Journal of Scientific and Industrial Research

Open Access Open Access  Restricted Access Subscription Access

Comparison of NDT Data Fusion for Concrete Strength using Decision Tree and Artificial Neural Network


Affiliations
1 Nuclear Recycle Board, Bhabha Atomic Research Centre, Mumbai 400 094, India
 

Fusion of Non-Destructive Test (NDT) data results in more accurate estimation of concrete strength when compared to any single NDT data. Estimation of concrete strength from NDT results assumes importance for health assessment and evaluation of existing concrete buildings, particularly those near the end of their design life. Application of machine learning tools and response surface method has found popularity in recent years for this purpose. In this study, universally popular Artificial Neural Network (ANN) and relatively un-explored Decision Tree (DT) are applied to estimate concrete strength from rebound number and ultrasonic pulse velocity data collected from literature, in single and combined forms. A ranking system based on ratios of multiple performance measures was demonstrated for cases where different models are adjudged better considering different performance measures. From the results, it was concluded that fusion of NDT data resulted in better accuracy, for both ANN and DT. Comparing the selected performance measures as well as the ranks of the two machine learning tools, ANN models were found to perform better as compared to the DT models. The narrow range of multiple performance metrics obtained for three different data divisions (into modelling and evaluation sets) in all cases imparted confidence in the robustness of the approach of model development adopted in this study.

Keywords

Design life, Multiple performance measures, Non-destructive testing, Rebound number, Ultrasonic pulse velocity.
User
Notifications
Font Size

  • Erdal M, Prediction of the compressive strength of vacuum processed concretes using artificial neural network and regression techniques, Sci Res Essay, 4(10) (2009) 1057–1065.

  • Amini K, Jalalpour M & Delatte N, Advancing concrete strength prediction using non-destructive testing: development and verification of a generalizable model, Constr Build Mater, 102 (2016) 762–768.

  • Ali-Benyahia K, Sbartai Z M, Breysse D, Kenai S & Ghrici M, Analysis of the single and combined non-destructive test approaches for on-site concrete strength assessment: General statements based on a real case-study, Case Stud Constr Mater, 6 (2017) 109–119, DOI: 10.1016/j.cscm.2017.01.004.

  • Dauji S, Bhalerao S, Srivastava P K & Bhargava K, Conservative characteristic strength of concrete from non destructive and partially destructive testing, J Asian Concr Fed, 5(1) (2019) 25–39, DOI: 10.18702/acf.2019.06.30.25.

  • Karahan S, Büyüksaraç A & Işık E, The relationship between concrete strengths obtained by destructive and non-destructive methods, Iran J Sci Technol–Trans Civ Eng, 44 (2020) 91–105, DOI: 10.1007/s40996-019-00334-3.

  • Sbartai Z M, Breysse D, Larget M & Balayssac J P, Combining NDT techniques for improved evaluation of concrete properties, Cem Concr Compos, 34 (2012) 725–733, DOI: 10.1016/j.cemconcomp.2012.03.005.

  • Poorarbabi A, Ghasemi M & Moghaddam M A, Concrete compressive strength prediction using non-destructive tests through response surface methodology, Ain Shams Eng J, 11 (2020) 939–949, DOI: 10.1016/j.asej.2020.02.009.

  • Tahwia A M, Heniegal A, Elgamal M S & Tayeh B A, The prediction of compressive strength and non-destructive tests of sustainable concrete by using artificial neural networks, Comput Concr, 27(1) (2021) 21–28, DOI: 10.12989/cac. 2021.27.1.021.

  • Silva F A N, Delgado J M P Q, Cavalcanti R S, Azevedo A C, Guimaraes A S & Lima A G B, Use of non-destructive testing of ultrasound and artificial neural networks to estimate compressive strength of concrete, Buildings, 11(44) (2021) 1–15, DOI: 10.3390/buildings11020044.

  • Bilgehan M, A comparative study for the concrete compressive strength estimation using neural network and neuro-fuzzy modelling approaches, Nondestruct Test Evaluation, 26(01) (2011) 35–55, DOI: 10.1080/105897 51003770100.

  • Park J Y, Yoon Y G & Oh T K, Prediction of concrete strength with P-, S-, R-wave velocities by support vector machine (SVM) and artificial neural network (ANN), Appl Sci, 9(4053) (2019) 1–18, DOI: 10.3390/app9194053.

  • Tenza-Abril A J, Villacampa Y, Solak A M & Baeza-Brotons F, Prediction and sensitivity analysis of compressive strength in segregated lightweight concrete based on artificial neural network using ultrasonic pulse velocity, Constr Build Mater, 189 (2018) 1173–1183, DOI: 10.1016/j.conbuildmat. 2018.09.096.

  • Demir A, Prediction of hybrid fibre-added concrete strength using artificial neural networks, Comput Concr, 15(4) (2015) 503–514, DOI: 10.12989/cac.2015.15.4.503.

  • Chun P, Ujike I, Mishima K, Kusumoto M & Okazaki S, Random forest-based evaluation technique for internal damage in reinforced concrete featuring multiple non-destructive testing results, Constr Build Mater, 253(119238) (2020) 1–11, DOI: 10.1016/j.conbuildmat.2020.119238.

  • Ahmed L, Obanishola S, Ibrahim B, Abdulfatai T L & Nii A O, A boosted tree machine learning alternative to predictive evaluation of non-destructive concrete compressive strength, in 18th IEEE International Conference on Machine Learning and Applications (ICMLA), (Boca Raton, FL, USA) 16–19 December 2019, DOI: 10.1109/ICMLA.2019.00060.

  • Du G, Bu L, Hou Q, Zhou J & Lu B, Prediction of the compressive strength of high-performance self-compacting concrete by an ultrasonic-rebound method based on a GA-BP neural network, PLoS ONE, 16(5)e0250795 (2021) 1–25, DOI: 10.1371/journal.pone.0250795.

  • Ahmad A, Farooq F, Niewiadomski P, Ostrowski K, Akbar A, Aslam F & Alyousef R, Prediction of compressive strength of fly ash based concrete using individual and ensemble algorithm, Materials, 14(794) (2021), DOI: 10.3390/ma140 40794.

  • Kocamaz A F, Ayaz Y, Karakoc M B, Turkmen I & Demirboga R, Prediction of compressive strength and ultrasonic pulse velocity of admixtured concrete using tree model M5P, Struct Concr, 1 (2020) 1–15. DOI: 10.1002/suco. 202000061.

  • Chopra P, Sharma R K, Kumar M & Chopra T, Comparison of machine learning techniques for the prediction of compressive strength of concrete, Adv Civ Eng, 5481705 (2018) 1–9, DOI: 10.1155/2018/5481705.

  • Dauji S, Prediction of compressive strength of concrete with decision trees, Int J Concr Technol, 2(1) (2016) 19–29. 21 Dauji S, Estimation of capacity of eccentrically loaded single angle struts with decision trees, Chall J Struct Mech, 5(1) (2019) 1–8, DOI: 10.20528/cjsmec.2019.01.001.

  • Dauji S & Deo M C, Improving numerical current prediction with model tree, Indian J Geo-Mar Sci, 49(08) (2020) 1350– 1358.

  • Dauji S, Prediction accuracy of underground blast variables: decision tree and artificial network, Int J Earthq Impact Eng, 3(1) (2020) 40–59. DOI: 10.1504/IJEIE.2020.105382.

  • Rafi A, Dauji S & Bhargava K, Estimation of SPT from coarse grid data by spatial interpolation technique, in Geotechnical Characterization and Modelling, Lect Notes Civ Eng 85, edited by Gali M L & RRP (Springer Nature, Singapore) 2020, 1079–1091, DOI: 10.1007/978-981-15-6086-6_87.

  • Tharwat A, Classification assessment methods, Appl Comput Inform, 17(1) (2021) 168–192, DOI 10.1016/j.aci.2018.08.003

  • Fawcett T, An introduction to ROC analysis, Pattern Recognit Lett, 27 (2006) 861–874, DOI:10.1016/j.patrec.2005.10.010

  • Piryonesi S M & El-Diraby T E, Data analytics in asset management: Cost-effective prediction of the pavement condition index, J Infrastruct Syst, 26(1) (2020) 04019036, DOI: 10.1061/(ASCE)IS.1943-555X.0000512.

  • Rokach L & Maimon O, Data Mining with Decision trees: Theory and Applications (World Scientific, Singapore) 2015.

  • Quinlan J R, C4.5: Programs for Machine Learning (Morgan Kaufmann, San Francisco) 1992.

  • Jekabsons G, M5PrimeLab: M5' regression tree, model tree, and tree ensemble toolbox for Matlab/Octave, [http://www.cs. rtu.lv/jekabsons/ (08 May 2016)].

  • Wasserman P D, Advanced Methods in Neural Computing (Van Nostrand Reinhold Company, New York) 1993.

  • Witten I H & Frank E, Data Mining: Practical Machine Learning Tools and Techniques (Morgan Kaufmann, San Francisco) 2000.

  • Bose N K & Liang P, Neural Networks Fundamentals with Graphs, Algorithms, and Applications (Tata-McGraw-Hill Publishing Company Limited, New Delhi, India) 1993.

  • Haykin S O, Neural Networks and Machine Learning (Pearson Education, New Delhi, India) 2008.


Abstract Views: 89

PDF Views: 53




  • Comparison of NDT Data Fusion for Concrete Strength using Decision Tree and Artificial Neural Network

Abstract Views: 89  |  PDF Views: 53

Authors

Saha Dauji
Nuclear Recycle Board, Bhabha Atomic Research Centre, Mumbai 400 094, India

Abstract


Fusion of Non-Destructive Test (NDT) data results in more accurate estimation of concrete strength when compared to any single NDT data. Estimation of concrete strength from NDT results assumes importance for health assessment and evaluation of existing concrete buildings, particularly those near the end of their design life. Application of machine learning tools and response surface method has found popularity in recent years for this purpose. In this study, universally popular Artificial Neural Network (ANN) and relatively un-explored Decision Tree (DT) are applied to estimate concrete strength from rebound number and ultrasonic pulse velocity data collected from literature, in single and combined forms. A ranking system based on ratios of multiple performance measures was demonstrated for cases where different models are adjudged better considering different performance measures. From the results, it was concluded that fusion of NDT data resulted in better accuracy, for both ANN and DT. Comparing the selected performance measures as well as the ranks of the two machine learning tools, ANN models were found to perform better as compared to the DT models. The narrow range of multiple performance metrics obtained for three different data divisions (into modelling and evaluation sets) in all cases imparted confidence in the robustness of the approach of model development adopted in this study.

Keywords


Design life, Multiple performance measures, Non-destructive testing, Rebound number, Ultrasonic pulse velocity.

References