ID: 46
Presenting Author: Hee-soo Hwang
Status: pending
AI PSS identifies carbon hotspots and drivers, prioritizing them for tailored mitigation policies, ensuring efficient and impactful environmental assessment.
This research addresses the critical challenge of efficiently allocating resources for carbon mitigation within urban environments, where traditional Environmental Impact Assessment (EIA) often lacks the spatial granularity required for actionable policy. We propose an AI-driven Planning Support System (PSS) designed to transform fine-grained carbon spatial data into prioritized policy actions for city planners.
The methodology is four-fold: First, we use Spatial Autocorrelation (Moran’s I/LISA) combined with Machine Learning (ML) clustering to detect dynamic carbon emission hotspots at a local, grid level. Second, ML regression models identify the primary urban drivers—such as land use, transportation, and building characteristics—linked to these high-emission areas. Third, we develop a novel Impact Assessment Framework that matches specific hotspot profiles with tailored policy interventions from a national/local carbon-neutral policy inventory, prioritizing strategies based on potential impact and resource efficiency. Finally, a validation study in selected regions will quantify the cost-benefit efficiency of the AI-based prioritization strategy compared to conventional approaches, proving its value as a proactive Decision Support Tool. This PSS is crucial for urban planners seeking to maximize limited resources, offering a scalable system designed for integration into existing urban planning workflows and administrative data structures.
Author is a researcher at the Korea Institute of Civil Engineering and Building Technology, specializing in AI-driven spatial systems for carbon-neutral and sustainable urban planning.
Coauthor 1: Seunghyun Jung