New UAV scheduling model helps drones fight wildfires with degradation-aware routing

A new paper from researchers Erdi Dasdemir, Esther Jose, and Rajan Batta tackles the intertwined problem of scheduling and routing for wildfire-fighting UAV fleets. The core idea is that fires degrade over time—if you don't act quickly, the fire grows, reduces the area's value, and spreads to adjacent zones, creating new jobs that demand response. The team formalizes this as a scheduling and routing problem with degradation-triggered job arrivals, where each job has a time window and a default reward that decreases as intervention is delayed. Their mixed-integer programming model maximizes total remaining reward, directly aligning with the proverb "a stitch in time saves nine."

The authors also propose a hybrid model based on dynamic constraint generation to overcome scalability issues, making it feasible for realistic wildfire scenarios. They validate their approach through computational experiments and a case study, showing practical viability. Notably, the codebase is open access, encouraging further research and implementation. While the paper focuses on aerial forest firefighting, the underlying framework could extend to other emergency response domains like disaster relief or infrastructure maintenance, where jobs degrade and spawn new ones. For tech professionals, this demonstrates how advanced optimization and AI-driven routing can turn drone fleets into proactive, value-preserving response systems rather than reactive tools.