SMART CONSTRUCTION ENGINEERING AND ROBOTICS-BASED AUTOMATION FOR SUSTAINABLE INFRASTRUCTURE DEVELOPMENT

Abstract

Smart construction engineering and robotics-based automation are increasingly transforming infrastructure development by improving productivity, sustainability, safety, and project delivery efficiency. This study investigates the role of intelligent construction technologies in supporting sustainable infrastructure development through the integration of robotics, artificial intelligence, Building Information Modeling, Internet of Things sensors, drone-based monitoring, and automated construction equipment. The research uses a secondary and simulation-based dataset designed to represent smart construction project conditions, including robotic equipment performance, construction productivity records, site safety observations, energy consumption data, material waste logs, labor utilization indicators, carbon emission estimates, and infrastructure project scheduling parameters. The final dataset contained 18,500 construction activity records collected from published construction automation studies, open infrastructure datasets, BIM-based simulation outputs, and MATLAB-generated robotic workflow scenarios. The dataset was categorized into five major performance dimensions: productivity efficiency, safety improvement, material waste reduction, energy optimization, and carbon emission control. The proposed framework was developed and evaluated using Python 3.11, MATLAB R2024a, Autodesk Revit, Navisworks Manage, Microsoft Excel, and SPSS. Python was used for data preprocessing, machine learning model development, performance evaluation, and visualization through Scikit-learn, Pandas, NumPy, Matplotlib, and TensorFlow. MATLAB was used for robotic motion simulation and automation workflow analysis, while Revit and Navisworks were used for BIM-based construction planning and clash detection. SPSS was used for statistical validation of sustainability indicators. Experimental results show that robotics-based automation improved construction productivity by 27.8%, reduced material waste by 22.4%, decreased site safety incidents by 31.6%, improved equipment utilization by 24.9%, and reduced estimated carbon emissions by 18.7% compared with conventional construction practices. The AI-supported prediction model achieved 94.2% accuracy, 93.6% precision, 94.0% recall, and 93.8% F1-score in classifying construction performance conditions. The findings confirm that smart construction engineering combined with robotic automation can significantly enhance sustainable infrastructure development by enabling data-driven decision-making, automated site operations, real-time monitoring, and resource-efficient project execution.