Increased Levels of Automation for the ATM Network As an applications oriented research topic, proposals must describe a specific improvement to operations and explain how it is proposed to undertake an initial validation. The improvement to operations can be either a new operational improvement or a new technical enabler (e.g. an improvement to traffic prediction and execution of the plan that would improve the performance of current network operations without changing the current network management procedures).Proposals must demonstrate how their work would go beyond what is currently under research in IR in Wave 1 (PJ07, PJ08 and PJ09) and in Wave 2.The scope of this sub-work area covers the following aspects:Application area 1: Digital Network Management OperationsThe research activities shall explore improvements to the Network Management (NM) function based on digital technologies such as:The potential use of new data sources (big data), machine learning algorithms (including neural networks), AI based decision support tools, behavioural economics, improved market modelling, complexity science, etc. to support network operations e.g. models and methods for improving demand, flow and complexity forecasting and resolution;The use of Big data analysis, machine learning and digital-twin techniques for better planning the actors (controllers, FMPs, AUs) reactions to potential operational improvements based on the emerging trends (e.g. incentives, etc.);The development of methodologies to analyse, quantify and manage the effects of weather and other uncertainties on the network for all phases of flights and in particular on: Trajectory planning (including MET forecasts evolving over time), aircraft performance, etc.;Storm avoidance (including reduction of thunderstorm prediction uncertainty);Sector demand and capacity balancing using the best available plan of action. The better consideration of airport events in network traffic prediction that minimises network disruption;Better consideration of the diversity in data quality from different ATC centres/airports and identify where improvements would bring the biggest gain for operations;Innovative DCB resolution algorithms, e.g. using radically different algorithms to what is used today, e.g. building on previous exploratory research project OPTIFRAME or using alternative approaches, or working on fine-tuning today’s methods, e.g. by considering additional inputs.The use of machine learning for: The identification and prediction of: major traffic flows, complexity assessment, calibration of airspace/sector capacity, flight delays, estimated arrival and overflight times, etc. with the objective of reducing NM capacity buffers;To improve the handling of AU priorities/preferences;To improved disruption management; The use of machine learning and/or advanced visual analytics for DCB decision support tools and automation e.g. hotspot resolution;The adaptation of applications that use models/techniques that are already applied for uncertainty management in other domains.The development of probabilistic approaches based on historical data mining techniques. Application area 2: Improved Integration of Airline Operations into the NetworkThe research shall address potential improvements of airline operations based on the use of advanced digital technologies, e.g. big data, machine learning algorithms, AI, IoT, behavioural economics, improved market modelling, complexity science, etc. such as:The potential use of new data sources (big data), machine learning algorithms, AI based decision support tools, etc. to support airline decision making in disruption scenarios in order to improve the resilience of the system;The development of new tools for improving the integration of airline operations into the network, in order to mitigate the impact of disruptions on the overall ATM network and/or improving operations in nominal conditions (e.g. earlier update of TTOT, better adherence to TSAT, more accurate turn-around time planning, more accurate 4D trajectory calculation for the eFPL by using AI to improve predictions, etc.).The identification of innovative applications to improve the collaboration between AOCs and Network management function and ATC, e.g. to support the involvement of AOC’s that track flights in trajectory revision, in particular for long-haul flights, or to facilitate the inclusion of airline preferences and priorities in the DCB processes or sequencing processes beyond what is already covered in IR Wave 1 and the scope of IR Wave 2 (candidates solutions 38, 39 and 40). The research may build on previous SESAR ER projects using complexity science to better understand delay propagation (e.g. NEWO, TREE) or use entirely new approaches.Application area 3: CASA EvolutionThe Computer-Assisted-Slot-Allocation (CASA) algorithm is used by the Network Manager to allocate departure slots. The Airline Operator (AO) files a flight plan and requests a slot. NM takes into account all the regulation requests from FMPs and allocates some aircraft a delay for entering the regulated area based on the principle of ‘First Planned - First Served’. On this basis, the system calculates the Calculated Take-Off Time (CTOT), which is transmitted to the concerned AOCs and to the control tower at the aerodrome of departure. On top of this basic process, there are a number of compensation mechanisms that take into consideration modifications to the flight plan, late received flight plans, etc., as well as for the cases when a flight crosses more than one regulated area.Proposals shall include activities that propose areas for the evolution of the current CASA and constraints reconciliation algorithm and slot allocation processes in order to improve efficiency and reduce the adverse impact of multiple regulations affecting the same flight or flows, e.g. by finding less constraining ways to handle flights that cross several regulated areas than the current most penalising regulation criterion., or facilitating a more collaborative approach between ANSPs by better allocating CASA delay minutes between ANSPs in the case of regulations that are not formally linked.Proposals must demonstrate in-depth knowledge of the CASA baseline and explain how they plan to assess the potential benefits of the proposed evolutions against it; the use of a non-regulated scenario as reference for comparison is not acceptable.The research may include the use of big data and machine learning to identify best practices regarding regulation strategies for particular traffic load patterns based on historical data. The analysis of historical data on regulation strategies should be complemented by the results from network simulations to develop optimized strategies for the most frequent traffic load situations in the European ATFCM network.The research may build on previous SESAR ER research (e.g. OPTIFRAME) or propose entirely new approaches. Please note that PJ09 results are expected to become publicly available at the end of 2019. Projects planning to work in this area must reserve effort for analysing PJ09’s output and incorporate it in their research. Please note also that proposals must show how they go beyond the IR Wave 2 scope of candidate solution 47.Application area 4: More automated ATFCMThe research shall identify mechanisms to allow the introduction of higher levels of automation in the coordination of DCB actions at the pre-tactical and tactical levels. The network is not sufficiently robust or resilient to react to significant perturbations e.g. meteo, industrial actions, etc.; this imposes unplanned/additional costs on airlines, which have a huge impact on airlines’ annual revenue. Airspace users’ full participation through their flight operations centres (FOC/WOC) into ATM collaborative processes is essential to minimise impacts of deteriorated operations for all stakeholders, including airspace users themselves. An improved recovery process offering more flexibility to accommodate AUs’ changing business priorities and equity in the ATM system is therefore needed. The proposed solutions under this topic aim at improving:Safety, thanks to better anticipating and managing demand-capacity imbalances;Robustness and resilience of the network to perturbations;Efficiency thanks to a better monitoring of the DCB measures and network performance and the implementation of corrective actions;Cost-efficiency: an improved network management will allow better planning of ATM resources, as well as making better use of existing capacities, which would lead to reduced ATC and airport costs;Flexibility: common awareness to all stakeholders of the network situation and access to opportunities in case of late changes in capacity or demand.