Predicting whole-life carbon emissions for buildings using different machine learning algorithms - A case study on typical residential properties in Cornwall, UK

Aug 5, 2026

Whole-life carbon emissions (WLCE) research evaluates buildings’ environmental impacts and supports sustainable design. Existing WLCE estimation methods remain time- and data-intensive, which restricts use during early design. This study proposes a novel machine-learning framework to predict building WLCE and WLCE intensity (normalised by floor area), using models trained on 150 typical residential properties in Cornwall (UK) and 28 survey-derived features. Ten algorithms are compared with R2, MAE, MSE, RMSE, and runtime, showing non-linear models outperform linear ones and Random Forest achieves the best accuracy, stability, and efficiency, enabling life-cycle integration early in tight design schedules.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Highlights

Novel ML approach for WLCE prediction

WLCE integration into early building design

UK residential dataset for robust modeling

Article Information

Keywords and abstract

Introduction

Buildings as major carbon emitters and need for WLCE prediction

Applied Energy 357 (2024) 122472 Contents lists available at ScienceDirect Applied Energy journal [homepage:](homepage: www.elsevier.com/locate/apenergy)[ www.elsevier.com/locate/apenergy](homepage: www.elsevier.com/locate/apenergy) | Predicting whole-life carbon emissions for buildings using different machine learning algorithms: A case study on typical residential properties in Cornwall, UK
Lin Zheng a, b, Markus Mueller b, c, d, Chunbo Luod, e, Xiaoyu Yana, b, *
a Renewable Energy Group, Engineering Department, Faculty of Environment, Science and Economy, University of Exeter, Penryn Campus, Penryn TR10 9FE, UK b Environment and Sustainability Institute, Faculty of Environment, Science and Economy, University of Exeter, Penryn Campus, Penryn TR10 9FE, UK c Department of Earth and Environmental Sciences, Faculty of Environment, Science and Economy, University of Exeter, Penryn Campus, Penryn TR10 9FE, UK d Institute for Data Science and Artificial Intelligence, University of Exeter, Exeter EX4 4RN, UK
e Department of Computer Science, Faculty of Environment, Science and Economy, University of Exeter, Streatham Campus, Exeter, EX4 4RN, UK | | | | --- | --- | --- | | H I G H L I G H T S | | | | • Novel method using machine learning algorithms for building WLCE prediction.
• Promotes WLCE integration into the early stages of building design.
• Leverage a comprehensive UK residential dataset for robust model development. | | | | A R T I C L E I N F O | | A B S T R A C T | | Keywords:
Whole life carbon emissions Life cycle thinking Machine learning algorithms Sustainable building Carbon reduction
Data-driven approaches | | Whole-life carbon emissions (WLCE) studies are critical in assessing the environmental impact of buildings and promoting sustainable design practices. However, existing methods for estimating WLCE are time-consuming and data-intensive, limiting their usefulness in the early building design stages. In response to this, this research introduces a novel approach by harnessing various machine learning algorithms to predict WLCE and WLCE intensity (normalised by floor area) for buildings. To evaluate the suitability of machine learning algorithms, we conducted an experiment involving ten algorithms to build the prediction models. These models were trained using data from 150 typical residential properties in Cornwall, UK, along with 28 features obtained from a comprehensive survey, including floor area, heating type, and occupant characteristics. The ten algorithms include Multiple Linear Regression, and non-linear algorithms such as Decision Tree, Random Forest. Performance evaluation metrics, such as coefficient of determination (R2), mean absolute error (MAE), means squared error (MSE), root-mean-square error (RMSE), and elapsed time, were employed. Our research contributes to the field by showcasing the effectiveness of machine learning models in predicting building WLCE. We reveal that all the tested machine learning algorithms have the capability to predict WLCE and WLCE intensity, non-linear models outperform linear ones, and the Random Forest (RF) model demonstrates superior performance in terms of accuracy, stability, and efficiency. This research encourages the integration of life cycle studies into the early design stage, even within tight building design schedules, offering practical guidance to architects and designers. Furthermore, these results also benefit a wide range of stakeholders, not only the architects but also the engineers, policymakers, and life cycle assessment (LCA) researchers, contributing to the advancement of data-driven sustainability approaches within the building sector. | Abbreviations: LCA, Life Cycle Assessment; WLCE, Whole-Life Carbon Emissions; R2 , Coefficient of Determination; MAE, Mean Absolute Error; MSE, Mean Squared Error; RMSE, Root Mean Squared Error; MLR, Multiple linear regression; LASSO, Least Absolute Shrinkage and Selection Op

FAQs

Why is predicting whole-life carbon emissions (WLCE) important in building design?

WLCE studies assess buildings’ environmental impacts across the full lifespan. Predicting WLCE helps support sustainable design practices and life cycle thinking, especially when early decisions affect long-term performance.

What machine learning approach does the research propose for WLCE estimation?

The research introduces multiple machine learning algorithms to predict WLCE and WLCE intensity (normalised by floor area). Models are trained using 150 typical residential properties in Cornwall, UK, with 28 features from a comprehensive survey.

Which algorithm performs best and how do results compare linear vs non-linear models?

All tested algorithms can predict WLCE and WLCE intensity. Non-linear models outperform linear ones, and Random Forest (RF) shows superior performance in accuracy, stability, and efficiency based on evaluation metrics like R2, MAE, MSE, RMSE, and runtime.

Aug 5, 2026

Upload by Seraphina