This paper presents a UAV-based integrative IoT platform that leverages UAVs to deliver different IoT services from height. One of the major tasks of the platform is to select the appropriate UAVs for a particular IoT task. This selection may be based on different criteria, such as UAV's equipment, energy budget and geographical proximity of the UAV to the area of interest. For the selection mechanism, this paper proposes and formulates two Linear Integer Problem (LIP) optimization solutions by aiming at minimizing the energy consumption and shortening the UAV operation time. These two solutions are dubbed Energy-Aware Selection of UAVs (EAS) and Delay- Aware Selection of UAVs (DAS). They are both evaluated through simulations. The obtained results show that if the objective is energy efficiency, EAS is more efficient than DAS in terms of reducing the total energy consumption by the UAVs. Additionally, if the time is the objective, DAS exhibits better performance than EAS in terms of operation time.