Introducing spatiotemporal order crowdedness into the on-demand delivery management data pipeline: A dynamic inverted two-step floating catchment area (Di2SFCA) method

Matching fluctuating delivery demand with local service availability across space and time is a key challenge in urban logistics. In Online-to-Offline (O2O) on-demand delivery service, concentrated order demand can place pressure on retailers and courier operations, with implications for the use of urban space. These spatial conflicts between high-frequency digital demand and limited physical infrastructure require dynamic assessment tools. However, traditional spatial-interaction models remain constrained by static measurements and a predominant focus on consumer access rather than supply-side order pressure. To address this gap, we propose a Dynamic Inverted Two-Step Floating Catchment Area (Di2SFCA) framework that reorients spatial-interaction modeling toward real-time retailer crowdedness. By integrating hourly demand fluctuations, time-varying retailer operating status, and mode-specific delivery catchments, the method introduces a dynamic measure of order crowdedness into the delivery-management data pipeline. Its application to Beijing, China, reveals distinct spatiotemporal rhythms. Order crowdedness was concentrated in business districts during the noon peak but shifted toward residential and mixed-use areas in the evening. The crowdedness-sales relationship was also non-linear and varied across delivery modes, suggesting that excessive crowdedness may reflect operational bottlenecks rather than stronger market demand alone. Di2SFCA provides a transferable framework for dynamic spatial assessment in O2O delivery and, with context-specific calibration, other time-sensitive urban service systems. The findings can support temporally adaptive transport policies and better coordination between urban activity patterns and infrastructure capacity.