Recommendations for Active-Learning Kriging Reliability Analysis of Bridge Structures
Publication: Journal of Bridge Engineering
Volume 30, Issue 1
Abstract
Active-learning Kriging (AK) was developed as a surrogate-aided reliability technique to address the need for efficient reliability estimation when assessing complex limit states. The results of AK analyses are sensitive to the choice of the regression function, correlation function, learning function and associated stopping criteria, and reliability estimation technique, with unique sets of these input parameters referred to as AK configurations. For the reliable use of AK analysis in bridge reliability assessment, recommendations regarding the best-performing AK configurations are needed to balance the desired accuracy-to-efficiency of the simulation. The objective of this study was to recommend sets of AK configurations for the reliability analysis of reinforced-concrete bridge girders and piers that can be readily used by engineers to perform AK analysis for bridge design optimization and assessment. An extensive parametric analysis, using 432 unique AK configurations and over 3,000 AK analyses, was performed, combined with the application of a comprehensive metric system to recommend the top five best-performing AK configurations for bridge analysis based on the root mean square error, the absolute average error, the degree of consistency, and total number of training points.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
Some or all data, models, or code generated or used during the study are proprietary or confidential in nature and may only be provided with restrictions. The AK analysis computer script is confidential.
Acknowledgments
The authors wish to acknowledge the financial contributions of Dalhousie University, the Natural Sciences and Engineering Research Council (NSERC), and the Nova Scotia Department of Public Works.
Notation
The following symbols are used in this paper:
- AIdeal
- area of idealized cross-sectional geometry;
- Ag
- gross cross-sectional area;
- area of steel in the ith layer of rebar;
- Ast
- total area of steel in the cross-section;
- B
- width of the section;
- Cr
- compressive force;
- db
- diameter of steel bar reinforcements;
- di
- depth to ith layer of rebar;
- EFF(x)
- effective feasibility learning function;
- regression function;
- F
- regression realizations;
- fy
- yield strength of steel;
- compressive strength of concrete;
- f(x)
- regression evaluation of x;
- G(x)
- responses generated by original model;
- responses generated by Kriging surrogate model;
- mean value of Kriging prediction;
- H
- height of the cross-section;
- Hc(x)
- corrected H learning function;
- H(x)
- H learning function;
- I1
- transformed RMSE;
- I2
- transformed AAE;
- I3
- transformed TTPs;
- I4
- transformed DoC;
- KO(x)
- KO learning function;
- L
- correlation length vector;
- Leffect
- live load effect;
- optimum correlation length vector;
- dead load moment due to self-weight;
- dead load moment due to wearing surface;
- ML
- live load moment;
- Mc
- compressive moment from concrete zone;
- Min
- vector of initial inputs;
- Mout
- vector of initial outputs;
- Mr
- moment resistance;
- tensile moment generated by the ith layer of reinforcing bars;
- m
- number of design sites within design of experiments;
- N
- minimum number of trials;
- NAK-MCS
- number of trials in the AK-MCS analysis;
- Nadded
- number of added training points;
- Nbl
- number of reinforcing bar layers within the cross-section;
- Ncalls
- summation number of initial and added training points;
- Ncs
- number of concrete sections with different base widths;
- Nf
- total number of failed trials;
- NI
- total number of identical analyses;
- Ninitial
- number of initial training set;
- NMCS
- total number of trials in a Monte Carlo simulation;
- Ns
- number of successful reliability analyses;
- n
- number of random variables;
- P
- number of initial design points required to generate Kriging surrogate model;
- PF
- professional factor;
- Pc
- axial force in concrete under compression;
- Pf
- probability of failure;
- probability of failure of AK-MCS;
- Probability of failure of the ith identical AK configuration;
- probability of failure of Monte Carlo simulation;
- Pr
- axial load resistance;
- maximum allowable axial load;
- Pro
- axial load resistance with an eccentricity value of zero;
- axial force in the ith steel rebar under tension;
- Py
- axial force from interaction diagram;
- expected probability of failure;
- q
- number of model outputs;
- expected probability of survival;
- R
- correlation matrix;
- REIF(x)
- reliability based expected improvement learning function;
- R(x)
- resistance model;
- r(x)
- correlation evaluation for x;
- Si
- vector containing the inputs from all design sites;
- Ti
- tensile force in the ith layer of rebar;
- U(x)
- U learning function;
- error measurements that determine the border of the desired range;
- V
- coefficient of variation;
- v1
- linear scale;
- v2
- exponential scale;
- v3
- logarithmic scale;
- X
- design site;
- x
- vector made up of randomly generated variables;
- Yi
- vector containing the outputs from all design sites;
- stochastic process;
- Z value for selected confidence level;
- β
- reliability index;
- βMCS
- reliability index for the MCS for a given geometry;
- βj
- matrix of regression coefficients;
- Kriging shape predictor factor;
- Kriging shape prediction factor;
- λ
- bias;
- μ
- mean value;
- mean value of the Kriging predictor;
- ρ
- reinforcement ratio;
- standard deviation of the Kriging predictor; and
- φ(x)
- variance of the Kriging predictor.
References
Bichon, B. J., M. S. Eldred, L. P. Swiler, S. Mahadevan, and J. M. McFarland. 2008. “Efficient global reliability analysis for nonlinear implicit preformance functions.” AIAA J. 135 (1): 2459–2468. https://doi.org/10.2514/1.34321.
Biondini, F., and D. M. Frangopol. 2017. “Time-variant redundancy and failure times of deteriorating concrete structures considering multiple limit states.” Struct. Infrastruct. Eng. 13 (1): 94–106. https://doi.org/10.1080/15732479.2016.1198403.
Cressie, N. 1990. “The origins of Kriging.” Math. Geol. 22: 239–251. https://doi.org/10.1007/BF00889887.
CSA (Canadian Standard Association). 2014. Concrete design handbook. CSA A23.3. Toronto, ON, Canada: Cement Association of Canada.
CSA (Canadian Standard Association). 2019a. Canadian highway bridge design code. CSA S6. Toronto, ON, Canada: CSA.
CSA (Canadian Standard Association). 2019b. Commentary on the Canadian highway bridge design code. CSA S6.1. Toronto, ON, Canada: CSA.
Echard, B., N. Gayton, and M. Lemaire. 2011. “AK-MCS: An active learning reliability method combining Kriging and Monte Carlo simulation.” Struct. Saf. 33 (2): 145–154. https://doi.org/10.1016/j.strusafe.2011.01.002.
Ghosn, M., and F. Moses. 1986. “Reliability calibration of bridge design code.” J. Struct. Eng. 112 (4): 745–763. https://doi.org/10.1061/(ASCE)0733-9445(1986)112:4(745).
Huang, C. W., A. El Hami, and B. Radi. 2016a. “Overview of structural reliability analysis methods—Part I: Local reliability methods.” Incertitudes et fiabilité des systèmes multiphysiques 17 (1): 1–10.
Huang, C. W., A. El Hami, and B. Radi. 2016b. “Overview of structural reliability analysis methods—Part II: Sampling methods.” Incertitudes et fiabilité des systèmes multiphysiques 17 (1): 1–10.
Huang, C. W., A. El Hami, and B. Radi. 2016c. “Overview of structural reliability analysis methods—Part III: Global reliability methods.” Incertitudes et fiabilité des systèmes multiphysiques 17 (1): 1–8
Immonen, A., and E. Niemelä. 2008. “Survey of reliability and availability prediction methods from the viewpoint of software architecture.” Software Syst. Model. 7 (1): 49–65. https://doi.org/10.1007/s10270-006-0040-x.
Journel, A. G. 1977. “Kriging in terms of projections.” Math. Geol. 9 (6): 563–586. https://doi.org/10.1007/BF02067214.
Kaymaz, I. 2005. “Application of Kriging method to structural reliability problems.” Struct. Saf. 27 (2): 133–151. https://doi.org/10.1016/j.strusafe.2004.09.001.
Kennedy, D. J. L., D. P. Gagnon, D. E. Allen, and J. G. MacGregor. 1992. “Canadian highway bridge evaluation: Load and resistance factors.” Can. J. Civ. Eng. 19 (6): 992–1006. https://doi.org/10.1139/l92-119.
Khorramian, K., A. Alhashmi, and F. Oudah. 2023a. “Optimized active learning Kriging reliability-based assessment of laterally loaded pile groups modeled using random finite element analysis.” Comuput. Geotech. 154: 105135. https://doi.org/10.1016/j.compgeo.2022.105135.
Khorramian, K., A. Alhashmi, and F. Oudah. 2023b. “Efficient representation of random fields for training the Kriging predictor in adaptive Kriging reliability assessment of civil structures.” In Automation in construction toward resilience, edited by E. N. Farsangi, M. Noori, T. T. Y. Yang, P. B. Lourenço, P. Gardoni, I. Takewaki, E. Chatzi, and S. Li, 485–502. Boca Raton, FL: CRC Press.
Khorramian, K., and F. Oudah. 2022. “Active learning Kriging-based reliability for assessing the safety of structures: Theory and application.” In Leveraging artificial intelligence into engineering, management, and safety of infrastructure, edited by M. Z. Naser, 184–231. Boca Raton, FL: CRC Press.
Khorramian, K., and F. Oudah. 2023. “New learning functions for active learning Kriging reliability analysis using a probabilistic approach: KO and WKO functions.” Struct. Multidiscip. Optim. 66 (8): 177. https://doi.org/10.1007/s00158-023-03627-4.
Khorramian, K., and F. Oudah. 2024. “Metric systems for performance evaluation of active learning Kriging configurations for reliability analysis.” ASCE-ASME J. Risk Uncertainty Eng. Syst. Part A: Civ. Eng. 10 (2): 04024024. https://doi.org/10.1061/AJRUA6.RUENG-1034.
Lophaven, S. N., H. B. Nielsen, and J. Søndergaard. 2002. “DACE: A Matlab kriging toolbox.” In Vol. 2 of IMM informatics and mathematical modelling, 1–34. Lyngby, Denmark: Technical Univ. of Denmark.
Lu, P., T. Hong, Y. Wu, Z. Xu, D. Li, Y. Ma, and L. Shao. 2022. “Kriging–KNN hybrid analysis method for structural reliability analysis.” J. Bridge Eng. 27 (4): 04022009. https://doi.org/10.1061/(ASCE)BE.1943-5592.0001837.
Lv, Z., Z. Lu, and P. Wang. 2015. “A new learning function for Kriging and its applications to solve reliability problems in engineering.” Comput. Math. Appl. 70 (5): 1182–1197. https://doi.org/10.1016/j.camwa.2015.07.004.
Melchers, R. 2007. “Structural reliability theory in the context of structural safety.” Civ. Eng. Environ. Syst. 24 (1): 55–69. https://doi.org/10.1080/10286600601025191.
Melchers, R., and A. Beck. 2018. Structural reliability analysis and prediction. 3rd ed. Hoboken, NJ: John Wiley & Sons.
Moustapha, M., S. Marelli, and B. Sudret. 2022. “Active learning for structural reliability: Survey, general framework and benchmark.” Struct. Saf. 96 (2): 102174. https://doi.org/10.1016/j.strusafe.2021.102174.
Ni, P., J. Li, H. Hao, and H. Zhou. 2021. “Reliability based design optimization of bridges considering bridge-vehicle interaction by Kriging surrogate model.” Eng. Struct. 246: 112989. https://doi.org/10.1016/j.engstruct.2021.112989.
Nowak, A. S., and M. Szerzen. 2003. “Calibration of design code for buildings: Part 1—Statistical models for resistance.” ACI Struct. J. 100 (3): 377–382.
Petrie, C., and F. Oudah. 2024. “Clustering-based active learning reliability of FRP strengthened RC beams with random field nonlinear finite element to model spatial variability.” J. Compos. Constr. 28 (5): 04024045.
Rakoczy, A. M., and A. S. Nowak. 2013. “Reliability-based sensitivity analysis for prestress concrete girder bridges.” PCI J. 58 (2): 81–92. https://doi.org/10.15554/pcij.09012013.81.92.
Ren, C., Y. Aoues, D. Lemosse, and E. S. De Cursi. 2022. “Ensemble of surrogates combining Kriging and artificial neural networks for reliability analysis with local goodness measurement.” Struct. Saf. 96: 102186. https://doi.org/10.1016/j.strusafe.2022.102186.
Stewart, M. G. 1998. “Reliability-based bridge design and assessment.” Prog. Struct. Mater. Eng. 1 (2): 214–222. https://doi.org/10.1002/pse.2260010215.
Sun, Z., J. Wang, R. Li, and C. Tong. 2017. “LIF: A new Kriging based learning function and its application to structural reliability analysis.” Reliab. Eng. Syst. Saf. 157: 152–165. https://doi.org/10.1016/j.ress.2016.09.003.
Tabsh, S. W., and A. S. Nowak. 1991. “Reliability of highway girder bridges.” J. Struct. Eng. 117 (8): 2372–2388. https://doi.org/10.1061/(ASCE)0733-9445(1991)117:8(2372).
Xiao, S., S. Oladyshkin, and W. Nowak. 2020. “Reliability analysis with stratified importance sampling based on adaptive Kriging.” Reliab. Eng. Syst. Saf. 197: 106852. https://doi.org/10.1016/j.ress.2020.106852.
You, X., M. Zhang, D. Tang, and Z. Niu. 2022. “An active learning method combining adaptive kriging and weighted penalty for structural reliability analysis.” Proc. Inst. Mech. Eng., Part O: J. Risk Reliab. 236 (1): 160–172. https://doi.org/10.1177/1748006X211016148.
Zhang, L., Z. Lu, and P. Wang. 2015. “Efficient structural reliability analysis method based on advanced Kriging model.” Appl. Math. Modell. 39 (2): 781–793. https://doi.org/10.1016/j.apm.2014.07.008.
Zhang, X., L. Wang, and J. D. Sørensen. 2019. “REIF: A novel active-learning function toward adaptive Kriging surrogate models for structural reliability analysis.” Reliab. Eng. Syst. Saf. 185: 440–454. https://doi.org/10.1016/j.ress.2019.01.014.
Zhang, Y., and G. Wu. 2019. “Seismic vulnerability analysis of RC bridges based on Kriging model.” J. Earthquake Eng. 23 (2): 242–260. https://doi.org/10.1080/13632469.2017.1323040.
Information & Authors
Information
Published In
Copyright
© 2024 American Society of Civil Engineers.
History
Received: Oct 2, 2023
Accepted: Jul 11, 2024
Published online: Oct 17, 2024
Published in print: Jan 1, 2025
Discussion open until: Mar 17, 2025
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.