Resilient Airline Scheduling and Operations
Optimize on-time performance based on delay risk prediction scheduling until day of Ops
Airlines should already consider delay risks in scheduling to increase punctuality
Flight delays are very costly and annoy passengers. Nevertheless todays flight schedulers have little information on delay risks and on typical disturbances when creating a schedule. Even so the schedule is the base for a punctual operation, punctuality aspects are barley considered in the scheduling phase.
To increase on-time performance with the same or even higher productivity, airlines should address delay risks already in scheduling. A punctual schedule structure is much easier build up without scarifying the airlines profitability at an early planning phase. In this way, the airline has much more planning flexibility e.g. to adjust airport slots or arrange rotations, than closer to day of operation without negatively effecting passengers, crews and other airline processes.
Data science research project based on operations data of more than 10 years
To build up a schedule which is more resilient in respect to delays we set up a data science research project as joint project of Lufthansa Systems and German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt, DLR). The project was co-financed by the German Federal Ministry for Economic Affairs and Energy (BMWi) as part of the aviation research LUFO V program.
Provide scheduling with reliable information on delay risks
Based on operations data of more than 10 years we developed a model to forecast delays. Thus we provide block- and ground times that balances productivity and punctuality. These scheduling parameters are as long as needed to be punctual, but are not too generous thus influencing negatively the profitability of an airline. In addition, typical constellations leading to systematic delays are identified and can be buffered by applying appropriate buffer in the schedule.
Use delay mitigation optimizer to reduce delays automatically
We also developed and tested different delay mitigation optimizers using meta-heuristic methods. By changing only the aircraft routing we can reduce the predicted delay by up to 15 %. The delay risks of a schedule can be significantly decreased further by applying minor time shifts of 5 minutes at some legs.
