Article: Using regression-based machine learning model to estimate constructions cost Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.17 pp.72 - 92 Abstract: Accurate construction cost estimation is crucial for effective project planning and resource management in the construction industry. Traditional estimation methods often suffer from inaccuracies due to the complexity and variability of construction projects. This study explores the application of regression-based machine learning models support vector machine (SVM), K-nearest neighbours (KNN), and multilayer perceptron (MLP) to improve the precision of construction cost predictions. The study evaluates the performance of these models using a construction-related dataset that includes factors such as material costs, labour expenses, and project characteristics. The results revealed that the SVM model outperforms the others, achieving an RMSE of 18,189 and an R<SUP align="right"><SMALL>2</SMALL></SUP> of 0.975, indicating its superior ability to predict construction costs accurately. The KNN and MLP models also demonstrated effectiveness, but with higher errors, particularly in more complex data scenarios. This research highlights the potential of machine learning techniques to revolutionise construction cost estimation, providing more reliable, data-driven insights for project planning and budgeting. Inderscience Publishers - linking academia, business and industry through research

Title: Using regression-based machine learning model to estimate constructions cost

Authors: Yanfei Shen

Addresses: Sanmenxia Polytechnic, Sanmenxia, Henan, 472000, China

Abstract: Accurate construction cost estimation is crucial for effective project planning and resource management in the construction industry. Traditional estimation methods often suffer from inaccuracies due to the complexity and variability of construction projects. This study explores the application of regression-based machine learning models support vector machine (SVM), K-nearest neighbours (KNN), and multilayer perceptron (MLP) to improve the precision of construction cost predictions. The study evaluates the performance of these models using a construction-related dataset that includes factors such as material costs, labour expenses, and project characteristics. The results revealed that the SVM model outperforms the others, achieving an RMSE of 18,189 and an R² of 0.975, indicating its superior ability to predict construction costs accurately. The KNN and MLP models also demonstrated effectiveness, but with higher errors, particularly in more complex data scenarios. This research highlights the potential of machine learning techniques to revolutionise construction cost estimation, providing more reliable, data-driven insights for project planning and budgeting.

Keywords: cost estimation; predictive analytics; project planning; cost prediction; construction industry.

DOI: 10.1504/IJICT.2025.146673

International Journal of Information and Communication Technology, 2025 Vol.26 No.17, pp.72 - 92

Received: 19 Feb 2025
Accepted: 08 Mar 2025
Published online: 11 Jun 2025 *

Title: Using regression-based machine learning model to estimate constructions cost

Keep up-to-date