İnşaat Mühendisliği Bölümü Koleksiyonu
Bu koleksiyon için kalıcı URI
Güncel Gönderiler
Öğe Neural Network Predictive Models for Alkali-Activated Concrete Carbon Emission Using Metaheuristic Optimization Algorithms(2024) Aydın, Yaren; Çakıroğlu, Celal; Bekdaş, Gebrail; Işıkdağ, Ümit; Kim, Sanghun; Hong, Junhee; Geem, Zong WooDue to environmental impacts and the need for energy efficiency, the cement industry aims to make more durable and sustainable materials with less energy requirements without compromising mechanical properties based on UN Sustainable Development Goals 9 and 11. Carbon dioxide (CO2) emission into the atmosphere is mostly the result of human-induced activities and causes dangerous environmental impacts by increasing the average temperature of the earth. Since the production of ordinary Portland cement (PC) is a major contributor to CO2 emissions, this study proposes alkali-activated binders as an alternative to reduce the environmental impact of ordinary Portland cement production. The dataset required for the training processes of these algorithms was created using Mendeley as a data-gathering instrument. Some of the most efficient state-of-the-art meta-heuristic optimization algorithms were applied to obtain the optimal neural network architecture with the highest performance. These neural network models were applied in the prediction of carbon emissions. The accuracy of these models was measured using statistical measures such as the mean squared error (MSE) and coefficient of determination (R2). The results show that carbon emissions associated with the production of alkali-activated concrete can be predicted with high accuracy using state-of-the-art machine learning techniques. In this study, in which the binders produced by the alkali activation method were evaluated for their usability as a binder material to replace Portland cement, it is concluded that the most successful hyperparameter optimization algorithm for this study is the genetic algorithm (GA) with accurate mean squared error (MSE = 161.17) and coefficient of determination (R2 = 0.90) values in the datasets.Öğe Explainable machine learning model for predicting punching shear strength of FRC flat slabs(2024) Liu, Tongxu; Çakıroğlu, Celal; Islam, Kamrul; Wang, Zhen; Nehdi, Moncef L.Reinforced concrete slabs are vulnerable to punching shear failure at the slab-column joint, which can initiate catastrophic progressive collapse. The addition of steel fibers in the concrete matrix has emerged as an effective strategy to mitigate such progressive failure. However, the effects of the diverse mixture proportions of the concrete matrix with different types and dosages of fibers have made the accurate prediction of the punching shear strength (PSS) of the fiber-reinforced concrete (FRC) flat slabs a complex task, where the existing me chanical models have several limitations. Therefore, this study proposes an explainable XGBoost model for predicting PSS of flat slabs made with different types of FRC based on a newly established comprehensive database of 251 flat slabs including normal strength FRC slabs, high-performance FRC slabs, and ultra-highperformance FRC slabs. A customized procedure was proposed to establish the XGBoost model considering data preparation, feature selection, hyperparameter tuning and model validation. The performance of the XGBoost model was then compared with that of existing mechanical models. Finally, sensitivity analysis and SHapley Additive exPlanations (SHAP) analysis were applied to identify the most influential parameters on the prediction of PSS. Results show that the proposed feature selection method is effective in identifying six influ ential parameters from the eleven parameters related to the PSS of FRC flat slabs. The developed XGBoost model yielded highest prediction accuracy and lowest variation, which outperformed the other mechanical models. Sensitivity analysis also indicated similar trends of parameters in both the XGBoost model and the mechanical models. The PSS of FRC flat slabs can be improved by increasing the concrete compressive strength, reinforce ment ratio, and fiber volume, and by decreasing the column width-to-depth ratio, water-to-binder ratio, and aggregate size ratio. The proposed XGBoost model could enhance the understanding of PSS of FRC flat slabs and guide future pertinent design code provisions.Öğe Cooling load prediction of a double-story terrace house using ensemble learning techniques and genetic programming with SHAP approach ((2024) Çakıroğlu, Celal; Aydın, Yaren; Bekdaş, Gebrail; Işıkdağ, Ümit; Sadeghifam, Aidin Nobahar; Abualigah, LaithSince the cooling systems used in buildings in hot climates account for a significant portion of the energy consumption, it is very important for both economy and environment to accurately predict the cooling load and consider it in building designs. This study aimed to maximize energy efficiency by appropriately selecting the features of a building that affect its cooling load. To this end, data-driven, accurate, and accessible tools were developed that enable the prediction of the cooling load of a building by practitioners. The study involves simulating the energy consumption of a mid-rise, double-story terrace house in Malaysia using building information modeling (BIM) and estimating the cooling load using ensemble machine learning models and genetic programming. Categorical Boosting (CatBoost), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Random Forest (RF) models have been developed and made available as an online interactive graphical user interface on the Streamlit platform. Furthermore, the symbolic regression technique has been utilized to obtain a closed-form equation that predicts the cooling load. The dataset used for training the predictive models comprised 94,310 data points with 10 input variables and the cooling load as the output variable. Performance metrics such as the coefficient of determination (R2), root mean squared error (RMSE), and mean absolute error (MAE) were used to measure the predictive model performances. The results of the machine learning models indicated successful prediction, with the CatBoost model achieving the highest score (R2 = 0.9990) among the four ensemble models and the predictive equation. The SHAP analysis determined the aspect ratio of the building as the most impactful feature of the building.Öğe Explainable Ensemble Learning and Multilayer Perceptron Modeling for Compressive Strength Prediction of Ultra-High-Performance Concrete(2024) Çakıroğlu, Celal; Bekdaş, Gebrail; Geem, Zong Woo; Aydın, Yarenfirst_pagesettingsOrder Article Reprints Open AccessArticle Explainable Ensemble Learning and Multilayer Perceptron Modeling for Compressive Strength Prediction of Ultra-High-Performance Concrete by Yaren Aydın 1ORCID,Celal Cakiroglu 2ORCID,Gebrail Bekdaş 1,*ORCID andZong Woo Geem 3,*ORCID 1 Department of Civil Engineering, Istanbul University-Cerrahpaşa, 34320 Istanbul, Turkey 2 Department of Civil Engineering, Turkish-German University, 34820 Istanbul, Turkey 3 Department of Smart City, Gachon University, Seongnam 13120, Republic of Korea * Authors to whom correspondence should be addressed. Biomimetics 2024, 9(9), 544; https://doi.org/10.3390/biomimetics9090544 Submission received: 27 June 2024 / Revised: 23 August 2024 / Accepted: 5 September 2024 / Published: 9 September 2024 (This article belongs to the Special Issue Bionic Design & Lightweight Engineering) Downloadkeyboard_arrow_down Browse Figures Versions Notes Abstract The performance of ultra-high-performance concrete (UHPC) allows for the design and creation of thinner elements with superior overall durability. The compressive strength of UHPC is a value that can be reached after a certain period of time through a series of tests and cures. However, this value can be estimated by machine-learning methods. In this study, multilayer perceptron (MLP) and Stacking Regressor, an ensemble machine-learning models, is used to predict the compressive strength of high-performance concrete. Then, the ML model’s performance is explained with a feature importance analysis and Shapley additive explanations (SHAPs), and the developed models are interpreted. The effect of using different random splits for the training and test sets has been investigated. It was observed that the stacking regressor, which combined the outputs of Extreme Gradient Boosting (XGBoost), Category Boosting (CatBoost), Light Gradient Boosting Machine (LightGBM), and Extra Trees regressors using random forest as the final estimator, performed significantly better than the MLP regressor. It was shown that the compressive strength was predicted by the stacking regressor with an average R2 score of 0.971 on the test set. On the other hand, the average R2 score of the MLP model was 0.909. The results of the SHAP analysis showed that the age of concrete and the amounts of silica fume, fiber, superplasticizer, cement, aggregate, and water have the greatest impact on the model predictions.Öğe Explainable ensemble learning graphical user interface for predicting rebar bond strength and failure mode in recycled coarse aggregate concrete(Elsevier, 2024) Çakıroğlu, Celal; Tusher, Tanvir Hassan; Shahjalal, Md.; Islam, Kamrul; Billah, A. H. M. Muntasir; Nehdi, Moncef L.Novel study deploys robust machine learning algorithms using newly built comprehensive dataset to predict reinforcing rebar-to-recycled coarse aggregate concrete (RCA) bond strength and failure mode. Prior investigations have solely concentrated on bond strength, resulting in a limited comprehension of the bond failure pattern. Considering the increasing significance of sustainable construction methods, it is crucial to examine both the failure pattern and bond strength to expand the versatility of RCA in various reinforced concrete structures. Accordingly, XGBoost, CatBoost, Random Forest, and LightGBM were trained for this purpose. Model performance was appraised using various statistical metrics, while failure classification performance was assessed using accuracy, recall, and precision indicators. Model performance was ranked using Copeland’s algorithm. Feature importance was quantified using SHAP. Coefficient of determination of 0.91 was achieved by XGBoost in predicting bond strength, outperforming other nine analytical models in literature. Failure mode was predicted with accuracy of 94% by CatBoost, XGBoost, and LightGBM. Embedment length and compressive strength features had greatest influence on bond strength and failure mode, respectively. User-friendly graphical interface was developed to harvest ML models in real-world engineering practice. Online free access accurately assigns to any given combination of input features corresponding accurate rebar bond strength and failure mode.Öğe Optimization of axial load carrying capacity of CFST stub columns(2022) Çakıroğlu, Celal; Bekdaş, GebrailConcrete filled steel tubular (CFST) columns are widely used due to their enhanced mechanical properties. The interaction between the concrete core and the steel casing increases structural stability and magnifies the compressive strength of concrete. Besides the structural performance, in alignment with the commitment of the concrete industry to reduce its environmental impact, lowering the carbon emissions caused by the production of concrete structures is gaining importance in recent years. The current paper gives an overview of the equations available in the literature that predict the axial load carrying capacity of rectangular CFST columns. A modified version of the Jaya metaheuristic algorithm is being proposed and the outcome of this algorithm is being presented. The algorithm is used in order to maximize the axial load-carrying capacity of a stub column. As an optimization constraint the CO2 emission associated with the production of the CFST column is being kept below a predefined level throughout the optimization process. The optimization process as well as the cross-sectional dimensions associated with the optimum solution are presented.Öğe Bir açık ocak örneğinde yamaç molozu akma mekanizmasının belirlenmesi(2023) Yavuz, Enver VuralOn 17 November 2016, 16 workers lost their lives as a result of a landslide in an open pit copper mine in Madenköy district of Siirt province in southeastern Turkey. The main aims of this study is to investigate the causes of the accident, to reveal its mechanism and to be prepared for such mass movements that may occur in the future. The study area consists from the north and getting younger towards the south. Further investigations have shown that hydrogeological conditions and geological structure are the most important factors affecting the units. A large number of boreholes were drilled to reveal the 3D features of the area and cross sections were obtained. The most critical geotechnical numerical cross section that best represents the mass movements was selected and the stability of the area was analysed by finite element method in dry and water saturated conditions using Plaxis 2D and 3D Connect Edition v22 software.Öğe Büyükçekmece İlçesi kentsel dönüşüm çalışmaları kapsamında açığa çıkan inşaat yıkıntı atıklarının beton üretiminde kullanımının araştırılması(2023) Katrancı, Ali Barış; Çelebi, Ramazan; Kutlu, Mehmet; Çakır, Özgür; Sepetçi, Arda; Tarhan, Muhittin; Kılcı, Rıza EvrenWithin the scope of the "Urban Transformation" studies that started in 2012, it is aimed to demolish and reconstruct the buildings in terms of earthquake safety. It is extremely important to recycle and reuse construction and demolition wastes as raw materials without causing environmental problems. In the experimental study, the rubble obtained from the demolished houses as part of urban transformation projects in Büyükçekmece District of Istanbul was selectively separated at the source and crushed with the help of a aggregate crusher. Recycled aggregates of 5.6 – 11.2 mm and 11.2 – 22.4 mm in size were replaced with natural aggregates in different proportions by volume and recycled aggregate concrete series were produced. In the production of concrete mixtures, silica fume and fly ash were used in different proportions as mineral additives. Compressive and splitting tensile strength tests were carried out 28 days after casting. It has been obtained that the use of the recycled aggregates with presence of mineral additives can contribute to both the environment and the economy in terms of sustainability in the construction sector.Öğe Numerical analysis of reinforced soil-retaining wall structures with cohesive and granular backfills(ICE Publishing, 2015) Demirkan, M. M.; Güler, Erol; Hamderi, MuratThe failure mechanisms of reinforced soil segmental walls with extensible reinforcements were studied by performing a numerical analysis using the finite element method. The numerical approach was first verified against the results of three instrumented full-scale structures reported in the literature. Finite element models with different combinations of reinforcement spacing, reinforcement length and backfill soil were analysed. The –c reduction method, which is a special shear strength parameter reduction technique, was applied to simulate the failure conditions. The results of –c reduction analysis were used to evaluate assumptions used in current design procedures for geosynthetic-reinforced soil walls. In particular, shear strains were used to identify failure surfaces. Interpretation of the results indicated that, for both granular and cohesive backfills, the potential failure surface gradually shifts to a direct sliding mode as the system approaches failure. As a result, under working loads the potential failure surface used in current design analysis is correct, but the failure plane of a geosynthetic-reinforced soil-retaining wall at failure approaches a direct sliding type or a bilinear plane, which starts from the toe of the wall with a very shallow slope.Öğe A 3D finite element analysis of the modular block retaining walls with corner turns(Transportation Research Board, 2015) Hamderi, Murat; Güler, ErolThe design manuals for Geosynthetic Reinforced Soil retaining walls include the methodology for various conditions except the case where the wall has curved corner turns. Lately, there has been an increase in the frequency of problems associated with these types of walls. One of the typical problems is cracking/separation of the segmental blocks. The most common method for identifying the cause of problems is a 2D plain-strain analysis which is insufficient for the current case. Therefore in this study, a more robust modeling approach was considered. A 3D finite element (FE) model which is capable of modeling corner turns was created. The main elements of the model are modular blocks, interface layers, soil and geosynthetic reinforcements. As the first step, a base model was defined. The base model included reinforcements with the lowest stiffness. The base model was evaluated in terms of block displacements and stresses. In the other models, reinforcement stiffness and the soil modulus were increased or decreased to evaluate its effect on block displacements and stresses. According to the modeling results, the elastic modulus of the reinforcements and the soil modulus are very effective on block separation and cracking. The separation of blocks could be decreased by increasing the reinforcement stiffness and proper soil compaction. It is considered that the cracking of blocks was related to excessive moments developing in those blocks. The moments are reduced when the reinforcement stiffness was increased. It can be concluded that the cracking of blocks is less likely to happen under reduced moments.Öğe Estimation of horizontal displacements for geosynthetic reinforced soil wall(2023) Güler, Erol; Hamderi, MuratNowadays, the geotechnical design trend is increasingly heading towards the serviceability limit state. However, the current practice in the design of geosynthetic reinforced soil (GRS) walls mostly relies on ultimate limit state. Commonly available design software programs provide typical factor of safety values against various failure modes. With increasing height and variability in GRS walls, the deformation characteristics of the GRS walls also become an important parameter in the design. In this paper, an expression has been developed to predict the horizontal deformation of a GRS wall using a set of data obtained from about sixty-four finite element model configurations. The horizontal deformation expression includes wall height, internal friction angle and elastic modulus of backfill, length, spacing and stiffness of geosynthetic reinforcements. It has been found out that all these parameters contribute to the horizontal displacement of a GRS wall. In addition, some comparisons have been made to reveal the influence of each individual parameter on the horizontal displacement of the wall.Öğe Açısal dönme tabanlı aktif ve pasif toprak itkisi(2023) Hamderi, MuratKonsol istinat duvarlarında oluşan aktif ve pasif itki duvarın dönme ve yer değiştirmesi ile yakından ilgilidir. Klasik zemin mekaniği çerçevesinde yapılan hesaplarda duvarın yeteri kadar döndüğü ve yer değiştirdiği kabul edilerek aktif ve pasif itkilerin kararlı durumdaki değerleri kullanılır. Özellikle pasif itki için kararlı durumdaki değerleri kullanmak hesabın doğruluğunu olumsuz etkilemektedir. Bu çalışma kapsamında aktif ve pasif itkiler için elde edilmiş açısal dönme tabanlı aktif ve pasif itki formülleri tanıtılacaktır. Literatürdeki diğer bazı formüller ve sonlu elemanlar modelinden elde edilen sonuçlar da kullanılarak bir konsol duvar örneği çözülecek, genel bir karşılaştırma yapılacaktır.Öğe Kazıklı radye temellerin oturma tahmini için yeni bir yöntem(2018) Hamderi, MuratLiteratürde kazıklı radye sistemlerin oturması için verilen ampirik formüller oturmayı sadece kaba bir yaklaşıklıkla tahmin edebilmektedir. Kazıklı radye sistemler için hassas oturma tahmini ancak 3-boyutlu sonlu elemanlar yöntemleri ile mümkün olmaktadır. Öte yandan sonlu elemanlar yöntemleri, formül tabanlı yaklaşımlara göre daha karmaşıktır ve bu yöntemlerin yürütülmesi için görece uzun bir süreye ihtiyaç duyulmaktadır. Bu süreyi kısaltmak amacıyla, bu çalışma kapsamında, 3-boyutlu sonlu elemanlar tabanlı bir kazıklı radye temel oturma formülü tanıtılmış ve formül 2 adet vaka çalışmasına uygulanmıştır. Formül, kazık çapını, boyunu, sıklığını, yanal ve uç direncini; radye kalınlığını, yayılı yükü ve 5 adet zemin tabakasının zemin modülünü bünyesine almaktadır. 3-boyutlu sonlu elemanlar analizi kalitesinde sonuç veren kazıklı radye temel formülünün, kazıklı radye temel sistemlerinin optimize edilmesinde pratik bir çözüm sağlayacağı düşünülmektedir.Öğe Kum zeminde p-y eğrileri için yeni bir yöntem(2023) Hamderi, MuratP-y eğrileri yatay kuvvetlere maruz kazıkların zeminin verdiği tepkiyi hesaplamak için kullanılırlar. Birçok p-y eğrisi 1970 yıllarında, petrol şirketlerinin okyanuslar üzerine kurduğu platformları destekleyen kazıkların yanal kapasitelerini hesaplamak için geliştirilmiştir. Bu eğriler kaba veya ince daneli zeminlerde yapılan yanal kazık yükleme deneylerinden elde edilen verilere dayanır. Deneysel verileri barındırması sebebiyle güvenilir ancak çok az sayıda deneye dayandıklarından kaba bir yaklaşıma sahiptirler. Öte yandan günümüzde 3 boyutlu yapı zemin etkileşimini modelleyen yazılımların yaygınlaşması ile yanal kazık kapasitesi ile ilgili bilgisayar ortamında değişik parametreleri içeren sayısız deney yapılabilmektedir. Bu çalışmada, bilgisayar ortamında yapılan kazık yükleme deneyleri sonucunda elde edilen kum zeminler için p-y eğrisi formülü tanıtılacaktır.Öğe New displacement method for free embedded cantilever walls in sand(2024) Hamderi, MuratIn the current literature, there is no practical formula to calculate the horizontal displace ment of cantilever walls. To fill this gap, in the present study, eight formulae for the estimation of wall displacement were developed based on 431 FE wall model configurations. Each formula considers factors such as the wall height, embedment depth, surcharge load, unit weight, internal friction angle, elastic modulus of the surrounding soil, and flexural rigidity of the wall. The FE model, which was used in the development of the formula, was also validated against a physical laboratory study. In addition, the outputs obtained from the formulae were compared with the results of two laboratory studies and a real site study. Finally, a parametric study was performed to estimate the influence of formula input parameters on wall displacement.Öğe Finite element-based p-y curves in sand(Springer, 2023) Hamderi, MuratP-y curves are used in the prediction of non-linear soil resistance resulting from the horizontal pile movement. Many of the p-y curves have been developed in the 1970s for the petroleum industry to predict the horizontal capacity of piles supporting oil platforms. These curves are based on the beam theory fed by strain-gauge data installed on several full-size pile tests. In this particular study, a p-y curve formula for sand was developed using 34 finite element combinations. The p-y curve formula includes the input of soil modulus, soil friction angle, soil unit weight and pile diameter. The results of the formula and the finite element program were validated against the test data available in the literature.Öğe Pilot-scale modeling of colloidal silica delivery to liquefiable sands(Elsevier B.V., 2015) Gallagher, Patricia M.; Hamderi, MuratPassive site stabilization is a developing technology for the in situ mitigation of the risk of liquefaction without surface disruption. It involves the injection of stabilizing materials into liquefiable saturated sand. In this study, a pilot-scale facility (243 cm by 366 cm in plan 122 cm deep) was used to inject a dilute colloidal silica stabilizer into liquefiable sand specimens. The grout advancement was monitored in real time using electrical conductivity cells embedded in the specimens. Injection rates ranging from 65 to 9000 ml/min/well were used to investigate the optimal rate of grout delivery. In tests with low injection rates, the delivery performance was low due to sinking, while at higher injection rates, sinking was less noticeable. After the treatment, the degree of grout penetration was evaluated by excavating the model. The in situ unconfined compressive strength was measured using a pocket penetrometer, and soil blocks were excavated for additional unconfined compressive testing. Moreover, the 3-D flood simulator, UTCHEM, was utilized to simulate the experimental results and to predict the injection rates for adequate stabilizer delivery. The results of the strength testing demonstrated that as little as 1% by weight of the colloidal silica provides a significant improvement in strength after a month of curing. The study also revealed the feasibility of delivering colloidal silica to liquefiable sands by implementing a large-scale treatment.Öğe Optimal dimensions of post-tensioned concrete cylindrical walls using harmony search and snsemble learning with SHAP(2023) Çakıroğlu, Celal; Bektaş, Gebrail; Geem, Zong Woo; Kim, SanghunThe optimal design of prestressed concrete cylindrical walls is greatly beneficial for economic and environmental impact. However, the lack of the available big enough datasets for the training of robust machine learning models is one of the factors that prevents wide adoption of machine learning techniques in structural design. The current study demonstrates the application of the well-established harmony search methodology to create a large database of optimal design configurations. The unit costs of concrete and steel used in the construction, the specific weight of the stored fluid, and the height of the cylindrical wall are the input variables whereas the optimum thicknesses of the wall with and without post-tensioning are the output variables. Based on this database, some of the most efficient ensemble learning techniques like the Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Categorical Gradient Boosting (CatBoost) and Random Forest algorithms have been trained. An R2 score greater than 0.98 could be achieved by all of the ensemble learning models. Furthermore, the impacts of different input features on the predictions of different machine learning models have been analyzed using the SHapley Additive exPlanations (SHAP) methodology. The height of the cylindrical wall was found to have the greatest impact on the optimal wall thickness, followed by the specific weight of the stored fluid. Also, with the help of individual conditional expectation (ICE) plots the variations of predictive model outputs with respect to each input feature have been visualized. By using the genetic programming methodology, predictive equations have been obtained for the optimal wall thickness.Öğe Predictive modeling of recycled aggregate concrete beam shear strength using explainable ensemble learning methods(2023) Çakıroğlu, Celal; Bekdaş, GebrailConstruction and demolition waste (CDW) together with the pollution caused by the production of new concrete are increasingly becoming a burden on the environment. An appealing strategy from both an ecological and a financial point of view is to use construction and demolition waste in the production of recycled aggregate concrete (RAC). However, past studies have shown that the currently available code provisions can be unconservative in their predictions of the shear strength of RAC beams. The current study develops accurate predictive models for the shear strength of RAC beams based on a dataset of experimental results collected from the literature. The experimental database used in this study consists of full-scale four-point flexural tests. The recycled coarse aggregate (RCA) percentage, compressive strength (f 0 c ), effective depth (d), width of the cross-section (b), ratio of shear span to effective depth (a/d), and ratio of longitudinal reinforcement (?w) are the input features used in the model training. It is demonstrated that the proposed machine learning models outperform the existing code equations in the prediction of shear strength. State-of-the-art metrics of accuracy, such as the coefficient of determination (R 2 ), mean absolute error, and root mean squared error, have been utilized to quantify the performances of the ensemble machine learning models. The most accurate predictions could be obtained from the XGBoost model, with an R 2 score of 0.94 on the test set. Moreover, the impact of different input features on the machine learning model predictions is explained using the SHAP algorithm. Using individual conditional expectation plots, the variation of the model predictions with respect to different input features has been visualized.Öğe Data-driven ensemble learning approach for optimal design of cantilever soldier pile retaining walls(Elsevier, 2023) Çakıroğlu, Celal; Islam, Kamrul; Bektaş, Gebrail; Nehdi, Moncef L.Cantilever soldier pile retaining walls are used to ensure the stability of excavations. This paper deploys ensemble machine learning algorithms towards achieving optimum design of these structures. A large dataset was developed consisting of 40,569 combinations of pile geometry, external loading, soil properties, and concrete unit cost, with two different values of soil reaction coefficient. Optimum pile diameter that minimizes the total cost of the retaining wall was computed considering the structural load-carrying capacity as the optimization constraint. The dataset was split into training and testing sets at 70% to 30% ratio. The predictive accuracy of the ensemble machine learning models was appraised on the testing dataset using various statistical metrics. Model performance was also evaluated for its ability in predicting the optimum pile diameter. The developed models demonstrated excellent predictive accuracy. Furthermore, the effect of different input variables on the model predictions was explained using the SHapely Additive exPlanations (SHAP) approach. Through the SHAP algorithm, the pile length was identified as the design variable having the most significant effect on the optimum pile diameter. The study demonstrates ensemble learning techniques as a viable alternative to the traditional techniques in the optimum design of cantilever soldier pile retaining walls.
