#	Pagina
attuale pagina	/open-h2020/projects/213200/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - ADAPT (Advanced prediction models for flexible trajectory-based operations)

Teaser

Summary

The scope of the ADAPT project is to propose a set of methods and tools at the strategic and pre-tactical level of network management that is conducive to the trajectory-based operations, which clearly demonstrates the flexibility, information exchange responsibilities, and benefits for all the stakeholders. In particular, the aim of the ADAPT project is to design and test models and metrics that enable strategic planning (early information sharing), by providing the information on flight flexibility and network hotspots, which can eventually be integrated into the Network Operations Plan (NOP) and serve as a basis for stakeholder collaboration. Previous research mainly targeted just the opposite by reducing flexibility with the aim to increase the predictability of the ATM system. Often, this approach resulted in negative consequences on the performance in other areas, like direct operating costs of airspace users (AUs). With this in mind, the ADAPT project proposes a solution to enhance the predictability of the ATM system, while at the same time preserving flexibility for AUs.

The European ATM system offers a high level of flight planning flexibility, as only the final flight plans need to be submitted several hours before departure. On the one hand, this allows AUs the possibility to account for previously uncertain factors like weather forecasts, and thus create flight plans that are most convenient for the day of operations. On the other hand, this flexibility makes ATM system less predictable, resulting in costs due to flow measures, and under-utilisations from a mismatch between available ATM capacity and traffic demand. When creating and subsequently submitting an initial flight plan, AUs do not have the information on airspace nominal capacities and do not need to consider it. Thus, a precise traffic load on the airspace network is only known on the day of operations, while the capacity provision (e.g. staffing levels) is usually planned starting more than a year ahead and is updated as time progresses. On the day of operations in cases when available airspace (and airport) capacity cannot accommodate planned air traffic, the ANSPs and Network Manager agree on the Air Traffic Flow Management (ATFM) measures to reduce the demand on the congested parts of the network. ATFM measures impose delays on flights crossing congested network volumes (AUs can re-route around the area in question). These delays and deviations are very costly to AUs (e.g. estimated by EUROCONTROL to be more than 1B euro in 2014).

A part of the capacity-demand imbalances, and consequent delays, is caused by unforeseen factors, like weather. The other part is triggered by the lack of accurate information exchange at the required time horizons. ANSPs plan capacity provision months ahead (information not used by AUs), while the more accurate traffic demand is known on the day of operations, leading inevitably to congestion in parts of the network. In order to close this gap, the ADAPT project develops a set of linked models that exploit the past and current information in the early planning stages, for mutual benefit of all stakeholders. The past and early-shared flight-route information can be used to find the best distribution of the proposed flights that respect the capacity of the proposed network configuration(s). Next, the trajectories obtained in this process are assigned a measure of flexibility, that we term time windows (TWs). TWs measure the flexibility of a flight in terms of time slack available to a flight, within which no capacity problems are expected to be created along the agreed trajectory. Apart from TWs assigned to flights, the saturated elements of the planned network configuration (i.e. sectors or airports) are also identified. Thus, already at the strategic level, an indication of the flexibility or constraint on flights and saturated network elements can be obtained and included in the rolling NOP. This approach to strategic

Work performed

The first phase of the development of the ADAPT solution, which is the formulation and implementation of a deterministic model (European Strategic Flight Planning (ESFP) model) to define flight trajectories and associated TWs at the strategic level has been achieved. The ESFP model builds on two integer programming models. The first model assigns a trajectory for each scheduled flight, in such a way that nominal capacities of the network are respected. When all flights have a trajectory and departure time assigned, these become inputs of a second integer programming model, called Time Window (TW) model. This model uses departure times as the starting position of each TW, and the objective is to guarantee the largest flexibility by maximising the total duration of all TWs, i.e., the sum of the duration of all individual TWs. The output of this second model are the trajectories, assigned TWs and the saturated elements of the network. Three variants of the TW model which differ in the approach to accounting capacity have been developed and tested on a day of real air traffic data, encompassing the entire European airspace.

Furthermore, the main sources of uncertainty around the tactical trajectory evolution have been identified, in order to apply them in the tactical assessment of the ADAPT strategic solution. A stochastic model that assesses strategic TWs by considering probabilistic aircraft trajectory times has been developed, and meteorological uncertainty (wind speeds) has been applied and tested on a small example.

A set of metrics able to detect network hotspots, and geared toward the assessment of the effectiveness of ADAPT strategic solutions in the tactical setting, has also been developed.

Final results

The state-of-the-art in the operational ATM management in Europe is largely tactical. AUs are offered a high level of flexibility in the trajectory planning as the flight plans need to be submitted only tactically, thus allowing to consider the latest information available (e.g. weather, aircraft availability) and plan the most convenient trajectories. When planning a trajectory, AUs do not have the information on airspace nominal capacities and do not need to consider it. The consequence of this practice is the fact that the detailed traffic demand on the airspace network is known only on the day of operations, while the capacity provision is typically planned a year in advance and updated a few days before. This points to a significant tactical-strategic gap between the operations planning horizons of AUs and ANSPs, which the ADAPT solution aims to bridge. In fact, to the best of our knowledge, this is the first time a strategic planning that can quantify flexibility of flights and severity of saturated element is proposed.

In the second part of the project, the assessment of ADAPT results both in strategic and tactical settings will be performed, and the results visualisation tool will be developed. These assessments will collect a range of indicators and metrics in order to analyse the benefits to various stakeholders. Having the information on trajectories, flexibility and saturated network elements would be a step beyond the state-of-the-art, which would lead to a significant socio-economic impact as it could diminish the delays and related costs in European ATM.