Abstract | The current European ATM system offers great flight planning flexibility, which is very tactical. Moving towards a more strategic way of planning will be acceptable as long as a viable compromise between flexibility and predictability can be achieved. The project ADAPT (Advanced prediction models for flexible trajectory-based operations) has proposed strategic models to predict the volume, flexibility and complexity of traffic demand considering both individual flights and network infrastructure (i.e. sectors and airports). The aim is to enable early flight information sharing in order to identify potential network bottlenecks and the degree of flexibility of all flights. The project findings are based on a mathematical programming model that quantifies flexibility for each flight trajectory. We term this flexibility measure time windows. Time windows are time intervals around each sequential operation (departure, arrival or entry into a sector) of a flight. As long as the flight operation is performed within the time window, the flight will not cause disturbances (i.e., delay) to any other flight in the system, at any time. If a flight has to be performed in a highly congested environment with a number of interdependent flights, a ‘small’ delay may cause a large downstream effect. It follows that such flights are constrained to operate closely to their assigned times, and we refer to them as constrained. On the contrary, a flight is unconstrained when operated in a non-congested area where the same amount of delay may not have any impact other than the delay on the flight itself. In other words, should an unconstrained flight depart ‘slightly after’ the assigned time, it will not cause disruptions in the system. Thus, the duration of a time window is a measure of the flexibility that can be granted to perform the flight operation: the longer the duration of the time window, the greater the flexibility, of course. Since constrained and unconstrained flights may coexist at the same time in the network, the duration of time windows may vary among flights. Therefore, the ADAPT model can be used in what-if scenarios testing the decision-making processes of AUs and ANSPs. Once time windows are assigned to flights, it is possible to identify the elements of the planned network configuration (i.e., sectors or airports) that are going to be saturated. Thus, already at the strategic/pre-tactical level, an indication of the flexibility or constraints imposed on flights, and saturated network elements can be obtained. Furthermore, it is also possible to define a criticality index of the saturated elements, which indicates to what extent time windows can be enlarged should a rise in capacity of the element occur. Having the information on the saturated sectors, and their criticality index, ANSPs could take mitigation actions to improve the situation on the day of operations. For example, a supervisor having one or two saturated sectors, both with the low criticality index, might decide that the planned configuration is good enough as even if the capacity ends up being violated it will be for a small number of flights, which in many cases is what already happens in every-day operations. However, if there are few sector-hours within an area control center with high criticality indexes, the supervisor might decide to change the configuration into a one that brings more capacity. Further, this change can be inserted into time window model to re-run it and check what impact it would have on this particular airspace, and the entire network. Computational experiments are run on real data (traffic, airspace configurations, etc.), and on a large instance (around 30 000 flights, 230 airports, 1500 sectors and 1440 time periods). Their results are encouraging and highlight the opportunity to further explore ATM strategic aspects in order to better manage the system on the day of operations. Acknowledgements. The work presented here is a result of the ADAPT project. This project has received funding from the SESAR Joint Undertaking under grant agreement No. 783264 under European Union’s Horizon 2020 research and innovation programme. The opinions expressed herein reflect the author’s view only. Under no circumstances shall the SESAR Joint Undertaking be responsible for any use that may be made of the information contained herein. |
---|