nataraj, K.L. McManis, Strength and deformation properties of soils reinforced with fibrillated fibers., 4 (1) (1997) 65-79.
[46]J. Nelder, R. Wedderburn, Generalized linear models, Journal of the Royal Statistical Society. Series A (General) 135 (3) (1972) 370-384.
[47]H.I. Park, Y.T. Kim, Prediction of strength of reinforced lightweight soil using an artificial neural network, Engineering Computations (Swansea, Wales) 28 (5) (2011) 600-615.
[48]A.L. Prodromidis, S.J. Stolfo, A comparative evaluation of meta-learning strategies over large and distributed data sets, Workshop on Meta-learning, Sixteenth Intl. Conf. Machine Learning, Citeseer, 1999, pp. 18-27.
[49]C.S. Rani, N.V.K. Togati, Artificial neural networks (ANNs) for prediction of engineering properties of soils, Int. J. Innov. Technol. Explor. Eng.(IJITEE) 3 (1) (2013) 123-130.
[50]G. Ranjan, R. Vasan, H. Charan, Probabilistic analysis of randomly distributed fiber-reinforced soil, Journal of Geotechnical Engineering 122 (6) (1996) 419-426.
[51]G. Ranjan, R.M. Vasan, H.D. Charan, Behaviour of plastic-fibre-reinforced sand, Geotextiles and Geomembranes 13 (8) (1994) 555-565.
[52]S. Sadek, S. Najjar, F. Freiha, Shear strength of fiber-reinforced sands, Journal of Geotechnical and Geoenvironmental Engineering 136 (3) (2010) 490-499.
[53]P. Samui, 14 - Slope Stability Analysis Using Multivariate Adaptive Regression Spline, in: Alavi (Ed.), Metaheuristics in Water, Geotechnical and Transport Engineering, Elsevier, Oxford, 2013, pp. 327-342.
[54]R. Santoni, J. Tingle, S. Webster, Engineering properties of sand-fiber mixtures for road construction, Journal of Geotechnical and Geoenvironmental Engineering 127 (3) (2001) 258-268.
[55]W. Shao, B. Cetin, Y. Li, J. Li, L. Li, Experimental investigation of mechanical properties of sands reinforced with discrete randomly distributed fiber, Geotechnical and Geological Engineering 32 (4) (2014) 901-910.
[56]G.L. Sivakumar Babu, A.K. Vasudevan, S. Haldar, Numerical simulation of fiber-reinforced sand behavior, Geotextiles and Geomembranes 26 (2) (2008) 181-188.
[57]A.O. Sykes, An introduction to regression analysis, Chicago, 1993.
[58]R. Timofeev, Classification and regression trees (cart) theory and applications, Humboldt University, Berlin, 2004.
[59]J. Tinoco, A. Gomes Correia, P. Cortez, Support vector machines applied to uniaxial compressive strength prediction of jet grouting columns, Computers and Geotechnics 55 (0) (2014) 132-140.
[60]Y. Tsompanakis, N.D. Lagaros, P.N. Psarropoulos, E.C. Georgopoulos, Simulating the seismic response of embankments via artificial neural networks, Advances in Engineering Software 40 (8) (2009) 640-651.
[61]V.N. Vapnik, The nature of statistical learning theory, New York, Springer-Verlag, 1995.
[62]D.H. Wolpert, Stacked generalization, Neural networks 5 (2) (1992) 241-259.
[63]T. Yetimoglu, O. Salbas, A study on shear strength of sands reinforced with randomly distributed discrete fibers, Geotextiles and Geomembranes 21 (2) (2003) 103-110.
[64]Y. Yu, B. Zhang, H. Yuan, An intelligent displacement back-analysis method for earth-rockfill dams, Computers and Geotechnics 34 (6) (2007) 423-434.
[65]J. Zornberg, Discrete framework for limit equilibrium analysis of fibre-reinforced soil, Géotechnique 52 (8) (2002) 593-604.