Journal of Technology Innovatons and Energy ISSN: 2957-8809 htps://doi.org/10.56556/j te.v2i2.511 Global Scientfc Research 1 RESEARCH ARTICLE Using Machine Learning to Predict Cost Overruns in Constructon Projects Theingi Aung 1 *, Sui Reng Liana 1 , Arkar Htet 1 , Amiya Bhaumik 1 1 Faculty of Business and Accountng, Lincoln University, 47301 Petaling Jaya, Selangor D. E., Malaysia Corresponding author: Theingi Aung; taung@lincoln.edu.my Received: 27 April, 2023, Accepted: 10 June, 2023, Published: 11 June, 2023 Abstract Addressing the persistent issue of cost overruns in construction projects, our study explores the potential of machine learning algorithms for accurately predicting these overruns, utilizing an expansive set of project parameters. We draw a comparison between these innovative techniques and traditional cost estimation methods, unveiling the superior predictive accuracy of machine learning approaches. This research contributes to existing literature by presenting a data-driven, reliable strategy for anticipating and managing construction costs. Our findings have significant implications for project management, offering a path towards more efficient and financially sound practices in the construction industry. The improved prediction capabilities could revolutionize cost management, facilitating better planning, risk mitigation, and stakeholder satisfaction. Keywords: construction projects; cost overruns; machine learning; cost estimation; project management; risk mitigation Introduction Complex projects, tight schedules, and budget limits characterize the construction business, resulting in cost overruns that can significantly impair project success, leading to delays, disagreements, and financial losses (Samiullah S., Abd, H. A., Sasitharan, N., Abdul, F., Kaleem, U., & Kanes,K., 2017). Accurate prediction of cost overruns is essential for effective project management and risk mitigation, as it enables stakeholders to make informed decisions and allocate resources efficiently (Odeh, A. M., & Battaineh, H. T., 2002). Traditional cost estimation methods, such as expert judgment and parametric estimation, have been used for decades but often yield inaccurate results due to their reliance on human expertise and historical data (Flyvbjerg, B., Holm, M. S., & Buhl, S., 2003). In recent years, advances in machine learning and data analytics have provided new opportunities for improving cost estimation in construction projects (Yang, C., Baabak, A., & Minsoo, B., 2018). Machine learning methods, such as linear regression, support vector machines, and artificial neural networks, have demonstrated potential in a variety of disciplines due to their capacity to learn from data and accurately anticipate outcomes (Li, Chengxi, Cheng, Peng, and Chris Cheng., 2023). As a result, there has been growing interest in applying machine learning techniques to construction cost estimation, with several studies reporting promising results (Abolfazl J., Iman, P., & Pete, B., 2021). This study aims to investigate the potential of machine learning algorithms in predicting cost overruns in construction projects, based on a comprehensive set of project parameters. We compare the performance of these algorithms with traditional cost estimation methods to determine their relative accuracy and effectiveness. By providing a more accurate prediction of cost overruns, this research has the potential to significantly impact project management practices, helping stakeholders better anticipate and manage construction project costs. Literature Review Challenges in construction cost estimation Construction cost estimation is a critical component of project management, as it influences decision-making, budget allocation, and project success (Zainab, H. A.,