Early Stage Embodied and Operational Analysis for Guiding Sustainable Architectural Design Decisions

Abstract

Buildings account for a significant portion of global energy consumption and greenhouse gas emissions. Simulating building performance in the early design stage allows architects and engineers to adjust design decisions to reduce embodied carbon and energy consumption. Life-cycle assessment (LCA) is one of the most comprehensive methodologies to evaluate the environmental impact of architectural production and operation. This thesis aims to address the challenges involved in applying LCA to architectural design in the early design stage. By conducting a literature review of the status quo of architectural LCA and identifying the gaps in existing research and tools, this paper continues the research of a novel workflow in Grasshopper that calculates greenhouse gas (GHG) emissions and costs from both embodied and operational phases. The workflow addresses the early-stage uncertainty through random inputs with a Monte Carlo approach and implements surrogate models to accelerate the process for each iteration. The author's contribution to the workflow includes improving its robustness and accuracy by redesigning the simulation model to generate more accurate training data and transitioning to a new machine-learning algorithm. The results of the study provide insights into design decisions that can reduce embodied and operational carbon. A parallel case study was conducted to assess the trade-offs between embodied and operational carbon with regard to construction material selection. In the end, the thesis also proposes possible future research directions.