Dynamic Reroute Planning under Uncertain Severe Convective Weather

In order to solve the reroute planning problem with uncertain severe convective weather, a state transition matrix was built using the uncertain measure of forecasting state and the random error of extrapolating position for rainstorm cloud cluster. The matrix was then used to predict the time-varying delay conditions of the reroute planning network. On this basis, a dynamic reroute planning model, with the least expected rerouting cost as its optimized objective, was developed under the constraints of path continuity and maximum penetrating risk, and solved by the generic algorithm subsequently. Finally, two kinds of weather scenarios with 16 initial states of cloud cluster were set up in a case study to verify the proposed model. Simulation results show that the rerouting paths will strategically penetrate clouds with unsafe initial state or detour clouds with safe initial state, by predicting the uncertain changes of cloud cluster states in uncertain weather scenarios. In addition, with the dynamic rerouting strategy under an uncertain scenario, the detouring distance decreases by 40% and the penetrating count decreases by 30.8%, compared with those under a certain weather scenario. Therefore, both safety and economy of rerouting paths are improved effectively.