The low battery autonomy of Unnamed Aerial Vehicles (UAVs or drones) can make smart farming (precision agriculture), disaster recovery, and the fighting against dengue vector applications difficult. This article considers two approaches, first enumerating the characteristics observed in these three IoT application types and then modeling an UAV's battery recharge coordination using the Agent-Based Simulation (ABS) approach. In this way, we propose that each drone inside the swarm does not communicate concerning this recharge coordination decision, reducing energy usage and permitting remote usage. A total of 6000 simulations were run to evaluate how two proposed policies, the BaseLine (BL) and ChargerThershold (CT) coordination recharging policy, behave in 30 situations regarding how each simulation sets conclude the simulation runs and how much time they work until recharging results. CT policy shows more reliable results in extreme system usage. This work conclusion presents the potential of these three IoT applications to achieve their perpetual service without communication between drones and ground stations. This work can be a baseline for future policies and simulation parameter enhancements.

The double hypothesis test (DHT) is a test that allows controlling Type I (producer) and Type II (consumer) errors. It is possible to say whether the batch has a defect rate, p, between 1.5 and 2%, or between 2 and 5%, or between 5 and 10%, and so on, until finding a required value for this probability. Using the two probabilities side by side, the Type I error for the lower probability distribution and the Type II error for the higher probability distribution, both can be controlled and minimized. It can be applied in the development or manufacturing process of a batch of components, or in the case of purchasing from a supplier, when the percentage of defects (p) is unknown, considering the technology and/or process available to obtain them. The power of the test is amplified by the joint application of the Limit of Successive Failures (LSF) related to the Renewal Theory. To enable the choice of the most appropriate algorithm for each application. Four distributions are proposed for the Bernoulli event sequence, including their computational efforts: Binomial, Binomial approximated by Poisson, and Binomial approximated by Gaussian (with two variants). Fuzzy logic rules are also applied to facilitate decision-making.