Spatiotemporal patterns of online food delivery services before the Covid-19 pandemic

Public health concerns of the Covid-19 pandemic and the popularity of on-demand mobility services have led to a recent and prominent increase in online food delivery (OFD) service adoption. While app-based services that offer restaurant customers the convenience of a freshly prepared meal delivered to any location have existed, their present and future impacts to urban transportation networks and landscapes have become ever more apparent since the pandemic’s onset and subsequent restrictions on restaurant dining. By analyzing route-level data collected by a ridehailing driver assistant app between October 2015 and October 2019, this study informs a baseline understanding of where and when these on-demand food delivery services were used within the Phoenix metro area prior to the pandemic. This identification of the spatiotemporal patterns of OFD service is accompanied by the estimation of traditional and spatially lagged negative binomial models of delivery counts using a robust set of predictors of the built environment and socioeconomic context found at the trip destination. Study results indicate that on-demand food delivery services were popular during dinner hours corresponding with evening peak travel and in neighborhoods characterized by higher activity density, greater drinking establishment access, and increased shares of residents under 45 years old.