Performance Evaluation of Machine Learning Algorithms for Seismic Retrofit Cost Estimation Using Structural Parameters

Document Type : Regular Article

Authors

1 Assistant Professor, Department of Civil Engineering, Bozorgmehr University of Qaenat, Qaen, Iran

2 Ph.D., Department of Civil Engineering, Hamedan Branch, Islamic Azad University, Hamedan, Iran

Abstract

Estimation of the seismic retrofit cost (SRC) is a complicated task in construction projects. In this study, the performance of four machine learning algorithms (MLAs), including Random Forest (RF), Extreme Learning Machine (ELM), Classification and Regression Tree (CART), and Multivariate Adaptive Regression Spline (MARS), was examined in estimating SRC values. The total floor area (TFA), number of stories (NS), seismic weight (SW), seismicity (S), soil type (ST), plan configuration (PC), and structural type (STT) were considered as structural input variables. To achieve the best performance of applied MLAs, twenty-two scenarios based on different combinations of input variables were considered. The correlation coefficient (r), Root Mean Squared Error (RMSE), Adjusted R-squared, and Nash-Sutcliffe efficiency (NSE) metrics together with the Taylor diagram were used to compare the accuracy of applied models. A sensitivity analysis using the RReliefF algorithm showed that TFA, SW, and PC are the most influential parameters, whereas the ST and STT have negative influences on SRC values. Comparison analysis results indicated that the ELM model with r of 0.896, RMSE of 0.081, and NSE of 0.758 had the best performance among other employed MLAs. Also, the RF regression achieved the second rank. In conclusion, the ELM model with single-layer feedforward neural network was superior to other data-driven models; therefore, it can be applied as an efficient tool for estimating SRC values using structural input parameters.

Graphical Abstract

Performance Evaluation of Machine Learning Algorithms for Seismic Retrofit Cost Estimation Using Structural Parameters

Highlights

  • Four machine learning algorithms was used for SRC prediction using structural parameters.
  • Sensitivity analysis of input parameters was performed using the RReliefF algorithm.
  • ELM had the best SRC prediction model compared with CART, MARS, and RF algorithms.
  • The TFA, SW, and PC are the most influential parameters affecting SRC values.
  • The soil type and structural type have a negative influence on SRC values.

Keywords

Main Subjects


[1]     Nasrazadani H, Mahsuli M, Talebiyan H, Kashani H. Probabilistic Modeling Framework for Prediction of Seismic Retrofit Cost of Buildings. J Constr Eng Manag 2017;143:04017055. doi:10.1061/(asce)co.1943-7862.0001354.
[2]     Chen WT, Huang Y. Approximately predicting the cost and duration of school reconstruction projects in Taiwan. Constr Manag Econ 2006;24:1231–9. doi:10.1080/01446190600953805.
[3]     Jafarzadeh R, Ingham JM, Wilkinson S. A seismic retrofit cost database for buildings with a framed structure. Earthq Spectra 2014;30:625–37. doi:10.1193/080713EQS226.
[4]     Jafarzadeh R, Ingham JM, Wilkinson S, González V, Aghakouchak AA. Application of Artificial Neural Network Methodology for Predicting Seismic Retrofit Construction Costs. J Constr Eng Manag 2014;140:04013044. doi:10.1061/(asce)co.1943-7862.0000725.
[5]     Jafarzadeh R, Wilkinson S, González V, Ingham JM, Amiri GG. Predicting Seismic Retrofit Construction Cost for Buildings with Framed Structures Using Multilinear Regression Analysis. J Constr Eng Manag 2014;140:04013062. doi:10.1061/(ASCE)CO.1943-7862.0000750.
[6]     Jafarzadeh R, Ingham JM, Walsh KQ, Hassani N, Ghodrati Amiri GR. Using Statistical Regression Analysis to Establish Construction Cost Models for Seismic Retrofit of Confined Masonry Buildings. J Constr Eng Manag 2015;141:04014098. doi:10.1061/(ASCE)CO.1943-7862.0000968.
[7]     Alizamir M, Kisi O, Zounemat-Kermani M. Modelling long-term groundwater fluctuations by extreme learning machine using hydro-climatic data. Hydrol Sci J 2018;63:63–73. doi:10.1080/02626667.2017.1410891.
[8]     Yaseen ZM, Deo RC, Hilal A, Abd AM, Bueno LC, Salcedo-Sanz S, et al. Predicting compressive strength of lightweight foamed concrete using extreme learning machine model. Adv Eng Softw 2018;115:112–25. doi:10.1016/j.advengsoft.2017.09.004.
[9]     Al-Shamiri AK, Kim JH, Yuan TF, Yoon YS. Modeling the compressive strength of high-strength concrete: An extreme learning approach. Constr Build Mater 2019;208:204–19. doi:10.1016/j.conbuildmat.2019.02.165.
[10]    Nayak, Sarat and Nayak, Sanjib and Panda S. Assessing Compressive Strength of Concrete with Extreme Learning Machine. J Soft Comput Civ Eng 2021;5:68–85. doi:10.22115/SCCE.2021.286525.1320.
[11]    Zhang W, Zhang R, Wu C, Goh ATC, Wang L. Assessment of basal heave stability for braced excavations in anisotropic clay using extreme gradient boosting and random forest regression. Undergr Sp 2020:1–9. doi:10.1016/j.undsp.2020.03.001.
[12]    Zhang W, Wu C, Li Y, Wang L, Samui P. Assessment of pile drivability using random forest regression and multivariate adaptive regression splines. Georisk 2021;15:27–40. doi:10.1080/17499518.2019.1674340.
[13]    Zhang R, Wu C, Goh ATC, Böhlke T, Zhang W. Estimation of diaphragm wall deflections for deep braced excavation in anisotropic clays using ensemble learning. Geosci Front 2021;12:365–73. doi:10.1016/j.gsf.2020.03.003.
[14]    Zhang Y, Wang Y, Zhou G, Jin J, Wang B, Wang X, et al. Multi-kernel extreme learning machine for EEG classification in brain-computer interfaces. Expert Syst Appl 2018;96:302–10. doi:10.1016/j.eswa.2017.12.015.
[15]    Yaseen ZM, Sulaiman SO, Deo RC, Chau K-W. An enhanced extreme learning machine model for river flow forecasting: State-of-the-art, practical applications in water resource engineering area and future research direction. J Hydrol 2019;569:387–408. doi:10.1016/j.jhydrol.2018.11.069.
[16]    Sattar AMA, Ertuğrul ÖF, Gharabaghi B, McBean EA, Cao J. Extreme learning machine model for water network management. Neural Comput Appl 2019;31:157–69. doi:10.1007/s00521-017-2987-7.
[17]    Wang X, Yang K, Kalivas JH. Comparison of extreme learning machine models for gasoline octane number forecasting by near-infrared spectra analysis. Optik (Stuttg) 2020;200:163325. doi:10.1016/j.ijleo.2019.163325.
[18]    Zhang W, Zhang R, Wu C, Goh ATC, Lacasse S, Liu Z, et al. State-of-the-art review of soft computing applications in underground excavations. Geosci Front 2020;11:1095–106. doi:10.1016/j.gsf.2019.12.003.
[19]    Zhang WG, Li HR, Wu CZ, Li YQ, Liu ZQ, Liu HL. Soft computing approach for prediction of surface settlement induced by earth pressure balance shield tunneling. Undergr Sp 2020. doi:10.1016/j.undsp.2019.12.003.
[20]    Siahkali MZ, Ghaderi A, Bahrpeyma A, Rashki M. Estimating Pier Scour Depth : Comparison of Empirical Formulations. J AI Data Min 2021;9:109–28. doi:10.22044/jadm.2020.10085.2147.
[21]    Singh R, Wagener T, Crane R, Mann ME, Ning L. A vulnerability driven approach to identify adverse climate and land use change combinations for critical hydrologic indicator thresholds: Application to a watershed in Pennsylvania, USA. Water Resour Res 2014;50:3409–27. doi:10.1002/2013WR014988.
[22]    Arifuzzaman M, Gazder U, Alam MS, Sirin O, Mamun A Al. Modelling of Asphalt’s Adhesive Behaviour Using Classification and Regression Tree (CART) Analysis. Comput Intell Neurosci 2019;2019. doi:10.1155/2019/3183050.
[23]    Choubin B, Moradi E, Golshan M, Adamowski J, Sajedi-Hosseini F, Mosavi A. An ensemble prediction of flood susceptibility using multivariate discriminant analysis, classification and regression trees, and support vector machines. Sci Total Environ 2019;651:2087–96. doi:10.1016/j.scitotenv.2018.10.064.
[24]    Salimi A, Faradonbeh RS, Monjezi M, Moormann C. TBM performance estimation using a classification and regression tree (CART) technique. Bull Eng Geol Environ 2018;77:429–40. doi:10.1007/s10064-016-0969-0.
[25]    Pandey S, Kumar V, Kumar P. Application and Analysis of Machine Learning Algorithms for Design of Concrete Mix with Plasticizer and without Plasticizer. J Soft Comput Civ Eng 2021;5:19–37. doi:10.22115/scce.2021.248779.1257.
[26]    Fung JF, Butry DT, Sattar S, McCabe SL. A methodology for estimating seismic retroft costs. Gaithersburg, MD: 2017. doi:10.6028/NIST.TN.1973.
[27]    Fung JF, Butry DT, Sattar S, McCabe SL. Estimating structural seismic retrofit costs for federal buildings. Gaithersburg, MD: 2018. doi:10.6028/NIST.TN.1996.
[28]    Fung JF, Sattar S, Butry DT, McCabe SL. A predictive modeling approach to estimating seismic retrofit costs. Earthq Spectra 2020;36:579–98. doi:10.1177/8755293019891716.
[29]    Li L-L, Sun J, Tseng M-L, Li Z-G. Extreme learning machine optimized by whale optimization algorithm using insulated gate bipolar transistor module aging degree evaluation. Expert Syst Appl 2019;127:58–67. doi:10.1016/j.eswa.2019.03.002.
[30]    Shariati M, Mafipour MS, Ghahremani B, Azarhomayun F, Ahmadi M, Trung NT, et al. A novel hybrid extreme learning machine–grey wolf optimizer (ELM-GWO) model to predict compressive strength of concrete with partial replacements for cement. Eng Comput 2020:1–23. doi:10.1007/s00366-020-01081-0.
[31]    Dai B, Gu C, Zhao E, Zhu K, Cao W, Qin X. Improved online sequential extreme learning machine for identifying crack behavior in concrete dam. Adv Struct Eng 2019;22:402–12. doi:10.1177/1369433218788635.
[32]    Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and Regression Trees. Routledge; 2017. doi:10.1201/9781315139470.
[33]    Friedman JH. Multivariate Adaptive Regression Splines. Ann Stat 2007;19:1–67. doi:10.1214/aos/1176347963.
[34]    Ghanizadeh AR, Rahrovan M. Modeling of unconfined compressive strength of soil-RAP blend stabilized with Portland cement using multivariate adaptive regression spline. Front Struct Civ Eng 2019;13:787–99. doi:10.1007/s11709-019-0516-8.
[35]    Shirzad A, Safari MJS. Pipe failure rate prediction in water distribution networks using multivariate adaptive regression splines and random forest techniques. Urban Water J 2019;16:653–61. doi:10.1080/1573062X.2020.1713384.
[36]    Breiman L. Random forests. Mach Learn 2001;45:5–32. doi:10.1023/A:1010933404324.
[37]    Benali L, Notton G, Fouilloy A, Voyant C, Dizene R. Solar radiation forecasting using artificial neural network and random forest methods: Application to normal beam, horizontal diffuse and global components. Renew Energy 2019;132:871–84. doi:10.1016/j.renene.2018.08.044.
[38]    Baez-Villanueva OM, Zambrano-Bigiarini M, Beck HE, McNamara I, Ribbe L, Nauditt A, et al. RF-MEP: A novel Random Forest method for merging gridded precipitation products and ground-based measurements. Remote Sens Environ 2020;239:111606. doi:10.1016/j.rse.2019.111606.
[39]    Tan K, Wang H, Chen L, Du Q, Du P, Pan C. Estimation of the spatial distribution of heavy metal in agricultural soils using airborne hyperspectral imaging and random forest. J Hazard Mater 2020;382:120987. doi:10.1016/j.jhazmat.2019.120987.
[40]    Samadianfard S, Ghorbani MA, Mohammadi B. Forecasting soil temperature at multiple-depth with a hybrid artificial neural network model coupled-hybrid firefly optimizer algorithm. Inf Process Agric 2018;5:465–76. doi:10.1016/j.inpa.2018.06.005.
[41]    Behbahani H, Amiri AM, Imaninasab R, Alizamir M. Forecasting accident frequency of an urban road network: A comparison of four artificial neural network techniques. J Forecast 2018;37:767–80. doi:10.1002/for.2542.
[42]    Robnik-Šikonja M, Kononenko I. Theoretical and Empirical Analysis of ReliefF and RReliefF. Mach Learn 2003;53:23–69.