Automating embodied and operational carbon assessment in urban sustainable development
Siavash Ghorbany, Ming Hu, Siyuan Yao, Matthew Sisk, Chaoli Wang
Abstract
The construction industry is a major contributor to global greenhouse gas emissions, with embodied carbon playing a key role. This paper introduces EcoSphere, an integrated software for automating sustainable urban development by analyzing trade-offs between embodied and operational carbon emissions, construction costs, and environmental impacts. It leverages National Structure Inventory data, computer vision, and large language models on Google Street View and satellite imagery to provide high-resolution, building-specific insights. Using a bottom-up approach, it categorizes buildings into archetypes to create a baseline emissions dataset. Designed for policymakers and non-experts, EcoSphere enables data-driven decision-making on policy scenarios and mitigation strategies. Case studies in Chicago and Indianapolis, USA, highlight its effectiveness in reducing emissions and costs. By simplifying complex data into actionable insights, EcoSphere empowers stakeholders to support carbon neutrality goals, making it a crucial tool for sustainable urban planning.