AI-Enabled GIS Decision Support for Detecting Demand–Supply Gaps in Bangkok’s Rental Real Estate Market

This study develops an AI-enabled GIS decision-support framework for detecting spatial demand–supply gaps in Bangkok’s rental real estate market. The objective is to identify neighbourhood-level opportunity areas and oversupply risks by integrating alternative platform-based indicators with spatial analytics. The study combines rental listing data, calendar availability, review activity, and geographic boundary data. After preprocessing, the analytical dataset contained 23,088 rental listings, 583,333 review records, and 50 Bangkok neighbourhoods. Calendar unavailability was used as a proxy for potential booked activity, while listing counts were used to measure supply pressure. These indicators were aggregated by neighbourhood and normalized to construct a demand index, supply index, demand–supply gap score, saturation-risk score, and review-validated opportunity score. GIS visualization was then used to map spatial market patterns and classify areas into opportunity, balanced-market, and oversupply-risk groups.

The results show that Pathum Wan had the strongest direct demand–supply opportunity, with the highest positive gap score and an average occupancy rate of 31.05%. Huai Khwang, Khlong Toei, and Phra Nakhon showed stronger oversupply risk, indicating high supply pressure relative to demand signals. When review momentum was included, Vadhana, Pom Prap Sattru Phai, Bang Rak, and Pathum Wan emerged as high-opportunity areas. Most Bangkok neighbourhoods were classified as balanced markets, but several districts displayed clear opportunity or saturation signals. The proposed framework supports rental market screening, investment planning, and urban spatial decision-making by combining demand proxies, supply indicators, review momentum, and GIS-based visualization.