Improving profitability through schedule optimization
We developed proprietary algorithms by combining in novel ways several operations research techniques, such as mixed integer programming, heuristics, very large-scale neighborhood search and parallel computing, and we demonstrated that clean-sheet flight scheduling was mathematically possible. We then modified our algorithms to enable incremental optimization of existing flight schedules.
We also combined flight scheduling with crew scheduling and optimized the problem holistically, and the system produced flight schedules that reduced crew costs significantly. We combined our optimization algorithms with an advanced decision support system, where users could define constraints, prepare scenarios, run different optimization modes and view schedules using advanced visualization tools. We hosted SkyMAX in the cloud to enable high performance, reliability and scalability.
- Maximized profitability: The system balances network profitability and operational constraints to find the best times (for both nonstop and connecting passengers) and allocates the right aircraft to each flight leg in the schedule.
- Improved aircraft utilization: Planners can develop more effective schedules that increase aircraft utilization by simultaneously adding a combination of fleeting and timing changes to the schedule.
- Faster schedule generation: Because the system rigorously builds flight schedules that satisfy all operational constraints, manual efforts are minimized.