Details
| Original language | English |
|---|---|
| Article number | 100296 |
| Number of pages | 12 |
| Journal | Developments in the Built Environment |
| Volume | 17 |
| Early online date | 12 Dec 2023 |
| Publication status | Published - Mar 2024 |
Abstract
The decision-making process in real-world implementations has been affected by a growing reliance on data-driven models. Recognizing the limitations of isolated methodologies - namely, the lack of domain understanding in data-driven models, the subjective nature of empirical knowledge, and the idealized assumptions in first-principles simulations, we explore their synergetic integration. We showed the potential risk of biased results when using data-driven models without causal analysis. Through a case study on energy consumption in building design, we demonstrate how causal analysis significantly enhances the modeling process, mitigating biases and spurious correlations. We concluded that: (a) Sole data-driven models' accuracy assessment or domain knowledge screening may not rule out biased and spurious results; (b) Data-driven models' feature selection should involve careful consideration of causal relationships, especially colliders; (c) Integrating causal analysis results aid to first-principles simulation design and parameter checking to avoid cognitive biases. We advocate for the routine integration of causal inference within data-driven models in engineering practices, emphasizing its critical role in ensuring the models' reliability and real-world applicability.
Keywords
- Architecture, engineering, and construction industry, Biased outcomes, Building energy performance, Causal inference, Cognitive biases, Data-driven models, Domain knowledge, Feature selection, First-principles simulations, Machine learning methods
ASJC Scopus subject areas
- Engineering(all)
- Architecture
- Engineering(all)
- Civil and Structural Engineering
- Engineering(all)
- Building and Construction
- Materials Science(all)
- Materials Science (miscellaneous)
- Computer Science(all)
- Computer Science Applications
- Computer Science(all)
- Computer Graphics and Computer-Aided Design
Sustainable Development Goals
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Developments in the Built Environment, Vol. 17, 100296, 03.2024.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Using causal inference to avoid fallouts in data-driven parametric analysis
T2 - A case study in the architecture, engineering, and construction industry
AU - Chen, Xia
AU - Sun, Ruiji
AU - Saluz, Ueli
AU - Schiavon, Stefano
AU - Geyer, Philipp
N1 - Funding Information: We gratefully acknowledge the German Research Foundation ( DFG ) support for funding the project under grant GE 1652/3-2 in the Researcher Unit FOR 2363 and under grant GE 1652/4-1 as a Heisenberg professorship.
PY - 2024/3
Y1 - 2024/3
N2 - The decision-making process in real-world implementations has been affected by a growing reliance on data-driven models. Recognizing the limitations of isolated methodologies - namely, the lack of domain understanding in data-driven models, the subjective nature of empirical knowledge, and the idealized assumptions in first-principles simulations, we explore their synergetic integration. We showed the potential risk of biased results when using data-driven models without causal analysis. Through a case study on energy consumption in building design, we demonstrate how causal analysis significantly enhances the modeling process, mitigating biases and spurious correlations. We concluded that: (a) Sole data-driven models' accuracy assessment or domain knowledge screening may not rule out biased and spurious results; (b) Data-driven models' feature selection should involve careful consideration of causal relationships, especially colliders; (c) Integrating causal analysis results aid to first-principles simulation design and parameter checking to avoid cognitive biases. We advocate for the routine integration of causal inference within data-driven models in engineering practices, emphasizing its critical role in ensuring the models' reliability and real-world applicability.
AB - The decision-making process in real-world implementations has been affected by a growing reliance on data-driven models. Recognizing the limitations of isolated methodologies - namely, the lack of domain understanding in data-driven models, the subjective nature of empirical knowledge, and the idealized assumptions in first-principles simulations, we explore their synergetic integration. We showed the potential risk of biased results when using data-driven models without causal analysis. Through a case study on energy consumption in building design, we demonstrate how causal analysis significantly enhances the modeling process, mitigating biases and spurious correlations. We concluded that: (a) Sole data-driven models' accuracy assessment or domain knowledge screening may not rule out biased and spurious results; (b) Data-driven models' feature selection should involve careful consideration of causal relationships, especially colliders; (c) Integrating causal analysis results aid to first-principles simulation design and parameter checking to avoid cognitive biases. We advocate for the routine integration of causal inference within data-driven models in engineering practices, emphasizing its critical role in ensuring the models' reliability and real-world applicability.
KW - Architecture, engineering, and construction industry
KW - Biased outcomes
KW - Building energy performance
KW - Causal inference
KW - Cognitive biases
KW - Data-driven models
KW - Domain knowledge
KW - Feature selection
KW - First-principles simulations
KW - Machine learning methods
UR - http://www.scopus.com/inward/record.url?scp=85180361755&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2309.1150
DO - 10.48550/arXiv.2309.1150
M3 - Article
AN - SCOPUS:85180361755
VL - 17
JO - Developments in the Built Environment
JF - Developments in the Built Environment
M1 - 100296
ER -