Airline Fleet Allocation for Unprofitable Flight Deletion Project

The repository focuses on Airline Fleet Allocation with a primary emphasis on unprofitable flight deletion. Within this repository, you will find code for solving this problem using the Gurobi solver and an improved Dijkstra algorithm to efficiently remove non-profitable airline cycles.

Files

• flight-loop_gurobi.py: Gurobi solver code for solving the airline fleet allocation problem
• preprocess.py: preprocesses the data into graph
• optimization.py: improved Dijkstra algorithm to efficiently remove non-profitable airline cycles
• main_optimization.py: main file for running the optimization

Notes

• While the enhanced Dijkstra algorithm offers improvements, it may not consistently deliver optimal performance. This limitation arises when the algorithm encounters scenarios where two vertices lack a direct connection within the optimal solution. For optimal results, I recommend utilizing the flight-loop_gurobi.py script for small-scale datasets, as I have not yet fine-tuned the enhanced Dijkstra algorithm for speed. Please refer to the Analysis_algorithm.md document for detailed information.

Update 20230916

• It is essential to note that the improved algorithm focuses on pruning flight loops rather than optimizing for maximum profits. Please refer to Analysis_algorithm.md for detaied information.