Estimation of the coefficient of permeability as an example of the application of the Random Forest algorithm in Civil Engineering

• Autorzy: Dzięcioł Justyna , Sas Wojciech

Identyfikatory

DOI

10.24425/ace.2024.149854

Warianty tytułu

PL

Szacowanie współczynnika filtracji jako przykład zastosowania algorytmu Random Forest w budownictwie lądowym

• Języki publikacji: EN

Abstrakty

EN

A new world record for crude steel production was recorded in 2021, which increased by 3.8% over 2020. This also affected the amount of slag produced with this production. Total waste from industrial and construction production throughout the European Union accounts for as much as 48%. Therefore, waste management should provide for the recovery of as many resources as possible. European Union strategies in line with the circular economy objectives focus on ensuring policy coherence in the areas of climate, energy efficiency, construction and demolition waste management and resource efficiency. Slags are a material of interest to researchers in terms of their use in construction. Slags, on the one hand, are materials that are becoming better understood on the other hand, we are making sure of the heterogeneity of these materials. The characteristics of physical properties of slags are influenced by many factors, including the furnace split in which they are produced. This prompts the search for tools to help determine the parameters of slags based on already available data. The study aimed to verify the hypothesis that it is possible to determine the parameter of the filtration coefficient, relevant to applications in earth structures using the machine learning algorithm – Random Forest. In the study, two types of material were analysed: blast furnace slag and furnace slag. The results of the analysis yielded a high coefficient of determination (R2) – 0.84-0.92. This leads us to believe that the algorithm may prove useful in determining filtration parameters in slags.

PL

Nowy światowy rekord produkcji stali surowej odnotowano w 2021 r., wzrost wynosił 3,8% w stosunku do roku 2020. Wpłynęło to jednocześnie na ilość wytwarzanego wraz z tą produkcją żużla. Całkowite ilości odpadów pochodzące z produkcji przemysłowej i budowlanej w całej Unii Europejskiej stanowią aż 48%. Dlatego, gospodarka odpadami powinna zapewniać odzysk jak największej ilości zasobów. Strategie Unii Europejskiej zgodne z celami gospodarki cyrkularnej koncentrują się na zapewnieniu spójności polityki w obszarach: klimatu, efektywności energetycznej, gospodarowania odpadami z budowy i rozbiórki oraz zasobooszczędności. Żużle są materiałem, który interesuje badaczy pod kątem ich zastosowania w budownictwie. Żużle z jednej strony są materiałami coraz lepiej poznanymi z drugiej upewniamy się o niejednorodności tych materiałów. Na charakterystykę właściwości fizycznych żużli wpływa wiele czynników m.in. rozdaj pieca w jakim powstają. Skłania to do poszukiwania narzędzi pomocnych w wyznaczaniu parametrów żużli, jak i materiałów antropogenicznych. Celem badań było zweryfikowanie hipotezy, że możliwe jest wyznaczanie parametru współczynnika filtracji, istotnego do zastosowań w konstrukcjach ziemnych z wykorzystaniem algorytmu uczenia maszynowego – Random Forest. W pracy przeanalizowano dwa rodzaje materiału: żużel wielkopiecowy oraz paleniskowy. Wyniki analizy pozwoliły na otrzymanie wysokiego współczynnika determinacji (R2) – 0.84-0.92. Pozwala to sądzić, że algorytm ten może okazać się użyteczny przy wyznaczaniu parametrów filtracyjnych w żużlach.

• Wydawca: Komitet Inżynierii Lądowej i Wodnej PAN ; Politechnika Warszawska, Wydział Inżynierii Lądowej
• Czasopismo: Archives of Civil Engineering
• Rocznik: 2024
• Tom: Vol. 70, nr 2
• Strony: 119--134
• Opis fizyczny: Bibliogr. 45 poz., il., tab.

Twórcy

autor

Dzięcioł Justyna

• Warsaw University of Life Sciences – SGGW, Institute of Civil Engineering, Warsaw, Poland

• https://orcid.org/0000-0002-2436-9748

autor

Sas Wojciech

• Warsaw University of Life Sciences – SGGW, Water Centre SGGW, Warsaw, Poland

• https://orcid.org/0000-0002-5488-3297

Bibliografia

• [1] J. Dzięcioł and M. Radziemska, “Blast furnace slag, post-industrial waste or valuable building materials with remediation potential?”, Minerals, vol. 12, no. 4, art. no. 478, 2022, doi: 10.3390/min12040478.
• [2] G.C. Ulubeyli and R. Artir, “Sustainability for blast furnace slag: use of some construction wastes”, Procedia Social and Behavioral Sciences, vol. 195, pp. 2191-2198, 2015, doi:10.1016/J.SBSPRO.2015.06.297.
• [3] R. Trach, M. Połoński, and P. Hrytsiuk, “Decision making in choosing a network organizational structure in integrated construction projects”, Archives of Civil Engineering, vol. 67, no. 2, pp. 195-208, 2021, doi: 10.24425/ACE.2021.137163.
• [4] J. Witkowska-Dobrev, et al., “Effect of sewage on compressive strength and geometric texture of the surface of concrete elements”, Structural Concrete, vol. 24, no. 1, pp. 468-484, 2023, doi: 10.1002/suco.202200467.
• [5] T. He, Z. Li, S. Zhao, X. Zhao, and X. Qu, “Study on the particle morphology, powder characteristics and hydration activity of blast furnace slag prepared by different grinding methods”, Construction and Building Materials, vol. 270, art. no. 121445, 2021, doi: 10.1016/J.CONBUILDMAT.2020.121445.
• [6] M. Radziemska, et al., “Recycling of blast furnace and coal slags in aided phytostabilisation of soils highly polluted with heavy metals”, Energies (Basel), vol. 14, no. 14, 2021, doi: 10.3390/en14144300.
• [7] M. Valcuende, F. Benito, C. Parra, and I. Miñano, “Shrinkage of self-compacting concrete made with blast furnace slag as fine aggregate”, Construction and Building Materials, vol. 76, pp. 1-9, 2015, doi: 10.1016/j.conbuildmat.2014.11.029.
• [8] N.D. Lagaros, “Artificial neural networks applied in civil engineering”, Applied Sciences, vol. 13, no. 2, art. no. 1131, 2023, doi: 10.3390/APP13021131.
• [9] I. Flood and N. Kartam, “Neural networks in civil engineering. I: principles and understanding”, Journal of Computing in Civil Engineering, vol. 8, no. 2, pp. 131-148, 1994, doi: 10.1061/(ASCE)0887-3801(1994)8:2(131).
• [10] I. Flood, “Towards the next generation of artificial neural networks for civil engineering”, Advanced Engineering Informatics, vol. 22, no. 1, pp. 4-14, 2008, doi: 10.1016/J.AEI.2007.07.001.
• [11] R. Trach, Y. Trach, and M. Lendo-Siwicka, “Using ANN to predict the impact of communication factors on the rework cost in construction projects”, Energies, vol. 14, no. 14, art. no. 4376, 2021, doi: 10.3390/EN14144376.
• [12] J. Dzięcioł and W. Sas, “Perspective on the application of machine learning algorithms for flow parameter estimation in recycled concrete aggregate”, Materials, vol. 16, no. 4, art. no. 1500, 2023, doi: 10.3390/MA16041500.
• [13] M.F. Hasan, O. Hammody, and K.S. Albayati, “Estimate final cost of roads using support vector machine”, Archives of Civil Engineering, vol. 68, no. 4, pp. 669-682, 2022, doi: 10.24425/ace.2022.143061.
• [14] L. Breiman, “Random forests”, Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
• [15] L. Breiman, “Bagging predictors”, Machine Learning, vol. 26, no. 2, pp. 123-140, 1996.
• [16] R. Caruana and A. Niculescu-Mizil, “An empirical comparison of supervised learning algorithms”, ACM International Conference Proceeding Series, vol. 148, pp. 161-168, 2006, doi: 10.1145/1143844.1143865.
• [17] G. James, D. Witten, T. Hastie, and R. Tibshirani, An introduction to statistical learning with applications in R. Springer, 2013.
• [18] G. Louppe, “Understanding Random Forests: from theory to practice”, PhD thesis, University of Liege, Belgium, 2014, doi: 10.48550/arxiv.1407.7502.
• [19] E. Scornet, “Tuning parameters in random forests”, ESAIM Proceedings and Surv, vol. 60, pp. 144-162, 2018, doi: 10.1051/proc/201760144.
• [20] M.N. Wright, S. Wager, and P. Probst, “Ranger: a fast implementation of Random Forests”, 2022.
• [21] S.J. Wright and B. Recht, Optimization for data analysis. Cambridge University Press, 2022, doi: 10.1017/9781009004282.
• [22] M.N. Wright and A. Ziegler, “Ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R”, Journal of Statistical Software, vol. 77, no. 1, pp. 1-17, 2017, doi: 10.18637/JSS.V077.I01.
• [23] A. Liaw and M. Wiener, “Classification and Regression by randomForest”, R News, vol. 2, no. 3, 2002.
• [24] M. Kuhn, The caret Package. 2019. http://jmlr.org/papers/v18/17-269.html
• [25] M. Kuhn, “Building Predictive Models in R Using the caret Package”, Journal of Statistical Software, vol. 28, no. 5, pp. 1-26, 2008, doi: 10.18637/JSS.V028.I05.
• [26] R. Genuer, J.-M. Poggi, and C. Tuleau, “Random Forests: some methodological insights”, Nov. 2008, doi: 10.48550/arxiv.0811.3619.
• [27] R. Genuer and J.-M. Poggi, Introduction to Random Forests with R. Cham: Springer International Publishing, 2020, pp. 1-8. doi: 10.1007/978-3-030-56485-8_1.
• [28] A. Parmar, R. Katariya, and V. Patel, “A Review on Random Forest: An Ensemble Classifier”, in Lecture Notes on Data Engineering and Communications Technologies, vol. 26. Springer Science and Business Media Deutschland GmbH, 2019, pp. 758-763, doi: 10.1007/978-3-030-03146-6_86.
• [29] P. Probst and A.-L. Boulesteix, “To Tune or Not to Tune the Number of Trees in Random Forest”, Journal of Machine Learning Research, vol. 18, pp. 1-18, 2018. http://jmlr.org/papers/v18/17-269.html
• [30] W. Sas, J. Dzięcioł , and A. Głuchowski, “Estimation of recycled concrete aggregate’s water permeability coefficient as earth construction material with the application of an analytical method”, Materials, vol. 12, no. 18, art. no. 2920, 2019, doi: 10.3390/ma12182920.
• [31] W. Sas and J. Dzięcioł , “Determination of the filtration rate for anthropogenic soil from the recycled concrete aggregate by analytical methods”, Scientific Review Engineering and Environmental Sciences, vol. 27, no. 2. pp. 236-248, 2018, doi: 10.22630/PNIKS.2018.27.2.23.
• [32] W. Sas, et al., “Geotechnical and environmental assessment of Blast Furnace Slag for engineering applications”, Materials, vol. 14, no. 20, art. no. 6029, 2021, doi: 10.3390/ma14206029.
• [33] PN-EN 13286-2:2007 Unbound and hydraulic binder mixtures: Part 2: Methods for determining density in relation to water content. Proctor compaction.
• [34] M.W. Browne, “Cross-Validation methods”, Journal of Mathematical Psychology, vol. 44, no. 1, pp. 108-132, 2000, doi: 10.1006/JMPS.1999.1279.
• [35] T. Fushiki, “Estimation of prediction error by using K-foldcross-validation”, Statistics and Computing, vol. 21, no. 2, pp. 137-146, 2011.
• [36] S. Janitza and R. Hornung, “On the overestimation of random forest’s out-of-bag error”, PLoS One, vol. 13, no. 8, art. no. e0201904, 2018, doi:10.1371/JOURNAL.PONE.0201904.
• [37] G. Martínez-Muñoz and A. Suárez, “Out-of-bag estimation of the optimal sample size in bagging”, Pattern Recognition, vol. 43, no. 1, pp. 143-152, 2010, doi: 10.1016/J.PATCOG.2009.05.010.
• [38] M.Vukovic and A. Soro, Determination of hydraulic conductivity of porous media from grain-size composition. Littleton: Water Resources Publications, 1992.
• [39] S.M. Lundberg and S.I. Lee, “A unified approach to interpreting model predictions”, in Neural Information Processing Systems, vol. 2017. 2017, pp. 4766-4775, doi: 10.48550/arxiv.1705.07874.
• [40] L. Merrick and A. Taly, “The Explanation Game: Explaining Machine Learning Models Using Shapley Values”, in Machine Learning and Knowledge Extraction. Lecture Notes in Computer Science, vol. 12279. Springer, 2020, pp. 17-38, doi: 10.1007/978-3-030-57321-8_2.
• [41] L.S. Shapley, “A Value for n-Person Games”, in Contributions to the Theory of Games (AM-28), vol. 2. Princeton University Press, 1953, pp. 307-318.
• [42] B. Rozemberczki, et al., “The Shapley Value in Machine Learning”, in Proceedings of the 31st International Joint Conference on Artifical Intelligence, IJCAI22. International Joint Conferences on Artificial Intelligence Organization, 2022, pp. 5572-5579, doi: 10.24963/ijcai.2022/778.
• [43] M. Anjum, K. Khan,W. Ahmad, A. Ahmad, M.N. Amin, and A. Nafees, “New SHapley Additive ExPlanations (SHAP) Approach to Evaluate the Raw Materials Interactions of Steel-Fiber-Reinforced Concrete”, Materials, vol. 15, no. 18, art. no. 6261, 2022, doi: 10.3390/ma15186261.
• [44] I.U. Ekanayake, D.P.P. Meddage, and U. Rathnayake, “A novel approach to explain the black-box nature of machine learning in compressive strength predictions of concrete using Shapley additive explanations (SHAP)”, Case Studies in Construction Materials, vol. 16, 2022, doi: 10.1016/j.cscm.2022.e01059.
• [45] H. Errousso, E.A. Abdellaoui Alaoui, S. Benhadou, and H. Medromi, “Exploring how independent variables influence parking occupancy prediction: toward a model results explanation with SHAP values”, Progress in Artificial Intelligence, vol. 11, no. 4, pp. 367-396, 2022, doi: 10.1007/S13748-022-00291-5.