Pouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101The Optimization of Mix Proportion of Hot Mix Asphalt for Sustainable Flexible Pavements: Experimental Study and Grey Taguchi Relational Analysis11915882710.22115/scce.2022.343508.1447ENSamanasa Krishna RaoProfessor, Civil Engineering Department, Vignana Bharathi Institute of Technology, Hyderabad, Telangana, IndiaP. SravanaProfessor, Civil Engineering Department, JNTUH College of Engineering, Hyderabad, Telangana, IndiaJournal Article20220521Most of the Indian black topped roads have been damaged due to adverse weather and heavy load distresses. Many researchers have focused on improving durability of Hot Mix Asphalt (HMA) pavements. The factors influencing durability of HMA are Binder content, Combined aggregate gradation, type of Filler and addition of Fiber. In order to optimize the combination of variables used in HMA mix design, Grey relational analysis by using Taguchi technique is used, where many parameters can be analyzed at a time with more accuracy. Multiple performance measurements like Stability, Flow, Bulk specific gravity of the mix (<em>G<sub>mb</sub></em>), Theoretical maximum specific gravity of the mix (<em>G<sub>mm</sub></em>), Voids in Mineral Aggregate (<em>VMA</em>), Air voids (<em>V<sub>a</sub></em>) and Voids Filled with Bitumen (<em>VFB</em>) are considered. Bituminous Concrete (BC) mix was optimized using L9 Orthogonal array considering four parameters such as Fiber content, Filler combination, Binder content and Combined Aggregate Gradation, with three levels having seven performance measurements. The most significant parameter and percent contribution of each parameter of BC mix are analyzed by Analysis of Variance (ANOVA) using Grey Taguchi technique. The analysis was done by considering two Gradation conditions (Coarse gradation and Fine gradation) based on voids in mix. From the analysis, it can be concluded that all parameters are significant except bitumen content for optimizing BC mix.https://www.jsoftcivil.com/article_158827_d0983e831634cafe708c211de2a2d4a8.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Models Development for Asphalt Pavement Performance Index in Different Climate Regions Using Soft Computing Techniques204216079810.22115/scce.2022.357135.1512ENAbdualmtalab AbdualazizAliPh.D. Candidate, Department of Civil Engineering, Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, Newfoundland, CanadaLecturer, Department of Civil Engineering, Faculty of Engineering, Azzaytuna University, Tarhuna, Libya0000-0002-4450-6607Usama HeneashAssistant Professor, Department of Civil Engineering, Faculty of Engineering, Kafr El Sheikh University, Kafr El Sheikh, EgyptAmgad HusseinAssociate Professor, Department of Civil Engineering, Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, Newfoundland, CanadaShaban Ismael AlbrkaAliAssociate Professor, Department of Civil Engineering, Faculty of Engineering, Near East University, North CyprusShahbaz KhanPostdoctoral Associate, Department of Civil Engineering, Faculty of Engineering, University of Florida, United StatesJournal Article20220822The Pavement Condition Index (PCI) is one of the most critical pavement performance indicators and ride quality. This study aims to develop PCI models based on pavement distress parameters using conventional technique and artificial neural network (ANN) technique across two climate regions in the U.S. and Canada. The long-term pavement performance (LTPP) database was used to obtain pavement distress data, including pavement age, rutting, fatigue cracking, block cracking, longitudinal cracking, transverse cracking, potholes, patching, bleeding, and ravelling, as input variables for predicting PCI. Forty-three flexible pavement segments were considered with 333 observations. The type, severity, and extent of surface damage and the PCI were determined for each pavement segment. Two modelling techniques were conducted to predict the PCI, namely, multiple linear regression (MLR) and artificial neural network (ANN). The coefficient of determination (R<sup>2</sup>), Root mean squared error (RMSE), and mean absolute error (MAE) were used to examine the performance of the two techniques adopted in this study. The models' results determined that both ANN and MLR models could predict PCI with high accuracy; ANN models were more accurate and efficient. ANN provided the highest accuracy in predicting PCI of pavement for wet and wet no-freeze climate regions, with R<sup>2</sup> values of 99.8%, 98.3 %: RMSE values of 0.44%, 1.413%, and MAE values of 0.44%, 1.022%, respectively. Whereas in the MLR method, R<sup>2 </sup>values of 86.8% and 89.4%: RMSE values of 7.195%, 7.324%, and MAE values of 5.616%, 5.79% for wet and wet no freeze climate regions, respectively.https://www.jsoftcivil.com/article_160798_76b4177491ef3e598d360c6c3d76f56c.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Performance Based Review and Fine-Tuning of TRM-Concrete Bond Strength Existing Models435516079910.22115/scce.2022.349483.1476ENHashem JahangirAssistant Professor, Department of Civil Engineering, University of Birjand, Birjand, Iran0000-0003-3099-4045Zahra NikkhahM.Sc. Student, Department of Civil Engineering, Ferdowsi University of Mashhad, Mashhad, Iran0000-0002-5849-3210Danial Rezazadeh EidgaheeDepartment of Civil Engineering, Ferdowsi University of Mashhad, Mashhad, Iran0000-0001-8023-2896Mohammad Reza EsfahaniProfessor of Structural Engineering, Department of Civil Engineering, Ferdowsi University of Mashhad, Mashhad, Iran0000-0002-2655-2466Journal Article20220628Textile reinforced mortars (TRMs) are new composite materials which were considered as a proper alternative for fiber reinforced polymers (FRPs) to strengthen various structural elements. In comparison to FRPs, the TRMs have more fire resistance, more environmental consistency and are safer the structural elements because of their better bond to substrate and various failure modes. There are a lot of existing models to calculate the bond strength between TRMs and concrete substrate. But, most of them originated from the FRP-concrete bond models and are not accurate enough to estimate the TRM-concrete bond strength. In this paper, new TRM-concrete bond models were calibrated to predict the bond strength between various TRM composites and the concrete substrate. To achieve this goal, a database including 221 experimental direct shear tests were compiled and a simple existing model was selected to be calibrated via soft computing techniques. It was found that the presented novel models could be accurately utilized to anticipate the TRM-concrete bond strength with various types of fibers and different geometrical features with R value of 0.6909 and NMAE error value of 12.62%.https://www.jsoftcivil.com/article_160799_61f58dc178858db7588e9afb76705a5a.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Predicting The Strength Properties of Self Healing Concrete Using Artificial Neural Network567116371110.22115/scce.2022.349792.1478ENA. Ravi ThejaAssistant Professor in SVR ENGINEERING COLLEGE, Nandyal, Andhra Pradesh, IndiaM. Srinivasula ReddyAssociate Professor of Civil Engineering, G Pulla Reddy Engineering College, Kurnool, AP, IndiaBharat Bhushan JindalAssistant Professor in School of Civil Engineering, Shri Mata Vaishno Devi University, Jammu and Kashmir, IndiaC. SashidharProfessor of Civil Engineering, JNTU College of Engineering, JNTUA, Ananthapuramu, Andhra Pradesh, IndiaJournal Article20220630An extensive simulation program is used in this study to discover the best ANN model for predicting the compressive strength of concrete with respect to the percentage of mineral admixture and percentage of crystalline admixture. To accomplish this, an experimental database of 100 samples is compiled from the literature and utilized to find the best ANN architecture. The main aim of this paper was to predict the strength properties of self-healing concrete (SHC) with crystalline admixture and different mineral admixtures using an artificial neural network (ANN). The samples, 100 in Number, with different mixes, were analyzed after 28 days of curing of the samples. ANN was fed with the experimental data containing four input parameters: mineral admixture (MA), percentage of mineral admixture (PMA), Percentage of crystalline admixture (PCA), and type of exposure (TE). Correspondingly, strength (Fc) was the output parameter. The experimental data showed a good correlation with the values predicted by ANN. In conclusion, ANN could be used to accurately evaluate SHC strength characteristics.https://www.jsoftcivil.com/article_163711_60ed8d5f430e1a5abdd691803919e519.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Optimal Design of MR Dampers Using NSGA-II Algorithm729216371010.22115/scce.2022.347247.1466ENMehdi BabaeiAssistant Professor, Department of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, Iran0000-0002-2597-9767Nastaran TaghaddosiDepartment of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, IranNavid SerajiDepartment of Civil Engineering, Faculty of Engineering, University of Zanjan, Zanjan, Iran0000-0002-9065-9815Journal Article20220614In recent years, new ideas and solutions have been proposed by scientists and researchers to control the response of structures against seismic excitations. In this article, The Optimization of semi-active control systems using MR dampers has been studied to reduce the structure's response under earthquake forces. For this purpose, three frames of five, eight, and eleven stories have been examined as numerical examples. A multi-objective optimization approach based on the NSGA-II algorithm is used to control the response of structures and the fuzzy logic algorithm is used to determine the force of these dampers. The values of maximum displacement, acceleration, and inter-story drift of the top floor have been selected as objective functions. The position of dampers has been optimized to obtain optimal practical solutions. The results show that the responses are significantly reduced when using a semi-active MR damper and the arrangement of the dampers has a great impact on the amount of this reduction.https://www.jsoftcivil.com/article_163710_112535483b91da371af70358005e8800.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Air Quality Prediction - A Study Using Neural Network Based Approach9311316370910.22115/scce.2022.352017.1488ENRaunaq SinghSuriSchool of Planning & Architecture, Bhopal, Madhya Pradesh, 462030, IndiaAjay KumarJainNational Institute of Technical Teachers’ Training and Research, Bhopal, Madhya Pradesh, 462002, IndiaNishant RajKapoorAcademy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, India0000-0002-4721-3332Aman KumarAcademy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, IndiaStructural Engineering Department, CSIR—Central Building Research Institute, Roorkee 247667, India0000-0002-4718-3903Harish ChandraAroraAcademy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, IndiaStructural Engineering Department, CSIR—Central Building Research Institute, Roorkee 247667, India0000-0001-6921-999XKrishna KumarDepartment of Hydro and Renewable Energy, Indian Institute of Technology, Roorkee 247667, IndiaHashem JahangirDepartment of Civil Engineering, University of Birjand, Birjand, 9717434765, Iran0000-0003-3099-4045Journal Article20220716India is the 7<sup>th</sup> largest country by area and 2<sup>nd</sup> most populated country in the world. The reports prepared by IQAir revels that India is 3<sup>rd</sup> most polluted country after Bangladesh and Pakistan, on the basis of fine particulates (PM<sub>2.5</sub>) concentration for the year 2020. In this article, the quality of air in six Indian cities is predicted using data-driven Artificial Neural Network. The data was taken from the 'Kaggle' online source. For six Indian cities, 6139 data sets for ten contaminants (PM<sub>2.5</sub>, PM<sub>10</sub>, NO, NO<sub>2</sub>, NH<sub>3</sub>, CO, SO<sub>2</sub>, O<sub>3</sub>, C<sub>6</sub>H<sub>6</sub> and C<sub>7</sub>H<sub>8</sub>) were chosen. The datasets were collected throughout the last five years, from 2016 to 2020, and were used to develop the predictive model. Two machine learning model are proposing in this study namely Artificial Intelligence (AI) and Gaussian Process Regression (GPR) The R-value of ANN and GPR models are 0.9611 and 0.9843 sequentially. The other performance indices such as RMSE, MAPE, MAE of the GPR model are 21.4079, 7.8945% and 13.5884, respectively. The developed model is quite useful to update citizens about the predicted air quality of the urban spaces and protect them from getting affected by the poor ambient air quality. It can also be used to find the proper abatement strategies as well as operational measures.https://www.jsoftcivil.com/article_163709_13249de4f8ae15930a44a4feaa08c517.pdfPouyan PressJournal of Soft Computing in Civil Engineering2588-28727120230101Deep Neural Network Regression with Advanced Training Algorithms for Estimating the Compressive Strength of Manufactured-Sand Concrete11413416370810.22115/scce.2022.349837.1485ENNhat-Duc HoangLecturer, Institute of Research and Development, Duy Tan University, Da Nang, 550000, VietnamLecturer, Faculty of Civil Engineering, Duy Tan University, Da Nang, 550000, VietnamVan-Duc TranLecturer, Faculty of Civil Engineering, Duy Tan University, Da Nang, 550000, VietnamLecturer, International School - Duy Tan University, Da Nang, 550000, VietnamJournal Article20220709Manufactured sand has high potential for replacing natural sand and reducing the negative impact of the construction industry on the environment. This paper aims at developing a novel deep learning-based approach for estimating the compressive strength of manufactured-sand concrete. The deep neural networks are trained by the advanced optimizers of Root Mean Squared Propagation, Adaptive Moment Estimation, and Adaptive Moment Estimation with Nesterov momentum (Nadam). In addition, the activation functions of logistic sigmoid, hyperbolic tangent sigmoid, and rectified linear unit activation are employed. A dataset including 132 samples has been used to train and verify the deep neural networks. Stone powder content, sand ratio, quantity of cement, quantity of water, quantity of coarse aggregate, quantity of water-reducer, quantity of manufactured sand, concrete slump, unit weight of concrete, and curing age are utilized as predictor variables. Based on experiments, the Nadam-optimized model used with the sigmoid activation function has achieved the most desired performance with root mean square error (RMSE) = 1.95, mean absolute percentage error (MAPE) = 3.04%, and coefficient of determination (<em>R</em><sup>2</sup>) = 0.97. Thus, this neural computing model is recommended for practical purposes because it can help to mitigate the time and cost dedicated to laboratory work.https://www.jsoftcivil.com/article_163708_91c6fe6dfd6d00aa41ee12f22dcd4de2.pdf