Scikit-decide is an open-source AI framework for reinforcement learning, automated planning, and scheduling that enables users to describe decision-making problems once and automatically match compatible solvers. The framework, initiated at Airbus with contributions from ANITI, TUPLES, and ANU, provides a growing catalog of domains and solvers including RL algorithms, planning solvers, and search methods, with extensibility to wrap state-of-the-art external domains and solvers.
A Boeing 787 flying from Newark to Rome requires about $68K in jet fuel, but adjusting flight paths for wind conditions could save thousands. Scikit-decide, an open-source reinforcement learning framework combined with fuel consumption models from Delft University and NOAA wind data, offers an alternative to commercial solutions like Jeppesen for optimizing flight paths and airline operations.