**2. User Requirements**

Assimilated nowcasting data of convective areas should be visualized and used for arrival scheduling to support ATCOs and reduce their workload in these kinds of nonnominal conditions. For the definition of user requirements, a survey with nine ATCOs was conducted within the project to obtain an overview about their demands and preferences in order to enhance the acceptance and usability of adverse weather visualizations [26]. The results of the survey facilitate suggestions to improve the presentation of adverse weather areas (such as convective cells) on a traffic situation display for aircraft guiding and 4D flight trajectory calculation with target times for significant waypoints [27]. These insights were used for the development of an individually configurable primary display with adverse weather presentation possibilities in order to increase acceptance of the support system by controllers.

#### *2.1. The Requirements Inquiry*

A twelve-page questionnaire divided into ten different structured sections was provided to the participants. There were questions about display variants, in which the participants could answer using a seven-point Likert scale, and questions that could be answered with either "yes" or "no". The latter ones had the additional option of specifying

one's answer or limiting its validity and scope via an associated comment field. After general questions about the person, five presentation variants of convective cells were introduced, for which the participants were asked to give an estimate of the support quality of the display. The participants were also asked to provide a possible order regarding their personal acceptance of the display modes. The third section concerned to what extent the activation of the display should be automated or manual and whether they wanted a weather display at all. After questions about dynamic and static weather visualizations, as well as questions about a possible forecast period when integrating nowcasts, the final section addressed ideas about the use of additional symbols on the aircraft labels on the radar display to mark aircraft affected by adverse weather and also focused on the integration of additional safety zones around measured and predicted convective cells in the airspace.

There were, in total, nine survey responses that were returned, though not all controllers answered the complete set of questions. The number of participants was not extensive. However, all of them were professional and licensed air traffic controllers. Hence, their answers provide meaningful and valuable insights into the requirements for adverse weather visualizations on traffic situation displays. The respondents were male and their ages were uniformly distributed between the considered age groups (Figure 1).

**Figure 1.** Age distribution and professional experience of the participating controllers.

#### *2.2. The Requirement Wishes of the Air Traffic Controllers*

Although the integration of adverse weather is intended as a support for advanced scheduling and sequencing, it was usually perceived by controllers as a taking over of additional responsibility. Some of the respondents emphasized that it is the responsibility of the pilot to ensure a safe flight, including evasive maneuvers due to adverse weather [26]. One of the most common comments regarded a possible overload of the controller's display, either with additional information or with its colored presentations. The next point was that the controller's display may only represent the actual state at any time and not a forecasted one. On the other hand, the respondents stressed that displaying severe weather would be beneficial for better planning and less interference with traffic flows. Due to the large differences in the reactions of the pilots, respondents assumed that meaningful and realistic categorization with regard to the dangerousness of a weather situation was impossible. Experience shows that one aircraft can fly on the left side of a convective cell, the next one on the right side, and the last one through some severe weather area depending on the experience of the pilot and the interpretation of available onboard weather radar. Generally, this supporting tool is conceivable for a planning controller. For executive controllers, there might be a risk of visual overload on the traffic display. In principle, it was noted that it is a very good approach to show current weather data in the radar traffic image. However, there should be the possibility of manually switching this information on or off with a button. This should be used by ATCOs if necessary for better planning of traffic in relation to sequence creation. The represented information should always match the actions of the controller. For instance, airspace being "usable" or "not usable" means that a display

with two possible states should be used without detailed information about meteorological airspace conditions.

The essence of the results from the requirements analysis for the development of the controller support system in the project can be summarized very well with the phrase *"less is more"*. Overall, eight out of the nine controllers that participated in the survey welcomed a way of quickly and directly accessing weather information that is relevant to them. One of their basic requirements is that any type of information, whether it is provided graphically or numerically, must be able to be activated and deactivated quickly and easily on the display. These requirements build the basis for further developments in the SINOPTICA project that can visualize adverse weather on a controller's traffic situation display.

#### **3. AMAN Air Traffic Controller Support**

Arrival Manager (AMAN) systems have been developed and deployed in Europe over the course of the last 25 years. They are primarily designed to provide automated sequencing support for ATCOs handling traffic arriving at an airport by continuously calculating arrival sequences and times for flights while considering the locally defined landing rate, the required spacing for flights arriving to the runway, and other criteria. The AMAN has to generate flyable 4D trajectories with target times for all significant waypoints, including runway thresholds.

During the last twenty-five years, the DLR's Institute of Flight Guidance in Braunschweig has developed arrival management systems for different kinds of scientific applications at various international airports. The latest version of DLR's previously developed arrival manager tools "COMPAS" [28] and "4D-Planner" [29] is called the 4-dimensional Cooperative Arrival Manager (4D-CARMA). Both previous versions were the result of research projects in close cooperation with the Deutsche Flugsicherung GmbH (DFS). Considering different constraints, such as weight classes, runway separation criteria, or runway allocation, 4D-CARMA uses radar data and additional information, such as the flight plans of all arriving aircraft, for sequencing and trajectory calculation. AMANs are pure suggestion systems and have a planning horizon of around one hour. They ease AT-COs' tasks by taking over the particularly difficult planning and optimization of approach sequences while considering all given constraints. This technical support in approach planning can have a clearly positive influence on the effectiveness of ATCOs' work since approaching aircraft are integrated at an early stage, the required distances are precisely considered, throughput is slightly increased, and approach trajectories are more direct and thus shorter [30]. This is especially true when the decision support systems are combined with speech recognition [31]. In recent years, developments in Arrival Managers have gone in two main directions: On the one hand, the planning horizon has been systematically extended to several hundred miles, giving aircraft a precise target time for the threshold and, thus, a position in the landing sequence already in the en-route phase [32]. On the other hand, attempts are being made to use machine learning (ML) methods to build pilot support systems that are based on trajectories that have actually been flown and can thus better represent the typical actions of controllers and pilots in certain situations than classic deterministic algorithms [33,34]. With the help of ML, attempts are also being made to support more environmentally friendly approach planning and thus further reduce the environmental impact of aviation [35].

The first arrival managers already developed the systematic base for ATCO support, and this has not changed in principle. Accordingly, the tasks of an AMAN can be divided into different levels: "Sequence Planning" calculates optimal landing sequences based on airspace structure, current air traffic situations, and performance criteria for all aircraft in the airspace. "Trajectory Calculation" creates optimal 4D routes for every individual aircraft to fulfil the planned sequence. "Advisory Generation" deduces the required instructions from air traffic controller to pilot to follow the calculated trajectory, and "Conformance Monitoring" tracks if aircraft follow the planned trajectories. In order to support adverse weather avoidance, the assistant functionalities of an AMAN have to be adapted and new

assistant functionalities need to be provided. The most important elements of an AMAN capable of supporting severe weather area diversions are conflict detection between routes and weather areas, alternate route finding, sequence calculation, and trajectory generation.
