Assessing and improving spatial equality in hyper-dense cities: a network analysis of amenity accessibility of Hong Kong’s public housing estates

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Introduction In hyper‐dense cities such as Hong Kong, unequal access to essential amenities has important implications for social equity, particularly within the public housing sector. However, accessibility differences between Public Rental Housing (PRH) and Home Ownership Scheme (HOS) estates have rarely been examined from both multimodal travel‐time and network‐structural perspectives. Methods This study examined 147 public housing estates in Kowloon, including 82 PRH and 65 HOS estates, and assessed access to five categories of public medical and educational facilities: general outpatient clinics, hospitals, specialist outpatient clinics, public primary schools, and public secondary schools. Travel times by driving, public transit, and walking were obtained from the Google Maps Distance Matrix API on weekdays in February 2025. Estate‐level accessibility was measured using average travel time to retained facilities, PRH‐HOS differences were tested statistically, and GIS‐based Kernel Density Estimation and Gephi network analysis were used to assess spatial concentration, closeness centrality, modularity, and network density. Results PRH estates had modestly but significantly longer driving times than HOS estates. Across all key facilities, the mean driving time was 0.47 min longer for PRH estates, corresponding to a 5.1% increase. Public-transit differences were generally small and non-significant, whereas walking differences varied by facility type. PRH estates also showed lower average closeness centrality (0.38 versus 0.52) and lower network density (0.22 versus 0.35), while the overall modularity score of 0.62 indicated distinct accessibility communities. Discussion Accessibility inequality within Hong Kong’s public housing sector is mode‐ and amenity‐specific and is reflected not only in travel‐time differences but also in the structural positions of estates within the housing‐amenity network. The findings support targeted amenity allocation, transit and pedestrian improvements, and accessibility‐sensitive planning for existing and future public housing estates.