Starting from March 2020, the social distancing policy has been issued across the world to prevent the spread of the virus. Specifically, no-contact delivery is required for food and drink delivery businesses, seeing a surge in demand, such as Instacart, in areas with an increasing number of confirmed cases. There are three options for restaurants to implement no-contact delivery, such as signing in food delivery apps, operating their own delivery service, or providing pickup service (Drive-thru).
In this work, we study the effects of the COVID-19 pandemic over the food delivery. Firstly, we build a susceptible infected removed (SIR) model to predict the future infections. Then, we construct a multiple linear regression (MLR) to forecast the restaurants' demand by analyzing the real-time data from the food delivery app and the virus spread data. Finally, from the perspective of restaurants, we compare the costs of two delivery options, operating own delivery service and signing in food delivery apps, through the predicted number of orders and randomly generated demand locations.
To design routes and time schedules for using multiple vehicles (e.g., school buses) to deliver school lunches to students in under-served communities.