Search

Abstract Concrete compressive strength (CS) is a crucial performance parameter in concrete structure design. Reliable strength prediction reduces costs and time in design and prevents material waste from extensive mixture trials. Machine learning techniques solve structural engineering challenges such as CS prediction. This study used Machine Learning (ML) models to enhance the prediction of CS, analyzing 1030 experimental CS data ranging from 2.33 to 82.60 MPa from previous research databases. The ML models included both non-ensemble and ensemble types. The non-ensemble models were regression-based, evolutionary, neural network, and fuzzy-inference-system. Meanwhile, the ensemble models consisted of adaptive boosting, random forest, and gradient boosting. There were eight input parameters: cement, blast-furnace-slag, aggregates (coarse and fine), fly ash, water, superplasticizer, and curing days, with the CS as the output. Comprehensive performance evaluations include visual and quantitative methods and k-fold cross-validation to assess the study’s reliability and accuracy. A sensitivity analysis using Shapley-Additive-exPlanations (SHAP) was conducted to understand better how each input variable affects CS. The findings showed that the Categorical-Gradient-Boosting (CatBoost) model was the most accurate prediction during the testing stage. It had the highest determination-coefficient (R2) of 0.966 and the lowest Root-Mean-Square-Error (RMSE) of 3.06 MPa. The SHAP analysis showed that the age of the concrete was the most critical factor in the predictive accuracy. Finally, a Graphical User Interface (GUI) was offered for designers to predict concrete CS quickly and economically instead of costly computational or experimental tests.

Reliable information plays a pivotal role in sustainable urban planning. With advancements in computer technology, geoinformatics tools enable accurate identification of land use and land cover (LULC) in both spatial and temporal dimensions. Given the need for precise information to enhance decision-making, it is imperative to assess the performance and reliability of classification algorithms in detecting LULC changes. While research on the application of machine learning algorithms in LULC evaluation is widespread in many countries, it remains limited in Zambia and Sri Lanka. Hence, we aimed to assess the reliability and performance of support vector machine (SVM), random forest (RF), and artificial neural network (ANN) algorithms for detecting changes in land use and land cover taking Lusaka and Colombo City as the study area from 1995 to 2023 using Landsat Thematic Mapper (TM), and Operational Land Imager (OLI). The results reveal that the RF and ANN models exhibited superior performance, both achieving Mean Overall Accuracy (MOA) of 96% for Colombo and 96% and 94% for Lusaka, respectively. Meanwhile, the SVM model yielded Overall Accuracy (OA) ranging between 77% and 94% for the years 1995 and 2023. Further, RF algorithm notably produced slightly higher OA and kappa coefficients, ranging between 0.92 and 0.97, when compared to both the ANN and SVM models, across both study areas. A predominant land use change was observed as the expansion of vegetation by 11,990 ha (60.4%), primarily through the conversion of 1,926 ha of bare lands into vegetation in Lusaka during 1995-2005. However, a noteworthy shift was observed as built-up areas experienced significant growth from 2005 to 2023, with a total increase of 25,110 ha (71%). However, despite the conversion of vegetation to built-up areas during the entire period from 1995 to 2023, there was still a net gain of over 11,000 ha (53.4%) in vegetation cover. In case of Colombo, built-up areas expanded by 1,779 ha (81.5%), while vegetation land decreased by 1,519 ha (62.3%) during concerned period. LULC simulation also indicated a 160-ha expansion of built-up areas during the 2023-2035 period in Lusaka. Likewise, Colombo saw a rise in built-up areas by 337 ha within the same period. Overall, the RF algorithm outperformed the ANN and SVM algorithms. Additionally, the prediction and simulation results indicate an upward trend in built-up areas in both scenarios. The resultant land cover maps provide a crucial baseline that will be invaluable for urban planning and policy development agencies in both countries. [ABSTRACT FROM AUTHOR]

Concrete compressive strength (CS) is a crucial performance parameter in concrete structure design. Reliable strength prediction reduces costs and time in design and prevents material waste from extensive mixture trials. Machine learning techniques solve structural engineering challenges such as CS prediction. This study used Machine Learning (ML) models to enhance the prediction of CS, analyzing 1030 experimental CS data ranging from 2.33 to 82.60 MPa from previous research databases. The ML models included both non-ensemble and ensemble types. The non-ensemble models were regression-based, evolutionary, neural network, and fuzzy-inference-system. Meanwhile, the ensemble models consisted of adaptive boosting, random forest, and gradient boosting. There were eight input parameters: cement, blast-furnace-slag, aggregates (coarse and fine), fly ash, water, superplasticizer, and curing days, with the CS as the output. Comprehensive performance evaluations include visual and quantitative methods and k-fold cross-validation to assess the study's reliability and accuracy. A sensitivity analysis using Shapley-Additive-exPlanations (SHAP) was conducted to understand better how each input variable affects CS. The findings showed that the Categorical-Gradient-Boosting (CatBoost) model was the most accurate prediction during the testing stage. It had the highest determination-coefficient (R2) of 0.966 and the lowest Root-Mean-Square-Error (RMSE) of 3.06 MPa. The SHAP analysis showed that the age of the concrete was the most critical factor in the predictive accuracy. Finally, a Graphical User Interface (GUI) was offered for designers to predict concrete CS quickly and economically instead of costly computational or experimental tests. [ABSTRACT FROM AUTHOR]

The Igla Ahmr region in the Central Eastern Desert (CED) of Egypt comprises mainly syenogranites and alkali feldspar granites, with a few tonalite xenoliths. The mineral potential maps were presented in order to convert the concentrations of total rare earth elements (REEs) and associated elements such as Zr, Nb, Ga, Y, Sc, Ta, Mo, U, and Th into mappable exploration criteria based on the line density, five alteration indices, random forest (RF) machine learning, and the weighted sum model (WSM). According to petrography and geochemical analysis, random forest (RF) gives the best result and represents new locations for rare metal mineralization compared with the WSM. The studied tonalites resemble I-type granites and were crystallized from mantle-derived magmas that were contaminated by crustal materials via assimilation, while the alkali feldspar granites and syenogranites are peraluminous A-type granites. The tonalites are the old phase and are considered a transitional stage from I-type to A-type, whereas the A-type granites have evolved from the I-type ones. Their calculated zircon saturation temperature TZr ranges from 717 °C to 820 °C at pressure < 4 kbar and depth < 14 km in relatively oxidized conditions. The A-type granites have high SiO2 (71.46–77.22 wt.%), high total alkali (up to 9 wt.%), Zr (up to 482 ppm), FeOt/(FeOt + MgO) ratios > 0.86, A/CNK ratios > 1, Al2O3 + CaO < 15 wt.%, and high ΣREEs (230 ppm), but low CaO and MgO and negative Eu anomalies (Eu/Eu* = 0.24–0.43). These chemical features resemble those of post-collisional rare metal A-type granites in the Arabian-Nubian Shield (ANS). The parent magma of these A-type granites was possibly derived from the partial melting of the I-type tonalitic protolith during lithospheric delamination, followed by severe fractional crystallization in the upper crust in the post-collisional setting. Their rare metal-bearing minerals, including zircon, apatite, titanite, and rutile, are of magmatic origin, while allanite, xenotime, parisite, and betafite are hydrothermal in origin. The rare metal mineralization in the Igla Ahmr granites is possibly attributed to: (1) essential components of both parental peraluminous melts and magmatic-emanated fluids that have caused metasomatism, leading to rare metal enrichment in the Igla Ahmr granites during the interaction between rocks and fluids, and (2) structural control of rare metals by the major NW–SE structures (Najd trend) and conjugate N–S and NE–SW faults, which all are channels for hydrothermal fluids that in turn have led to hydrothermal alteration. This explains why rare metal mineralization in granites is affected by hydrothermal alteration, including silicification, phyllic alteration, sericitization, kaolinitization, and chloritization. [ABSTRACT FROM AUTHOR]

Water availability is a key factor in territorial sustainable development. Moreover, groundwater constitutes the survival element of human life and ecosystems in arid oasis areas. Therefore, groundwater potential (GWP) identification represents a crucial step for its management and sustainable development. This study aimed to map the GWP using ten algorithms, i.e., shallow models comprising: multilayer perceptron, k-nearest neighbor, decision tree, and support vector machine algorithms; hybrid models comprising: voting, random forest, adaptive boosting, gradient boosting (GraB), and extreme gradient boosting; and the deep learning neural network. The GWP inventory map was prepared using 884 binary data, with "1" indicating a high GWP and "0" indicating an extremely low GWP. Twenty-three GWP-influencing factors have been classified into numerical data using the frequency ration method. Afterwards, they were selected based on their importance and multi-collinearity tests. The predicted GWP maps show that, on average, only 11% of the total area was predicted as a very high GWP zone and 17% and 51% were estimated as low and very low GWP zones, respectively. The performance analyses demonstrate that the applied algorithms have satisfied the validation standards for both training and validation tests with an average area under curve of 0.89 for the receiver operating characteristic. Furthermore, the models' prioritization has selected the GraB model as the outperforming algorithm for GWP mapping. This study provides decision support tools for sustainable development in an oasis area. [ABSTRACT FROM AUTHOR]

Increasing agricultural production is a major concern that aims to increase income, reduce hunger, and improve other measures of well-being. Recently, the prediction of soil-suitability has become a primary topic of rising concern among academics, policymakers, and socio-economic analysts to assess dynamics of the agricultural production. This work aims to use physico-chemical and remotely sensed phenological parameters to produce soil-suitability maps (SSM) based on Machine Learning (ML) Algorithms in a semi-arid and arid region. Towards this goal an inventory of 238 suitability points has been carried out in addition to14 physico-chemical and 4 phenological parameters that have been used as inputs of machine-learning approaches which are five MLA prediction, namely RF, XgbTree, ANN, KNN and SVM. The results showed that phenological parameters were found to be the most influential in soil-suitability prediction. The validation of the Receiver Operating Characteristics (ROC) curve approach indicates an area under the curve and an AUC of more than 0.82 for all models. The best results were obtained using the XgbTree with an AUC = 0.97 in comparison to other MLA. Our findings demonstrate an excellent ability for ML models to predict the soil-suitability using physico-chemical and phenological parameters. The approach developed to map the soil-suitability is a valuable tool for sustainable agricultural development, and it can play an effective role in ensuring food security and conducting a land agriculture assessment. [ABSTRACT FROM AUTHOR]

In the pursuit of understanding surface water quality for sustainable urban management, we created a machine learning modeling framework that utilized Random Forest (RF), Cubist, Extreme Gradient Boosting (XGB), Multivariate Adaptive Regression Splines (MARS), Gradient Boosting Machine (GBM), Support Vector Machine (SVM), and their hybrid stacking ensemble RF (SE-RF), as well as stacking Cubist (SE-Cubist), to predict the distribution of water quality in the Howrah Municipal Corporation (HMC) area in West Bengal, India. Additionally, we employed the ReliefF and Shapley Additive exPlanations (SHAP) methods to elucidate the underlying factors driving water quality. We first estimated the water quality index (WQI) to model seven water quality parameters: total hardness (TH), pH, total dissolved solids (TDS), dissolved oxygen (DO), biochemical oxygen demand (BOD), calcium (Ca), magnesium (Mg). Then six independent factors were utilized (i.e. Precipitation (Pr), Maximum Temperature (Tmax), Minimum Temperature (Tmin), Normalized Difference Turbidity Index (NDTI), Normalized Difference Chlorophyll Index (NDCI), and Total Dissolved Solids (TDS)) for predicting the WQI mapping through the different ML models. This study demonstrated that the SE-Cubist model outperforms other ML models. During the testing phase, it achieved the best modeling results with an R2 = 0.975, RMSE = 0.351, and MAE = 0.197. The ReliefF and SHAP analyses identified Pr and Tmax as the most significant factors influencing WQI within the study area. [Display omitted] • Different machine learning (ML) models were applied to predict urban surface water quality. • ReliefF and SHAP analyses identified Pr and Tmax as the most significant factors influencing water quality. • The SE-Cubist model showed superior prediction results and SVM was the worst model. • The hybrid stacking model demonstrated powerful performance prediction models. [ABSTRACT FROM AUTHOR]