Introducing causal inference in the energy-efficient building design process

03/14/2022
by   Xia Chen, et al.
0

"What-if" questions are intuitively generated and commonly asked during the design process. Engineers and architects need to inherently conduct design decisions, progressing from one phase to another. They either use empirical domain experience, simulations, or data-driven methods to provide consequential feedback. We take an example from an interdisciplinary domain of energy-efficient building design to argue that the current methods for decision support have four limitations: 1. Less carefully inspected parametric independence raises the risks of biased results and spurious relationships. 2. The integration gap between data-driven methods and knowledge-based approaches. 3. Less explicit model interpretability for informed decision-making. 4. Ambiguous boundaries for machine assistance during the design process. In this study, we first clarify the nature of dynamic experience in individuals and constant principal knowledge in design. Sequentially, we introduce the causal inference into the energy-efficient design domain by proposing a four-step process to reveal and analyze the parametric dependencies within the design space by identifying the design causal diagram with interventions. The causal diagram provides a nexus for integrating domain knowledge with data-driven methods and allows interpretability and testability against the domain experience. The extraction of causal structures from the data is close to the nature design reasoning process. As an illustration, we applied the properties of the proposed estimators through simulations. The paper concludes with a feasibility study that demonstrates the realization of the proposed framework.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/01/2023

Causalvis: Visualizations for Causal Inference

Causal inference is a statistical paradigm for quantifying causal effect...
research
01/16/2020

Combining Offline Causal Inference and Online Bandit Learning for Data Driven Decisions

A fundamental question for companies is: How to make good decisions with...
research
05/12/2021

Bayesian Model Averaging for Data Driven Decision Making when Causality is Partially Known

Probabilistic machine learning models are often insufficient to help wit...
research
08/20/2022

Data-Driven Causal Effect Estimation Based on Graphical Causal Modelling: A Survey

In many fields of scientific research and real-world applications, unbia...
research
08/10/2020

Learning to Learn in Collective Adaptive Systems: Mining Design Patterns for Data-driven Reasoning

Engineering collective adaptive systems (CAS) with learning capabilities...
research
12/11/2018

Learning Predictive Models That Transport

Classical supervised learning produces unreliable models when training a...

Please sign up or login with your details

Forgot password? Click here to reset