We were inspired by the increasing number of global wildfires and the urgent need for more intelligent, quicker, and more effective ways to respond. Witnessing the devastating effects on communities, the environment, and emergency response infrastructure, we wanted to explore how quantum-inspired optimization could be used to improve real-world disaster management—starting with firefighting resource allocation.
YQuantum 2025 is a wildfire response optimization tool that applies quantum-inspired algorithms to deploy firefighting crews to currently burning fire areas in terms of distance, severity, crew capacity, and urgency. It addresses all areas of fire while optimizing efficiency and minimizing response time and resource consumption. We used Python with D-Wave's Ocean SDK and neal (Simulated Annealing) to define and solve a QUBO (Quadratic Unconstrained Binary Optimization) problem.
Fire and team data were generated randomly to simulate a real-world scenario, e.g., GPS coordinates, intensity levels, and resource capacities. We computed distances using NumPy and calculated scores for both team ability and fire severity. Our objective function and constraints were applied to the QUBO model, which was then optimized with a simulated quantum annealer.
It was tricky to balance constraints and objective weights—ensuring fires are all included without having unrealistic team allocations. Choosing right scaling and normalization for scores like severity and ability so that the optimizer behaved rationally. Debugging QUBO logic and translating real-world priorities into a mathematical model were more challenging than expected.
Effectively modeled a real-world issue in a quantum-compatible form. Created a working simulation that assigns meaningful team-to-fire allocations with simulated annealing. Developed the model so that it was flexible enough to expand with real-world data in the future.
What we learned - How to express optimization problems in QUBO form. The strengths and weaknesses of current quantum-inspired solvers. How small adjustments to weights and constraints affect solution behavior immensely. The potential of quantum computation in disaster management and logististic. Include real-time fire data (e.g., from drones or satellites). Model time-varying fire spread and incorporate dynamic team reassignment.
