Measuring walking convenience of Hong Kong resource recycling services: Utilizing a spatially refined model framework of multiple recyclable wastes

Proper recycling leads to environmental benefits. However, there is still a lack of research on the convenience of walking recycling. We introduced the Walking Recycling Capability Index (WRCI) for paper, metal, plastic, and glass, and conducted a case study in Hong Kong. We initially trained machine learning models for municipal solid waste (MSW) generation and non-negative linear regression models for RW proportions, with R² values ranging from 0.68 to 0.95. By synthesizing the above models, we developed a stacking model (R² = 0.6731) to downscale RWs to 0.25 km² scale, which served as the demand-side parameter of the Gaussian Two-Step Floating Catchment Area (Ga2SFCA) for calculating the WRCI. By 2025, nearly 70 % of the areas will have a zero or low WRCI in Hong Kong. Our study innovatively provides a high-resolution RWs generation for walking accessibility analysis, offering decision support for enhancing residents' participation in the circular economy.