A Machine Learning-Based Framework for the Continuous Improvement of Economic Assessment in Construction Projects

A Machine Learning-Based Framework for the Continuous Improvement of Economic Assessment in Construction Projects

Mahmoud Golabchi1 Mohammad Hossein Mohseni2

1) Professor, College of Fine Arts, University of Tehran
2) Pars University of Art and

IOI Code : AAHE-ABGAKE
Publication : National Conference on Civil Engineering, Design and Construction Technologies, Energy, Materials and Advanced Machinery(icaeu.ir)
Abstract :
    This study investigates the role of machine learning in the continuous improvement of economic studies in construction projects. Economic studies constitute one of the fundamental components of project feasibility assessment and provide the basis for investment justification, cost-benefit estimation, risk evaluation, and economic performance analysis. However, the complexity of construction projects, fluctuations in material and energy prices, exchange rate volatility, market uncertainties, financial constraints, and environmental changes often result in significant discrepancies between initial estimates and actual project outcomes. Under such conditions, conventional approaches alone are insufficient to meet the growing need for accurate, dynamic, and continuously updated economic analyses. Accordingly, this research adopts an analytical approach based on a comprehensive literature review to examine the potential of machine learning in enhancing the quality of economic studies for construction projects. The findings indicate that machine learning algorithms can improve the prediction accuracy of project costs, schedule performance, cash flow, economic risks, and financial indicators through the analysis of historical and real-time project data. Furthermore, the capability of continuously retraining models using newly available data transforms economic studies from a static process into an intelligent, adaptive, and continuously learning system. The results suggest that integrating machine learning with continuous improvement frameworks, particularly the Plan–Do–Check–Act (PDCA) cycle, can provide an effective mechanism for continuously revising assumptions, reducing estimation errors, identifying critical variables, monitoring deviations, and supporting economic decision-making. Nevertheless, successful implementation of this approach requires high-quality data, adequate digital infrastructure, interpretable models, interdisciplinary expertise, and a data-driven decision-making culture. Overall, machine learning can be regarded as a strategic tool for improving the accuracy, transparency, and efficiency of economic studies and enhancing the success of construction projects.

Keywords : continuous improvement economic assessment machine learning