Earthquakes and Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Seismic vulnerability of reinforced concrete structures using machine learning

  • Ioannis Karampinis (Department of Civil Engineering, Democritus University of Thrace) ;
  • Lazaros Iliadis (Department of Civil Engineering, Democritus University of Thrace)
  • Received : 2024.03.08
  • Accepted : 2024.03.27
  • Published : 2024.08.25
⟨ Previous Next ⟩

Abstract

The prediction of seismic behavior of the existing building stock is one of the most impactful and complex problems faced by countries with frequent and intense seismic activities. Human lives can be threatened or lost, the economic life is disrupted and large amounts of monetary reparations can be potentially required. However, authorities at a regional or national level have limited resources at their disposal in order to allocate to preventative measures. Thus, in order to do so, it is essential for them to be able to rank a given population of structures according to their expected degree of damage in an earthquake. In this paper, the authors present a ranking approach, based on Machine Learning (ML) algorithms for pairwise comparisons, coupled with ad hoc ranking rules. The case study employed data from 404 reinforced concrete structures with various degrees of damage from the Athens 1999 earthquake. The two main components of our experiments pertain to the performance of the ML models and the success of the overall ranking process. The former was evaluated using the well-known respective metrics of Precision, Recall, F1-score, Accuracy and Area Under Curve (AUC). The performance of the overall ranking was evaluated using Kendall's tau distance and by viewing the problem as a classification into bins. The obtained results were promising, and were shown to outperform currently employed engineering practices. This demonstrated the capabilities and potential of these models in identifying the most vulnerable structures and, thus, mitigating the effects of earthquakes on society.

Keywords

References

  1. Alam, N., Alam, M.S. and Tesfamariam, S. (2012), "Buildings seismic vulnerability assessment methods: A comparative study", Nat. hazard., 62, 405-424. https://doi.org/10.1007/s11069-011-0082-4.
  2. Bansal, A. and Jain, A. (2021), "Analysis of focussed under-sampling techniques with machine learing classifiers", 2021 IEEE/ACIS 19th International Conference on Software Engineering Research, Management and Applications (SERA), Kanazawa, Japan, June.
  3. Barbat, A.H., Carreno, M.L., Pujades, L.G., Lantada, N., Cardona, O.D. and Marulanda, M.C. (2010), "Seismic vulnerability and risk evaluation methods for urban areas. A review with application to a pilot area", Struct. Infrastr. Eng., 6(1-2), 17-38. https://doi.org/10.1080/15732470802663763.
  4. Bergstra, J. and Bengio, Y. (2012), "Random search for hyper-parameter optimization", J. Mach. Learn. Res., 13(2), 281-305. https://doi.org/10.5555/2503308.2188395.
  5. Brena, M. and Batagelj, V. (2006), "The metric index", Croatica Chem. Acta, 79(3), 399-410.
  6. Buckland, M. and Gey, F. (1994), "The relationship between recall and precision", J. Am. Soc. Informat. Sci., 45(1), 12-19. https://doi.org/10.1002/(SICI)1097-4571(199401)45:1%3C12::AID-ASI2%3E3.0.CO;2-L.
  7. Cicirello, V.A. (2019), "Kendall tau sequence distance: Extending Kendall tau from ranks to sequences", arXiv preprint, 2019, arXiv:1905.02752. https://doi.org/10.48550/arXiv.1905.02752.
  8. Cunningham, P. and Delany, S.J. (2021), "K-nearest neighbour classifiers-a tutorial", ACM Comput. Survey. (CSUR), 54(6), 1-25. https://doi.org/10.1145/3459665.
  9. EAK 2000 (2000), Greek Code for Seismic Resistant Structures, https://iisee.kenken.go.jp/worldlist/23_Greece/23_Greece_Code.pdf
  10. Facchinei, F., Fischer, A. and Kanzow, C. (1998), "On the accurate identification of active constraints", SIAM J. Optimizat., 9(1), 14-32. https://doi.org/10.1137/S1052623496305882.
  11. Fawagreh, K., Gaber, M.M. and Elyan, E. (2014), "Random forests: From early developments to recent advancements", Syst. Sci. Control Eng.: Open Access J., 2(1), 602-609. https://doi.org/10.1080/21642583.2014.956265.
  12. Geurts, P., Ernst, D. and Wehenkel, L. (2006), "Extremely randomized trees", Mach. Learn., 63, 3-42. https://doi.org/10.1007/s10994-006-6226-1.
  13. Ghasemi, S.H., Bahrami, H. and Akbari, M. (2020), "Classification of seismic vulnerability based on machine learning techniques for RC frames", J. Soft Comput. Civil Eng., 4, 13-21. https://doi.org/10.22115/scce.2020.223322.1186.
  14. Gutierrez, P.A., Perez-Ortiz, M., Sanchez-Monedero, J., Fernandez-Navarro, F. and Hervas-Martinez, C. (2015), "Ordinal regression methods: survey and experimental study", IEEE Trans. Knowled. Data Eng., 28(1), 127-146. https://doi.org/10.1109/TKDE.2015.2457911.
  15. Harirchian, E. and Lahmer, T. (2020), "Improved rapid assessment of earthquake hazard safety of structures via artificial neural networks", IOP Conf. Ser.: Mater. Sci. Eng., 897(1), 012014. https://doi.org/10.1088/1757-899X/897/1/012014.
  16. Harirchian, E., Kumari, V., Jadhav, K., Raj Das, R., Rasulzade, S. and Lahmer, T. (2020), "A machine learning framework for assessing seismic hazard safety of reinforced concrete buildings", Appl. Sci., 10(20), 7153. https://doi.org/10.3390/app10207153.
  17. Harirchian, E., Kumari, V., Jadhav, K., Rasulzade, S., Lahmer, T. and Raj Das, R. (2021), "A synthesized study based on machine learning approaches for rapid classifying earthquake damage grades to RC buildings", Appl. Sci., 1(16), 7540. https://doi.org/10.3390/app11167540.
  18. Herbrich, R. (1999), "Support vector learning for ordinal regression", 9th International Conference on Artificial Neural Networks: ICANN '99, Edinburgh, UK, September.
  19. Hossain, R. and Timmer, D. (2021), "Machine learning model optimization with hyper parameter tuning approach", Glob. J. Comput. Sci. Technol. D Neural Artif. Intell., 21(2), 7-13.
  20. Impedovo, S. and Mangini, F.M. (2012), "A novel technique for handwritten digit classification using genetic clustering", 2012 International Conference on Frontiers in Handwriting Recognition, Bari, Italy, September.
  21. Karabinis, A. (2004), Calibration of Rapid Visual Screening in Reinforced Concrete Structures Based on Data after a Near Field Earthquake (7.9. 1999 Athens - Greece), School of Civil Engineers, Reinforced Concrete Laboratory, Zografos, Attica, Greece.
  22. Karampinis, I. and Iliadis, L. (2023), "A machine learning approach for seismic vulnerability ranking" International Conference on Engineering Applications of Neural Networks, Springer Nature, Cham Switzerland.
  23. Kosub, S. (2019), "A note on the triangle inequality for the Jaccard distance", Patt. Recogn. Lett., 120, 36-38. https://doi.org/10.48550/arXiv.1612.02696.
  24. Kotsiantis, S.B. (2013), "Decision trees: A recent overview", Artif. Intell. Rev., 39, 261-283. https://doi.org/10.1007/s10462-011-9272-4.
  25. Kotsiantis, S.B., Zaharakis, I. and Pintelas, P. (2007), "Supervised machine learning: A review of classification techniques", Emerg. Artif. Intell. Applicat. Comput. Eng., 160(1), 3-24.
  26. Kumari, R. and Srivastava, S.K. (2017), "Machine learning: A review on binary classification", Int. J. Comput. Applicat., 160(7), 11-15. https://doi.org/10.5120/IJCA2017913083.
  27. Lang, K. and Bachmann, H. (2003), "On the seismic vulnerability of existing unreinforced masonry buildings", J. Earthq. Eng., 7(3), 407-426. https://doi.org/10.1080/13632460309350456
  28. Li, L. and Lin, H.T. (2006), "Ordinal regression by extended binary classification", Advances in Neural Information Processing Systems 19: Proceedings of the 2006 Conference, MIT Press, Cambridge, MA, USA.
  29. Liu, Y., Li, X., Kong, A.W.K. and Goh, C.K. (2016), "Learning from small data: A pairwise approach for ordinal regression", 2016 IEEE Symposium Series on Computational Intelligence (SSCI), Athens, Greece, December.
  30. Lizundia, B., Durphy, S., Griffin, M., Holmes, W., Hortacsu, A., Kehoe, B., ... and Welliver, B. (2015), "Update of FEMA P-154: rapid visual screening for potential seismic hazards", Improving the Seismic Performance of Existing Buildings and Other Structures 2015, San Francisco, CA, USA, December.
  31. Luo, H. and Paal, S.G. (2019), "A locally weighted machine learning model for generalized prediction of drift capacity in seismic vulnerability assessments", Comput. Aid. Civil Infrastr. Eng., 34(11), 935-950. https://doi.org/10.1111/mice.12456.
  32. Mantovani, R.G., Rossi, A.L., Vanschoren, J., Bischl, B. and De Carvalho, A.C. (2015), "Effectiveness of random search in SVM hyper-parameter tuning", 2015 International Joint Conference on Neural Networks (IJCNN), Killarney, Ireland, July.
  33. Marom, N.D., Rokach, L. and Shmilovici, A. (2010), "Using the confusion matrix for improving ensemble classifiers", 2010 IEEE 26-th Convention of Electrical and Electronics Engineers in Israel, Eilat, Israel, November.
  34. Mohammed, R., Rawashdeh, J. and Abdullah, M. (2020), "Machine learning with oversampling and undersampling techniques: Overview study and experimental results", 2020 11th International Conference on Information and Communication Systems (ICICS), Irbid, Jordan, April.
  35. Nanda, R.P. and Majhi, D.R. (2013), "Review on rapid seismic vulnerability assessment for bulk of buildings", J. Inst. Eng. (India): Ser. A, 94, 187-197. https://doi.org/10.1007/s40030-013-0048-5.
  36. Natekin, A. and Knoll, A. (2013), "Gradient boosting machines, a tutorial", Front. Neurorobot., 7, 21. https://doi.org/10.3389/fnbot.2013.00021.
  37. Newaz, A., Hassan, S. and Haq, F.S. (2022), "An empirical analysis of the efficacy of different sampling techniques for imbalanced classification", arXiv preprint, 2022, arXiv: 2208.11852.
  38. Ningthoujam, M.C. and Nanda, R.P. (2018), "Rapid visual screening procedure of existing building based on statistical analysis", Int. J. Disaster Risk Reduct., 28, 720-730. https://doi.org/10.1016/j.ijdrr.2018.01.033
  39. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., ... and Duchesnay, E. (2011), "Scikit-learn: Machine learning in Python", J. Mach. Learn. Res., 12, 2825-2830. https://doi.org/10.48550/arXiv.1201.0490.
  40. Peter Flach and Meelis Kull (2015), "Precision-recall-gain curves: PR analysis done right", Adv. Neural Informat. Pr. Syst., 28, 838-846. https://doi.org/10.5555/2969239.2969333.
  41. Price Code (2005), Eurocode 8: Design of Structures for Earthquake Resistance-Part 1: General Rules, Seismic Actions and Rules for Buildings, European Committee for Standardization, Brussels, Belgium.
  42. Roeslin, S., Ma, Q., Juarez-Garcia, H., Gomez-Bernal, A., Wicker, J. and Wotherspoon, L. (2020), "A machine learning damage prediction model for the 2017 Puebla-Morelos, Mexico, earthquake", Earthq. Spectra, 36(2), 314-339. https://doi.org/10.1177/8755293020936714.
  43. Rossetto, T. and Elnashai, A. (2002), "Derivation of vulnerability functions for RC buildings based on observation Data", European Commission, Brussels, Belgium.
  44. Rosti, A., Rota, M. and Penna, A. (2022), "An empirical seismic vulnerability model", Bull. Earthq. Eng., 20(8), 4147-4173. https://doi.org/10.1007/s10518-022-01374-3.
  45. Ruggieri, S., Cardellicchio, A., Leggieri, V. amd Uva, G. (2021), "Machinelearning based vulnerability analysis of existing buildings", Automat. Constr., 132, 103936. https://doi.org/10.1016/j.autcon.2021.103936.
  46. Sajan, K.C., Bhusal, A., Gautam, D. and Rupakhety, R. (2023), "Earthquake damage and rehabilitation intervention prediction using machine learning", Eng. Fail. Anal., 144, 106949. https://doi.org/10.1016/j.engfailanal.2022.106949.
  47. Singh, A., Prakash, B.S. and Chandrasekaran, K. (2016), "A comparison of linear discriminant analysis and ridge classifier on Twitter data", 2016 International Conference on Computing, Communication and Automation (ICCCA), Greater Noida, India, April.
  48. Snoek, J., Larochelle, H. and Adams, R.P. (2012), "Practical bayesian optimization of machine learning algorithms", Adv. Neural Informat. Pr. Syst., 25, 1.
  49. So, Y. (1995), A Tutorial on Logistic Regression, SAS White Papers, Cary, NC, USA.
  50. State of California (1999), A Tutorial on Logistic Regression, SAS White Papers, Cary, NC, USA.
  51. Tesfamariam, S. and Saatcioglu, M. (2008), "Risk-based seismic evaluation of reinforced concrete buildings", Earthq. Spectra, 24(3), 795-821. https://doi.org/10.1193/1.2952767.
  52. Vicente, R., Parodi, S., Lagomarsino, S., Varum, H. and Silva, J.M. (2011), "Seismic vulnerability and risk assessment: Case study of the historic city centre of Coimbra, Portugal", Bull. Earthq. Eng., 9, 1067-1096. http://doi.org/10.1007/s10518-010-9233-3.
  53. Visa, S., Ramsay, B., Ralescu, A.L. and Van Der Knaap, E. (2011), "Confusion matrixbased feature selection", MAICS, 710(1), 120-127.
  54. Yang, L. and Shami, A. (2020), "On hyperparameter optimization of machine learning algorithms: Theory and practice", Neurocomput., 415, 295-316. https://doi.org/10.1016/j.neucom.2020.07.061.
  55. Yen, S. and Lee, Y. (2006), "Under-sampling approaches for improving prediction of the minority class in an imbalanced dataset", Intelligent Control and Automation: International Conference on Intelligent Computing, ICIC 2006, Kunming, China, August.
  56. Yu, T. and Zhu, H. (2020), "Hyper-parameter optimization: A review of algorithms and applications", arXiv preprint, 2020, arXiv:2003.05689. https://doi.org/10.48550/arXiv.2003.05689.
  57. Zhang, B. and Srihari, S.N (2003), "Properties of binary vector dissimilarity measures", Proceedings of JCIS International Conference on Computer Vision, Pattern Recognition, and Image Processing.