Periodic planning of UAVs' fleet mission with the uncertainty of travel parameters

This paper presents a variant of the Periodic Vehicle Routing Problem, where a fleet of homogeneous Unmanned Aerial Vehicles (UAVs) in periodically repeated missions, serves spatially dispersed customers, over a planning horizon. The plan of mission includes both routing and scheduling decisions. The fleet's size and UAV's lifting capacity allow for the implementation of missions. However, the changing weather conditions affect the energy consumption of UAVs, and limits their range because of an earlier depletion of batteries. That may result in only partly fulfilment of the demand. Thus, the planning of subsequent missions should take into consideration, previously unmet demand, as well as the current weather forecast. We aim to propose a new approach to an online, proactive periodic planning, which guarantees timely delivery of the required quantity of goods, while taking into account the forecasted weather changes. We implement a constraint programming approach with ordered fuzzy numbers. That allows coping with nonlinearity of system's characteristics and computational accumulation of inaccuracies generated by the uncertainty of travel parameters. The proposed approach is evaluated on several instances, including real-life scenarios. The computational experiments have shown that the developed model is capable of providing feasible plans of UAVs' mission in an online mode.