Presenting a thermal energy efficiency management model in the construction industry: a data mining approach

Ghaderi Rahaghi, Majid; Rahimi, Kamran; Agha Gholizadeh Sayar, Alireza

doi:10.48301/vjtve.2026.524178.1116

Presenting a thermal energy efficiency management model in the construction industry: a data mining approach

Articles in Press

Document Type : Original Article

Authors

Majid Ghaderi rahaghi ¹

Kamran Rahimi ¹

Alireza Agha gholizadeh sayar ²

¹ Faculty Member, Department of Humanities, National University of Skills (NUS), Tehran, Iran

² PhD of Industrial Management, Department of Humanities, National University of Skills, Tehran, Iran

10.48301/vjtve.2026.524178.1116

Abstract

In recent years, the significant increase in global energy demand, combined with the limited availability of non-renewable resources, has highlighted the critical need to optimize energy consumption and reduce waste, particularly in the building sector. In Iran, the building and housing sector accounts for a disproportionately high share of total energy consumption, exceeding global averages, which underscores the importance of targeted energy efficiency measures. This study investigates twelve key variables across 768 diverse building samples to identify and analyze the main factors influencing heating energy efficiency. Utilizing data mining techniques alongside artificial neural network (ANN) algorithms implemented in SPSS Modeler software, the study quantitatively evaluates the impact of these variables. Results indicate that overall building height, roof area, and window area are the most significant parameters affecting heating energy consumption, collectively explaining over 68% of the observed variance. The ANN model further demonstrates high predictive performance, achieving an accuracy rate of 96.3%. These findings provide practical, quantitative insights that can assist engineers, architects, and policymakers in developing effective strategies to improve energy efficiency in heating systems within buildings.

Keywords

Productivity management

energy economics

neural network

data mining

Subjects