Contribution of Machine Learning Methods to the Construction Industry: Prediction of Compressive Strength

Abstract

Highly accurate prediction of high performance concrete (HPC) compressive strength is very important issue. In recent years, a variety of modeling approaches and methodologies have been applied to predict HPC’s compressive strength from a wide range of variables, with different ratios of success. In this study, an appropriate machine learning method, using different mixing ratios for the prediction of compressive strength of HPC, is investigated. In recent years, rather developing machine learning methods; Artificial Neural Networks (ANN) and Support Vector Machines (SVM)’s applicabilities for the prediction, handled in this study, are being investigated and extremely high results were obtained. In this paper, it’s obtained that prediction success of SVM has been found more satisfactory than ANN's. It is concluded that the SVM’s can be used effectively as an alternative method by research labs and the concrete firms for predicting the strength.

Keywords

• High performance concrete, Artificial neural networks, Support vector machines

Makine Öğrenmesi Yöntemlerinin İnşaat Sektörüne Katkısı: Basınç Dayanımı Tahminlemesi

Öz

Yüksek performanslı beton (high performance concrete, HPC)’un eksenel basınç dayanımının yüksek doğrulukla tahmini son derece önemli bir konudur. Geçtiğimiz yıllarda, çeşitli gelişmiş modelleme yaklaşımları ve metodolojileri kullanılarak farklı başarı oranları ile HPC basınç dayanımı tahminlemeleri uygulanmıştır. Bu çalışmada farklı karışım oranları kullanılarak HPC’lerin eksenel basınç dayanımının tahmininde uygun bir makine öğrenmesi yöntemi araştırılmıştır. Son yıllarda makine öğrenmesinde oldukça gelişmekte olan Yapay Sinir Ağları (YSA) ve Destek Vektör Makineleri (DVM)’nin bu tahminde uygulanabilirliği incenmiş ve son derece yüksek tahmin sonuçları elde edilmiştir. Bu çalışmada DVM’lerin tahmin başarısının YSA’lara oranla daha tatmin edici sonuçlar verdiği görülmüştür. DVM yönteminin araştırma laboratuvarları ve beton firmaları tarafından dayanım tahmininde alternatif bir yöntem olarak etkin bir şekilde kullanılabileceği sonucuna varılmıştır.

Anahtar Kelimeler

• Yüksek performanslı beton, Yapay sinir ağları, Destek vektör makineleri

Kaynakça

1. Machine [1] Yeh IC. “UCI Compressive Repository: Concrete
2. http://archive.ics.uci.edu/ml/datasets/Concrete+Compr
3. essive+Strength (20.03.2014). Strength Data Set” [2] Yeh IC. “Modeling of Strength of High-Performance
4. Concrete Using Artificial Neural Networks”. Cement and
5. Concrete Research, 28(12), 1797-1808, 1998. [3] Hong-Guang N, Ji-Zong W. “Prediction of Compressive
6. Strength of Concrete by Neural Networks”. Cement and
7. Concrete Research, 30(8), 1245-1250, 2000. [4] Seyhan AT, Tayfur G, Karakurt M, Tanoğlu M. “Artificial
8. Neural Network (ANN) Prediction of Compressive

9. Strength of VARTM Processed Polymer Composites”.
10. Computational Materials Science, 34(1), 99-105, 2005. [5] Gupta R, Kewalramani MA, Goel A. “Prediction of Concrete
11. Strength Using Neural-Expert System”. Journal of
12. Materials in Civil Engineering, 18(3), 462-466, 2006. [6] Topcu IB, Sarıdemir M. “Prediction of Compressive
13. Strength of Concrete Containing Fly Ash Using Artificial
14. Neural Networks and Fuzzy Logic”. Computational
15. Materials Science, 41(3), 305-311, 2008. [7] Raghu Prasad BK, Eskandari H, Venkatarama Reddy BV.
16. “Prediction of Compressive Strength of SCC and HPC with
17. High Volume fly ash Using ANN”. Construction and
18. Building Materials, 23(1), 117-128, 2009. [8] Khan MI. “Predicting Properties of High Performance
19. Concrete Containing Composite Cementitious Materials
20. Using Artificial Neural Networks”. Automation in
21. Construction, 22, 516-524, 2012. [9] Atıcı U. “Prediction of the Strength of Mineral Admixture
22. Concrete Using Multivariable Regression Analysis and an
23. Artificial Neural Network”. Expert Systems with
24. Applications, 38(8), 9609-9618, 2011. [10] Nikoo M, Zarfam P, Sayahpour H. “Determination of
25. Compressive Strength of Concrete Using Self Organization
26. Feature Map (SOFM)”. Engineering with Computers, 31(1),
27. 113-121, 2015. [11] Fazel-Zarandi MH, Türksen IB, Sobhani J, Ramezanianpour AA.
28. Approximation of the Compressive Strength of Concrete”.
29. Applied Soft Computing, 8(1), 488-498, 2008. for [12] Yeh IC, Lien LC. “Knowledge Discovery of Concrete
30. Material Using Genetic Operation Trees”. Expert Systems
31. with Applications, 36(3), 5807–5812, 2009. [13] Cheng MY, Peng HS, Wu YW, Chen TL. “Estimate at
32. Completion for Construction Projects Using Evolutionary
33. Support Vector Machine Inference Model”. Automation in
34. Construction, 19(5), 619–629, 2010. [14] Wang YR, Yu CY, Chan HH. “Predicting Construction Cost
35. and Schedule Success Using Artificial Neural Networks
36. Ensemble and Support Vector Machines Classification
37. Models”. International Journal of Project Management,
38. 30(4), 470-478, 2012. [15] Cheng MY, Chou JS, Roy AFV, Wu YW. “High-performance
39. Concrete Compressive Strength Prediction using Time
40. Weighted Evolutionary Fuzzy Support Vector Machines
41. Inference Model”. Automation in Construction, 28,
42. 106–115, 2012. [16] Cheng MY, Firdausi PM, Prayogo D. “High-performance
43. Concrete Compressive Strength Prediction using Genetic Weighted
44. Engineering Applications of Artificial Intelligence.
45. 29, 104-113, 2014. Tree (GWPOT)”. [17] Chou JS, Chiu CK, Farfoura M, Al-Taharwa I. “Optimizing
46. the Prediction Accuracy of Concrete Compressive Strength
47. Based on A Comparison of Data-mining Techniques”.
48. Journal of Computing in Civil Engineering, 25(3), 242-253, 2011. [18] Abraham A, Philip NS, Saratchandran P. “Modeling Chaotic
49. Behavior of Stock Indices Using Intelligent Paradigms”.
50. Neural, Parallel & Scientific Computations, 11(1-2), 143- 160, 2003. [19] Huang S, Wu T. “Integrating GA-based Time-scale Feature
51. Extractions with SVMs for Stock Index Forecasting”.
52. Expert Systems with Applications, 35(4), 2080-2088, 2008. [20] Öztemel E. Yapay Sinir Ağları. İstanbul, Türkiye, Papatya
53. Yayıncılık, 2003. [21] Elmas Ç. Yapay Sinir Ağları Kuram, Mimari, Eğitim,
54. Uygulama. 1. Baskı, Ankara, Türkiye, Seçkin Yayıncılık, 2003. [22] Bayramoğlu MF. Finansal Endekslerin Öngörüsünde
55. Yapay Sinir Ağı Modellerinin Kullanılması: İMKB Ulusal
56. 100 Endeksinin Gün İçi En Yüksek ve En Düşük Değerlerinin
57. Yayımlanmamış Yüksek Lisans Tezi. Karaelmas
58. Üniversitesi, Zonguldak, Türkiye, 2007. Bir Uygulama.
59. Based Aerodynamic Analysis of Cable Stayed Bridges”.
60. Advances in Engineering Software, 40(9), 830-835, 2009.