Article: BIM-based carbon emissions tracing and diagnosis for building operations and maintenance using constraint-based causal discovery Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.89 pp.1 - 19 Abstract: During the building operation and maintenance stage, the factors influencing carbon emissions are complex and highly coupled. The existing diagnostic methods based on correlation analysis or black-box models are unable to reveal the underlying causal mechanisms. This paper integrates the static topology and equipment attributes in the building information model as prior knowledge into a constrained causal discovery framework, learns causal structures from the dynamic data of the internet of things, and realises the diagnosis from correlation descriptions to causal tracing. Experiments on public datasets show that the accuracy rate, recall rate, and area under the curve value of this method in identifying key influencing factors are 0.92, 0.88, and 0.95 respectively, which are 14%, 11%, and 7% higher than the standard Peter-Clark algorithm, and can track key causal chains such as 'outside temperature, air conditioning set temperature, carbon emissions'. Inderscience Publishers - linking academia, business and industry through research

Title: BIM-based carbon emissions tracing and diagnosis for building operations and maintenance using constraint-based causal discovery

Authors: Yan Ma

Addresses: Fujian Sanming Yihong Construction Engineering Co., Ltd., Sanming, 365000, China

Abstract: During the building operation and maintenance stage, the factors influencing carbon emissions are complex and highly coupled. The existing diagnostic methods based on correlation analysis or black-box models are unable to reveal the underlying causal mechanisms. This paper integrates the static topology and equipment attributes in the building information model as prior knowledge into a constrained causal discovery framework, learns causal structures from the dynamic data of the internet of things, and realises the diagnosis from correlation descriptions to causal tracing. Experiments on public datasets show that the accuracy rate, recall rate, and area under the curve value of this method in identifying key influencing factors are 0.92, 0.88, and 0.95 respectively, which are 14%, 11%, and 7% higher than the standard Peter-Clark algorithm, and can track key causal chains such as 'outside temperature, air conditioning set temperature, carbon emissions'.

Keywords: building information model; causal discovery; carbon emission; traceability diagnosis.

DOI: 10.1504/IJICT.2026.155690

International Journal of Information and Communication Technology, 2026 Vol.27 No.89, pp.1 - 19

Received: 07 Apr 2026
Accepted: 23 May 2026
Published online: 10 Aug 2026 *

Title: BIM-based carbon emissions tracing and diagnosis for building operations and maintenance using constraint-based causal discovery

Keep up-to-date