Evaluating urban accessibility: leveraging open-source data and analytics to overcome existing limitations

We revisit the standard methodology for evaluating proximity to urban services and recommend enhancements to address existing limitations. Existing approaches often simplify their measure of proximity by using large areal units and by imposing arbitrary distance thresholds. By doing so, these approaches risk overlooking vulnerable, access-poor populations - the very populations that such studies are often trying to identify. These limitations are primarily motivated by computational constraints. However, recent advances in computational power, open data, and open-source analytics permit high-resolution proximity analyses on large scales. Given the impetus for equitable accessibility in our communities, this is of fundamental importance for researchers and practitioners. In this paper, we present an approach that leverages these open source advances to (a) measure proximity using network distance at the building level, (b) estimate population at that level, and (c) present the resulting distributions so vulnerable populations can be identified. Using three cities and modes of transport, we demonstrate how the approach enhances existing measures and identifies service-poor populations where the previous methods fall short. The proximity results could be used alone, or as inputs to access metrics. Our collating of these components into an open source code provides opportunities for researchers and practitioners to explore fine-resolution, city-wide accessibility across multiple cities and the host of questions that follow.