Drones-as-a-service: a simulation-based analysis for on-drone decision-making

Abstract

Drone services are expected to emerge in many areas around the world in the near future and this is generating increasing interest. While there is a proliferation of ideas for various applications that can be delivered via drone services, the subject of drone service provisioning has received relatively less attention and, hence, is not well understood. We envision that a variety of drone applications (e.g. renting a drone as a security guard, to pick up and deliver something or to take an interesting photo) can be delivered as a service, using a common set of underlying service provisioning principles, such as making decisions as to whom to service next. In this paper, we study by simulation how different decision-making strategies for drones impact clients and service providers. In particular, the trade-offs between maximising provider revenue versus maximising a client’s personal satisfaction, and different combinations of factors that influence drone behaviour (e.g. speed, distribution of clients, criteria for judging client requests, service duration, effect of battery and whether drones are allowed to change their target client on-the-fly) are investigated. This research has implications for scholars as well as drone service providers and application developers.