관심지점기반 도시철도 역별 이용수요 추정모형 개발 = Development of an Urban Railway Demand Estimation Model Based on Point of Interest

다국어 초록 (Multilingual Abstract)

Past studies on subway ridership in Korea have largely relied on socio-economic indicators and aggregated land-use variables (e.g., population, number of households and employees, residential and commercial floor area) measured at the level of administrative districts or traffic analysis zones. While useful for capturing broad demand patterns, such approaches have limited ability to explain how the size and spatial distribution of facilities within each station area affect station-level ridership. This study addresses this gap by developing a ridership model based on Point of Interest (POI) data that explicitly reflect the internal spatial structure of station areas.
Using National Point of Interest (N-POI) data for Seoul, POI categories are reclassified into twelve functional groups and linked to road-name address building footprints and the Building Register to derive building floor area by function. Park areas are obtained from the “Living Area Plan Facilities (Parks)” dataset. On this basis, functional facility floor area is constructed for 500m catchment areas around stations on Seoul Metro Line 2. Station-level subway demand is defined as average weekday boardings derived from smart-card transaction data and is used as the dependent variable.
An intercept-free multiple linear regression model is estimated, and to alleviate multicollinearity and degrees-of-freedom issues, the twelve functional categories are consolidated into four: commercial/service, residential/business/education, medical, and religious facilities. The final model achieves an uncentered R² of 0.856 and a root mean squared error of 14,268 passengers, indicating that a limited set of facility-based variables explains station-level ridership patterns to a considerable extent. The results highlight that POI-based functional structure, combined with building floor area, provides a useful basis for constructing fine-scale public transport demand models.

목차 (Table of Contents)

• I. 서론 1
• 1.1 연구 배경 및 목적 1
• 1.2 논문 구성 3
• II. 선행 연구 분석 4
• 2.1 도시철도 이용수요모형 4
• 2.2 역세권 공간 범위 및 기능 구조 8
• 2.2.1 역세권 정의 및 범위 8
• 2.2.2 관심지점 기반 공간구조 분석 9
• 2.2.3 관심지점기반 수요변수 구축 10
• 2.3 관심지점자료 분류체계·품질관리 13
• 2.4 선행 연구 시사점 15
• III. 공간정보 자료 구축 및 정제 17
• 3.1 공간정보 자료 구축 17
• 3.2 공간정보 자료 정제 23
• 3.2.1 관심지점정보(POI) 정제 23
• 3.2.2 건물 도형 및 면적 정제 38
• 3.2.3 공원 도형 및 면적 정제 41
• 3.2.4 관심지점정보 및 도형·면적 결합 42
• IV. 모형 설정 및 분석 47
• 4.1 변수 설정 47
• 4.2 모형 설정 49
• 4.3 모형 추정 및 결과 52
• V. 결론 및 향후 과제 63