Abstract: The layout of urban waste cooking oil (WCO) pretreatment facilities critically determines the efficiency of collection and transportation as well as operational costs. It is a key link in connecting the entire chain of “production, collection, transportation, treatment and utilization”, and also the core support for achieving resource recycling. Taking Beijing as a case study, a site selection optimization model integrating spatial gridding and network location allocation was constructed. Firstly, the catering Point-of-Interest (POI) data were aggregated into a 500 m×500 m regular grid to characterize the spatial heterogeneity of demand. Secondly, a road network dataset with travel time as impedance was constructed based on the actual road network data, and the differentiated service time thresholds (30 minutes in the central urban area, 45 minutes in the plain new towns, and 60 minutes in the ecological conservation area) were set according to the urban functional zones. Finally, to balance system efficiency and service equity, the residual-demand P-median model (RPM) was employed to minimize total transport impedance and a capacitated maximal covering location model (CMCLM) was introduced to maximize coverage under capacity constraints, to carry out multi-scenario optimization solution. The results revealed that the three existing facilities could cover only 44.11% of the city’s demand under the capacity-constrained service threshold of 16 minutes, exposing a significant service deficit in the northern region of Beijing. Three highly robust nodes of Changping Asuwei, Chaoyang and Haidian were consistently identified across both models, forming a rationalized collection network with the existing facilities. The marginal benefits of the newly added facilities exhibited diminishing returns, with two to three new facilities representing the optimal increment range in the near future, while with three new facilities could basically eliminate the blind spots throughout the region. The research showed that the differentiated threshold approach and the assignment of facility weights effectively balanced the transportation efficiency and spatial equity. The grid-network coupling method based on ArcGIS can significantly improve the computational efficiency while ensuring accuracy, providing a replicable decision-support for the refined layout of environmental sanitation facilities in megacities.

Key words: waste cooking oil, pretreatment facilities, location optimization, spatial gridding, location-allocation model, GIS