Amir Hosseini
Civil Engineer
M.Sc of Construction management at Ferdowsi University of Mashhad (FUM)
Civli and Facade Engineer
Download CV
Send Message
Contact Info.
Amir Hosseini (Full name: Amirhossein Hosseini Sarchashmeh) holds a Bachelor’s degree in Civil Engineering and a Master’s degree in Construction Engineering and Management from Ferdowsi University of Mashhad. His research has focused on the application of machine learning algorithms in developing green concrete, with an emphasis on environmental issues and sustainable development principles. Solutions aimed at reducing the environmental impact of the construction industry through innovative technologies have been proposed in his work. Additionally, he has accumulated extensive professional experience in various civil engineering and construction management projects, gaining valuable expertise in project execution and management. Committed to advancing knowledge and practice in civil engineering, he consistently seeks opportunities for innovation and improvement in construction processes, prioritizing sustainability and efficiency.
Contact Info:
Academic Emails:
hosseini.s1@um.ac.ir
Gmail:
Education
Msc Construction Engineering and Management
Thesis Title:
Prediction Compressive Strength of Recycled Aggregate Concrete Using Machine Learning Algorithms
Supervisor:
Dr. Mansour Ghalehnovi
Abstract:
Concrete, as the second most widely used material in the construction industry worldwide, consumes a significant amount of natural aggregates extracted from natural resources such as mountains, riverbeds, and lakes each year. The production of this volume of aggregates leads to the depletion of natural resources and the generation of greenhouse gases.
Meanwhile, the demolition of structures at the end of their useful life produces a massive amount of construction and demolition waste, complicating the process of managing such large volumes of materials. Therefore, in recent years, the use of recycled aggregates in new concrete has emerged as an innovative approach to sustainable construction. This study aims to create a machine learning model to predict the compressive strength of concrete made with recycled aggregates using six models: Random Forest, Decision Tree, LightGBM, Adaboost, Gradient Boosting, and XGBoost. For this purpose, 225 experimental datas collected from previous research were utilized for training and testing the models. Additionally, two methods random search and Bayesian optimization were employed to tune the optimal hyperparameters of the models. The results indicated that the Gradient Boosting algorithm exhibited the best performance in predicting the compressive strength of recycled aggregate concrete, with evaluation metrics (R²= 0.954, MAE= 2.433 MPa, and RMSE= 3.116 MPa). In contrast, the Adaboost algorithm demonstrated the weakest performance in predictions. Sensitivity analysis and parametric analysis were conducted to assess the importance of input features on the compressive strength of recycled aggregate concrete and their varying impacts. The results revealed that the characteristics of superplasticizer dosage, Maximum Aggregate Size, and Effective Water to Cement Ratio are the most influential factors in estimating the compressive strength of recycled aggregate concrete. A positive correlation was observed between changes in superplasticizer dosage and the compressive strength of recycled aggregate concrete, while a negative correlation was noted between both the Maximum Aggregate Size and the Effective Water to Cement Ratio with the output parameter. Ultimately, this study seeks to develop a comprehensive machine learning model for estimating the compressive strength of recycled aggregate concrete, considering the complex relationships among influential features, thus contributing to facilitating sustainable construction.
September 2021 – August 2023
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.
Bsc Civil Engineering
September 2017- August 2021
Facade Engineer
Amir Hosseini has accumulated 2 years of hands-on experience. Joining the engineering team in 2022, Amir swiftly established himself as a crucial member of the Building Envelope team. His primary responsibilities encompass wind load calculations, Finite Element modelling, and the structural design of aluminum, glass, and steel components. Amir’s adeptness in ensuring structural integrity and safety is evident in his wind load calculations, while his expertise in Finite Element modeling contributes to the development of robust structural designs. (Link)
September 2022 – August 2024
Publications
Title: Hyperparameters’ role in machine learning algorithm for modeling of compressive strength of recycled aggregate concrete
Authors: Amirhossein Hosseini Sarcheshmeh, Hossein Etemadfard, Alireza Najmoddin & Mansour Ghalehnovi
Abstract:
RAC is a kind of concrete made from Recycled Concrete Aggregates instead of natural aggregates. The use of RAC has been popular in recent years due to the environmental benefits of reducing waste and preserving natural resources. However, one of the RAC-using challenges is accurately predicting its compressive strength. This is a crucial factor in determining its suitability for various structural applications. In this research, eight ML algorithms were trained using a dataset of RAC samples to predict their compressive strength. They were Random Forest, support vector regression (SVR), K nearest neighbors (KNN), light gradient boosting machine (Light GBM), adaptive boosting (Adaboost), gradient boosting, extreme gradient boosting (XGBoost), and multi-layer perceptron (MLP). The best hyperparameters for each algorithm obtained using different hyperparameter tuning methods include Grid Search, Random Search, Successive Halving, Bayesian Optimization with Gaussian Processes (BOGP), Bayesian Optimization Random Forest (BORF), and Bayesian optimization gradient boosted trees (BOGB). The study’s findings indicated that Gradient Boosting has the highest performance in predicting the compressive strength of RAC after applying the hyperparameter tuning methods, with R2 and RMSE equal to 0.86 and 5.46 MPa, respectively. In addition, a sensitivity analysis was performed to determine the effect of various input parameters on the compressive strength of RAC. This indicated that the Effective Water-Cement ratio and the RAC Nominal maximum size had the most significant effect. The results show the potential of machine learning algorithms to predict the compressive strength of RAC, which can contribute to the development of more sustainable building materials.
Title: Multi-output machine learning for predicting the mechanical properties of BFRC
Authors: Alireza Najmoddin, Hossein Etemadfard, Amirhossein Hosseini.S, Mansour Ghalehnovi
Abstract:
This investigation delves into the mechanical characteristics of Basalt Fiber Reinforced Concrete (BFRC), with a specific focus on compressive, flexural, and splitting tensile strengths. Employing a Multi-Output approach, six Machine Learning (ML) algorithms, namely Adaptive Boosting (AdaBoost), Light Gradient-Boosting Machine (LightGBM), Gradient Boosting, Extreme Gradient Boosting (XGBoost), K-Nearest Neighbors (KNN), and Random Forest, were used to predict the three output variables concurrently. The SHapley Additive exPlanations method facilitated sensitivity analysis, identifying influential factors, while Partial Dependence Plots (PDP) enhanced the comprehension of input impacts on the output values. The study revealed that the XGBoost algorithm exhibited superior performance, achieving an impressive R-squared value of 0.94 in predicting BFRC mechanical properties. Key parameters affecting compressive, flexural, and tensile strengths were pinpointed, emphasizing the critical roles of the water-to-cement ratio and coarse aggregates. PDP diagrams further unveiled optimal parameter ranges. The innovation of this research lies in its simultaneous prediction of multiple outputs, an approach that enhances the comprehensive assessment of BFRC mechanical properties. Furthermore, the utilization of SHapley Additive Explanations offers a robust method for interpreting results, enhancing transparency in model predictions. Lastly, the identification of critical parameters using PDP contributes valuable insights into the nuanced relationships governing BFRC behavior. Together, these innovations propel the field towards more accurate, interpretable, and insightful predictions in the realm of concrete technology.
مهارت های من
زبان های جهان :
انگلیسی : حرفه ای
فرانسوی : حرفه ای
فارسی : زبان اصلی
وبلاگ من
مقاله تست
لورم ایپسوم متن ساختگی با تولید سادگی نامفهوم از صنعت چاپ و با استفاده از طراحان گرافیک است. چاپگرها و متون بلکه روزنامه و مجله در ستون و سطرآنچنان که لازم است و برای شرایط
مقاله تست
لورم ایپسوم متن ساختگی با تولید سادگی نامفهوم از صنعت چاپ و با استفاده از طراحان گرافیک است. چاپگرها و متون بلکه روزنامه و مجله در ستون و سطرآنچنان که لازم است و برای شرایط