Future Scenarios and Relevant HF Concepts

EXPLORATORY RESEARCH

Future scenarios and relevant HF concepts

D1.1 STRESS Grant: 699381 Call: H2020-SESAR-2015-1 Topic: Sesar-01-2015 Automation in ATM Consortium coordinator: Deep Blue Edition date: 22nd November 2016 Edition: 00.02.00

EDITION 00.02.00

Authoring & Approval (DBL)

Authors of the document Name/Beneficiary Position/Title Stefano Bonelli/DBL Project Coordinator Géraud Granger/ENAC WP1-WP3 Leader Paola Tomasello/DBL WP5 Leader Martina Ragosta/DBL WP6 Leader Uğur Turhan/AU WP4 Leader Birsen Açikel/AU Project member Ali Ozan/AU Project member Gianluca Borghini/UniSap Project member Gianluca Di Flumeri/ UniSap Project member Fabio Babiloni/ UniSap WP2 Leader Pietro Aricò/ UniSap Project member Fabrice Drogoul/ECTL Project member

Reviewers internal to the project Name/Beneficiary Position/Title Date Stefano Bonelli/DBL Project Coordinator 20/09/2016 Francesca Piazza/DBL Project Assistant 20/09/2016 Francesca Piazza/DBL Project Assistant 12/10/2016 Stefano Bonelli/DBL Project Coordinator 14/10/2016

Approved for submission to the SJU By — Representatives of beneficiaries involved in the project Name/Beneficiary Position/Title Date Stefano Bonelli/DBL Project Coordinator 14/10/2016 Géraud Granger/ENAC WP1-WP3 Leader 14/10/2016

Rejected By - Representatives of beneficiaries involved in the project Name/Beneficiary Position/Title Date

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Document History

Edition Date Status Author Justification 00.00.01 21/07/2016 Draft Stefano Bonelli Proposed table of contents 00.00.02 05/08/2016 Draft Géraud Granger Toc revised and responsibilities assigned 00.00.03 29/08/2016 Draft Martina Ragosta Added “Chapter 3 – Section 3.1” 00.00.04 09/09/2016 Draft Martina Ragosta Added “Chapter 4 – Section 4.1” 00.00.05 20/09/2016 Quality Check Francesca Piazza First light Quality Check 00.00.06 20/09/2016 Internal review Stefano Bonelli Internal review #1 00.00.07 22/09/2016 Post 1st review draft Martina Ragosta, Paola Addressed issues from Tomasello review #1 and first quality check 00.00.08 26/09/2016 Draft Gianluca Di Flumeri, Added “Chapter 4.3 – Gianluca Borghini, Section 4.3.2” and the Pietro Aricò, Fabio related subsections Babiloni 00.00.09 27/09/2016 Draft Géraud Granger Added “Chapter 4.2 – Section 4.4.2” and the related subsections 00.00.10 27/09/2016 Draft Paola Tomasello Added “Chapter 3 – Section 3.2” and the related subsections 00.00.11 27/09/2016 Draft Uğur Turhan, Birsen Added “Chapter 4.2 – Açikel, Ali Ozan Section 4.4.1” and the related subsections 00.00.12 29/09/2016 Draft Martina Ragosta, Paola Harmonised different Tomasello partners’ contributions and added Introduction and “Chapter 2 – Section 2.3” 00.00.13 03/10/2016 Draft Paola Tomasello Added “Chapter 3 – Section 3.3” and the related subsections 00.00.14 04/10/2016 Draft Gianluca Di Flumeri, Reviewed “Chapter 4.3 – Gianluca Borghini, Section 4.3.2” and the Pietro Aricò, Fabio related subsections Babiloni 00.00.15 05/10/2016 Consolidated Draft Martina Ragosta, Paola Harmonised and revised

EDITION 00.02.00

Tomasello the whole document 00.00.16 06/10/2016 Consolidated Draft Géraud Granger Added “Abstract”, “Executive summary” and revised the whole document 00.00.17 10/10/2016 Consolidated Draft Martina Ragosta Addressed minor comments and added references 00.00.18 11/10/2016 Consolidated Draft Paola Tomasello Revised “Abstract”, “Executive summary” and added “Results” 00.00.19 12/10/2016 Quality Check Francesca Piazza Second Quality Check 00.00.20 13/10/2016 Post 2nd review draft Martina Ragosta Addressed issues from second quality check 00.00.21 14/10/2016 Proposed draft for PC Paola Tomasello and Harmonised and revised approval Martina Ragosta the whole document - ready for PC approval 00.01.00 14/10/2016 Final release Stefano Bonelli Deliverable approved for submission to the SJU 00.02.00 22/11/2016 Updated document Paola Tomasello Implementation of SJU comments

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

STRESS HUMAN PERFORMANCE NEUROMETRICS TOOLBOX FOR HIGHLY AUTOMATED SYSTEMS DESIGN

This deliverable has received funding from the SESAR Joint Undertaking under grant agreement No 699381 under European Union’s Horizon 2020 research and innovation.

Abstract This is the first deliverable of the STRESS project (human performance neurometricS Toolbox foR highly automatEd SystemS design).

STRESS intends to support the transition to higher automation levels in aviation, by addressing, analysing and mitigating its impact on the Human Performance aspects associated to the future role of Air Traffic Controllers. To achieve this, STRESS will carry out two validation activities. For each condition tested, the overall human response to the proposed scenarios will be collected, finding the human response configuration associated to specific levels of automation and/or to non-nominal events. Recommendations and guidelines for future systems design will derive as one of the main project expected outcomes.

The current deliverable intends to provide an overview of the potential operational scenarios that will be simulated in the framework of the STRESS validation activities, and to define the Human Factors aspects (with related neurophysiological indicators) that are deemed relevant in shaping the Human Performance in that scenarios. The methodology used for selecting the most appropriate set of potential validation scenarios includes the following criteria::

 Compliance with the project objectives  Compliance with the lessons learned derived from the extensive literature review of SESAR future concepts  Compliance with the initial operational needs collected during preliminary focus group with controllers  Validation platforms capabilities  Capability to measure the controllers mental status through neurophysiological indicators As outcome, the deliverable presents the possible operational environments, the set of highly automated systems and some potential non-nominal situations that could be simulated. The set of possible tools that will be used to measure the neurophysiological indicators of controllers mental status are described as well.

EDITION 00.02.00

Table of Contents Abstract ...... 5 1 Executive Summary ...... 7 2 Introduction ...... 9 2.1 Purpose and Scope of this Document ...... 10 2.2 Deliverable Structure ...... 10 2.3 Acronyms and Terminology ...... 11 3 Review ...... 18 3.1 Future Scenarios ...... 18 3.2 Automation in future scenarios ...... 24 3.2.1 Review on automation: frameworks and models ...... 24 3.2.2 Classifying automation: the Level of Automation Taxonomy ...... 28 3.2.3 Automation support to ATCOs and pilots tasks ...... 35 3.3 HF relevant concepts in future scenarios ...... 36 3.3.1 Stress ...... 36 3.3.2 Attention ...... 38 3.3.3 Mental workload ...... 40 3.3.4 Cognitive control ...... 40 3.4 Human Performance ...... 41 3.4.1 Automation impact on Human Performance ...... 41 4 Scenarios selection ...... 45 4.1 Method for STRESS scenarios selection ...... 45 4.1.1 Compliance with STRESS objectives ...... 45 4.1.2 Compliance with the literature review ...... 46 4.1.3 Compliance with the initial operational needs ...... 46 4.1.4 Assumptions and limitations ...... 47 4.2 Environments and platform capabilities ...... 48 4.2.1 At Anadolu University ...... 48 4.2.2 At École Nationale de l'Aviation Civile ...... 61 4.3 Neurophysiological Indicators ...... 68 5 Results ...... 76 5.1 STRESS validation activities ...... 76 5.2 Operational environments ...... 77 5.3 Potential Scenarios ...... 78 5.3.1 Automation ...... 78 5.3.2 Non-nominal scenarios examples ...... 79 6 References ...... 83 7 Appendix 1: AU _TATCA Focus Group ...... 97

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

1 Executive Summary

This is the first deliverable of the STRESS project (human performance neurometricS Toolbox foR highly automatEd SystemS design). STRESS addresses Human Performance in future SESAR scenarios. The European ATM system is expected to face challenging situations, with the growth of air traffic, the increase of its complexity, the introduction of innovative concepts and increased automation. The roles and tasks of air traffic controllers (ATCOs) will change in the future and it is vital to enhance the comprehension of human responses to their role changing, that is, from active control to monitoring of complex situations and managing unexpected system disruptions. ATCOs performance is recognised to be impacted by several aspects such as stress, emotions, attentional resources available, attention focus and so on. In the recent years, the concept of Human Performance Envelope (HPE) has been introduced as new paradigm in Human Factors. Rather than focusing on one individual factor (e.g. fatigue, situation awareness, etc.), the HPE considers their full range, mapping how they work alone or in combination leading to a decreased performance that could affect safety. At the European level, there are projects currently addressing the research goal of monitoring the team performance, including monitoring some of the above aspects. However, most of these research activities focus on pilots and airplane cockpits. In line with this, there is a clear definition of the future scenario for pilots and of the corresponding HP implications, while a corresponding work on the ATCO role is still to be performed. STRESS deals with it. To achieve this, STRESS proposes a research plan based on two validation activities. For each condition tested, the overall human response to the proposed scenarios will be collected through neurophysiological indicators, finding the human response configuration associated to specific levels of automation and/or to non-nominal events. Recommendations and guidelines for future systems design will derive as one of the main project expected outcomes. The current deliverable intends to provide an overview of the potential operational scenarios that will be simulated in the framework of the STRESS validation activities, and to define the Human Factors aspects (with related neurophysiological indicators) that are deemed relevant in shaping the Human Performance in that scenarios.

The methodology used for selecting the most appropriate set of potential validation scenarios includes the following criteria:  Compliance with the project objectives  Compliance with the lessons learned derived from the extensive literature review of SESAR future concepts  Compliance with the initial operational needs collected during preliminary focus group with controllers

EDITION 00.02.00

 Validation platforms capabilities  Capability to measure the controllers mental status through neurophysiological indicators According to these criteria, the initial features of STRESS validation scenarios are defined, as follows:

 Radar environment in Anadolu and En-route environment in ENAC have been chosen as operational environments to be simulated.  the Free Route Airspace could be a possible concept to be studied in the STRESS simulations.  Stress, emotions, workload and level of cognitive control on operational tasks have been recognized to be the most impacted Human Factors issues to be investigated. Neurophysiological measurement tools to assess them include electroencephalography (EEG), eye-tracker and skin-conductance-response measurement tool. Possible non-nominal scenarios examples include label transfer failure, wrong information on the label and STCA failure.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

2 Introduction

Complex adaptive socio-technical systems (CASs) such as Air Traffic Management (ATM) are featured by lots of interaction within the System, its sub-systems and across many organizations. Indeed, ATM is considered complex as it encompasses interactions involving multiple kinds of operators, various dedicated computing systems, multiple regulatory authorities and their performance is deeply influenced by environmental aspects (i.e. weather, organizational variability (e.g. strikes) or system failures). Moreover, the current European ATM System is becoming increasingly dynamic, facing the need to be improved for coping with the growth in air traffic forecasted for next years. Traffic is likely to become more heterogeneous integrating Remotely Piloted Aircraft Systems (RPAS) and possibly civil aircraft with reduced flying crew (also known as single pilot operations). These two aspects require evolutions to increase ATM capacity and to guarantee that safety level is not reduced. One way towards these objectives is to increase automation in the computing part of the ATM. In that case, automation has to be considered as an enabler to improve capacity (maintaining safety) of the ATM System, considering this improvement not from any partial view point, but from an overall System performance perspective [1], [2]. In line with this, dealing with such large scale implementation of increasing automation requires multi-disciplinary approaches, namely electronics and computer science (for the technological part), psychology and human factors (for the human part) and organization sciences (for the organisation part of CASs). Of course as the operator is interacting with the computing system, Human-Computer Interaction (HCI) issues arise and have to be addressed adequately. Beyond that, these CASs are most of the time large scale distributed systems where groups of operators interact altogether to achieve the required functions of the CASs. This adds all the disciplines related to collaborative aspects of Computer Supported Cooperative Work (CSCW) including distributed systems and Human-Computer Interaction (at coordination, production and communication levels). Due to the deep interleaving of these parts of the CAS, their evolution can sometimes be (e.g. under adverse conditions) non-linear making its overall performance hardly predictable. One of the objectives of the STRESS project (human performance neurometricS Toolbox foR highly automatEd SystemS design) is to support the aforementioned transition to higher automation levels, by addressing, analysing and mitigating its impact on the Human Performance. In fact, the roles and tasks of air traffic controllers will change in the future and it is vital to enhance the comprehension of human response to changes in role, monitoring of complex situations, unexpected disruptions. It is also vital to develop tools to investigate such aspects and to monitor in real time controllers’ fitness to the task, anticipating risks and problems.

EDITION 00.02.00

At the European level there are projects currently addressing the research goal of monitoring the team performance. However, most of these projects share a common assumption (and its consequence): - Research is mainly being carried out on pilots and cockpits. This was also highlighted in the OPTICS project (Observation Platform for Technological and Institutional Consolidation of research in Safety) assessment of research progress towards Progress towards the ACARE (Advisory Council for Aviation Research and innovation in Europe) Goals set in Flightpath 2050 [3]; - In line with this, there is a clear definition of the future scenario for pilots and of the corresponding Human Performance (HP) implications, while a corresponding work on the ATCOs role is still to be performed. This is the first deliverable of the project. It provides an overview of the potential operational scenarios that will be simulated in the framework of the STRESS validation activities, and defines the Human Factors aspects (with related neurophysiological indicators) that are deemed relevant in shaping the Human Performance in that scenarios.

The consolidated set of selected scenarios will be presented in deliverable 4.1 Validation Plan. STRESS will plan two validation activities. The first one will validate the links among HF concepts and neurophysiological indicators. For each condition tested, the overall human response to the proposed scenarios will be collected, finding HF concepts configuration associated to specific levels of automation and/or to non-nominal events. The second one will consist in the assessment of the ATCOs performance in high automation nominal and failure scenarios. Its outcome will be the main input for the preparation of the automation design guidelines. 2.1 Purpose and Scope of this Document

As outcome, the deliverable presents the possible operational environments, the set of highly automated systems and some potential non-nominal situations that could be simulated. The set of possible tools that will be used to measure the neurophysiological indicators of controllers mental status are described as well.

2.2 Deliverable Structure

This document is divided into five sections:

 Section 1 is an executive summary  Section 2 defines the structure, scope and purpose of the document.  Section 3 provides future Single European Sky ATM Research (SESAR) concepts review, with a focus on the role of automation. Based on this, it identifies the impacted human factors

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

aspects and how they may impact on the human performance associated to future ATCOs roles.  Section 4 presents the method to select scenarios among the different scenarios available.  Section 5 presents selected scenarios. 2.3 Acronyms and Terminology

The following table reports the acronyms used in this deliverable.

Term Definition

ACARE Advisory Council for Aviation Research and innovation in Europe

ACC Area Control Centre

ADS-B Automatic Dependent Surveillance -– Broadcast

AFL Actual Flight Level

AIP Aeronautical Information Publication

AMAN Arrival MANager

ANS Autonomic Nervous System

ANSPs Air Navigation Service Providers

AP/FD Auto Pilot/Flight Director

APM Approach Path Monitor

APW Area Proximity Warning

ASAS – ASPA Airborne Separation ASsistance – Airborne SPAcing system

A-SMGCS Advanced - Surface Movements Guidance and Control System

ATC Air Traffic Control

ATCO Air Traffic Control Officer

ATM Air Traffic Management

ATS Air Traffic Services

ATSAW - SURF Air Traffic Situation AWareness for SURFace operations

EDITION 00.02.00

Term Definition

ATSPs Air Traffic Service Providers

AU Anadolu

AUTOPACE project facilitating the AUTOmation PACE

BOLD Blood Oxygen Level Dependent

CANSO Civil Air Navigation Services Organization

CASs Complex Adaptive socio-technical Systems

CDG roissy Charles De Gaulle airport

CFL Cleared Flight Level

CLAM Cleared Level Adherence Monitoring

CNS Central Nervous System

CPDLC Controller-Pilot Data Link Communication

CSCW Computer Supported Cooperative Work

CTA Controlled Time of Arrival

CWP Controller Working Position

DHMI The Turkish Air Navigation Service Provider

DMAN Departure MANager

D-SAM Down-linked Selected Altitude Monitoring

DSNA The French Air Navigation Service Provider

E-AMAN Extended Arrival MANagement

EASA European Aviation Safety Agency

ECG ElectroCardioGraphy/ElectroCardioGram

EEG ElectroEncephaloGraphy/ElectroEncephaloGram

EOG ElectroOculoGraphy/ ElectroOculoGram

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Term Definition

ERAM En Route Automation Modernization

ERATO En-Route Air Traffic Organizer

ERD Event Related Desynchronization

ERP Event Related Potential

ERS Event Related Synchronization

E-TML Enhanced-Task Monitoring Load

EUROCONTROL European Organisation for the Safety of Air Navigation

FAA Federal Aviation Administration

FAB Functional Airspace Block

FDP Flight Data Processing

FIR Flight Information Region

fMRI functional Magnetic Resonance Imaging

FMS Flight Management System

fNIR functional Near-InfraRed

fNIRs functional Near-InfraRed spectroscopy

FOS Fast Optical Signal

FRA Free Route Airspace

GAS General Adaptation Syndrome

GSR Galvanic Skin Response

HALA! Higher Automation Levels in Aviation

HCI Human Computer Interaction

HF Human Factors

EDITION 00.02.00

Term Definition

HF High Frequency

HMI Human Machine Interface

HP Human Performance

HPE Human Performance Envelope

HRV Heart Rate Variability

IATA International Air Transport Association

ICAO International Civil Aviation Organization

iFACTS interim Future Area Control ToolS

KPAs Key Performance Areas

LF Low Frequency

LOA Level Of Automation

LOAT Level Of Automation Taxonomy

MABA – MABA Men Are Best At – Machines Are Best At

MEG MagnetoEncephaloGraphy/MagnetoEncephaloGram

MINIMA project MItigating Negative Impacts of Monitoring high levels of Automation

MONA MONitoring Aids

MOTA project MOdern TAxiing

MSAW Minimum Safe Altitude Warning

MTCD Medium Term Conflict Detection

NASA National Aeronautics and Space Administration

Next Generation Air Transportation Systems (the programme which NextGen defines the ATM Research and Development activities and Projects within USA)

Observation Platform for Technological and Institutional Consolidation of OPTICS project research in Safety

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Term Definition

PDRC Precision Departure Release Capability

PEL Planned Entry Level

PFC Pre-Frontal Cortex

PNS Parasympathetic Nervous System

PPC Posterior Parietal Cortex

R&D Research & Development

RA Resolution Advisory

RAM Route Adherence Monitoring

RNAV aRea NAVigation

RPAS Remotely Piloted Aircraft Systems

RT Radio Transmission

SAM Sympathetic Adrenal Medullary

SARDA Spot And Runway Departure Advisor

SCL Skin Conductance Level

SCR Skin Conductance Response

Single European Sky ATM Research (the programme which defines the SESAR Research and Development activities and Projects within Europe)

SID Standard Instrument Departure route

SJU SESAR Joint Undertaking (Agency of the European Commission)

The programme which addresses all activities of the SESAR Joint SJU Work Programme Undertaking Agency.

SME Subject Matter Expert

SNR Signal to Noise Ratio

EDITION 00.02.00

Term Definition

SNS Sympathetic Nervous System

SRK model Skill Rule and Knowledge model

SSR Special Service Request

STAR STandard Arrival Route

STCA Short Term Conflict Alert

human performance neurometricS Toolbox foR highly STRESS project automatEd16utomated SystemS design

SWIM System-Wide Information Management

TA Traffic Alert

TBS Time Based Separation

TCAS Traffic Collision Avoidance System

TCSA Traffic Control using Speed Adjustment

TCT Tactical Controller Tool

tfICA time-frequency Independent Component Analysis

TMA Terminal Manoeuvring Areas

TRA Temporary Restricted Area

TRL Technology Readiness Level

TSA Temporary Segregated Airspace

UAC Upper Area Control

UIR Upper Information Region

UTA Upper Traffic Area

VFR Visual Flight Rules

VHF Very High Frequency

VLF Very Low Frequency

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Term Definition

VOR VHF Omnidirectional Range

WP Working Position

EDITION 00.02.00

3 Review

In this chapter, a review of future SESAR concepts is provided, with a focus on the role of automation. Based on this, the impacted human factors aspects are identified. The chapter concludes with the description of the HPE and how it is impacted by automation. 3.1 Future Scenarios

The latest SESAR Global demonstrations, such as the one regarding the System-Wide Information Management (SWIM), have shown that the development and evolution of the “next” generation of innovative and unconventional ideas, concepts and technologies that define the performance of the future European Air Traffic Management (ATM) system, and contribute to its successful evolution, are already in place. Indeed, as in the case of SWIM, this is no longer a concept on paper, but is progressively becoming a reality that will propel aviation into a new era of global connectivity [4]. This new era not only concerns the introduction of advanced technologies, but it will deal with a revised founding principles and building blocks of information sharing, service orientation, federation, open standards, and information and service lifecycle management. In compliance with the International Civil Aviation Organization (ICAO) and European Aviation Safety Agency (EASA) regulations and directives [5], [6], SESAR is delivering the performance necessary to meet the growing demand for air transport from a worldwide perspective in order to achieve performance ambition levels for 2035 (reported in Figure 1).

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 1: SESAR performance ambition levels for 2035 These levels are subject to the optimal development and deployment of the Operational Changes made possible through SESAR Solutions [7]. Their realisation follows strategic orientations described by 4 Key Features (represented in the figure below), which evolve through an ongoing Deployment and supporting Research & Development (R&D) programme.

Figure 2: The four areas of ATM (Key Features)

EDITION 00.02.00

The operational changes are enabled through improvements to technical systems, procedures, human factors and institutional changes supported by standardisation and regulation. The Master Plan [8] includes roadmaps of the identified changes, ensuring that their deployment is planned in a performance-driven and synchronised way (e.g. between ground and air deployments) to maximise the benefits gained.

For each key area, there are dedicated solutions. In particular, “Advanced Air Traffic Services” and “Optimised ATM network services” concern “Traffic Synchronisation” which covers all aspects related to improving arrival/departure management. It aims to achieve an optimum traffic sequence resulting in significantly less need for Air Traffic Control (ATC) tactical intervention, and the optimisation of climbing and descending traffic profiles” [9]. Free Route Airspace (FRA) Concept of Operations is one of the solutions belonging to these areas. This solution offers additional flight planning route options on a large scale across Flight Information Regions (FIRs), such that overall planned leg distances are reduced in comparison with the fixed route network and are therefore fully optimised. This solution is particularly relevant for cross border control centres located in high and very high complexity environments. Indeed, in current operations, airspace users must follow a fixed route which takes into account sector boundaries to allow for adequate controller support. With the move to a free route environment, airspace users can generate their own optimised route, which is more efficient for them.

By definition, FRA is a specified airspace within which users may freely plan a route between a defined entry point and a defined exit point, with the possibility to route via intermediate (published or unpublished) way points, without reference to the Air Traffic Services (ATS) route network, subject to airspace availability. Within this airspace, flights remain subject to air traffic control. It is a concept of providing air traffic services in which an operator can choose their route subject to only a few limitations, e.g. fixed entry and exit points and the need to avoid danger areas, Temporary Restricted Areas (TRAs) or Temporary Segregated Airspace (TSA) as opposed to the situation where standard airways should be used. In most cases the straight line between an entry point and an exit point will be chosen. If for some reason this is not appropriate (e.g. a danger area needs to be avoided) additional turning points can be specified. These can be navigational aids, published navigational points or points with specified coordinates. The following diagram gives an overview of the main FRA rules:

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 3: Example of allowed and not allowed FRA routes to be considered during the pre-flight planning In the example FIR depicted, INTRO and ENTER are entry points, ALTAV and EXITO are exit points, SNA is a VHF Omnidirectional Range (VOR) point and REKRA is an (aRea NAVigation) RNAV point. When FRA is implemented, the green routes would be accepted and the red routes would be rejected by the Air Traffic Control (ATC) flight plan processing system. The reasons for rejection include the crossing of a danger area (INTRO-ALTAV) and the requested route not remaining within the FRA (ENTER-ALTAV). The approved routes can be either direct from an entry to an exit point (e.g. ENTER-EXITO) or with intermediate points (navigational aids (SNA), published points (REKRA) or randomly selected points (42°39’26” N, 23°22’42” E)).

The overall scope of the FRA Concept of Operations is to provide an enabling framework for the harmonised implementation of Free Route Operations in Europe whenever a State/Air Navigation Service Provider (ANSP), a group of States/ANSPs or a Functional Airspace Block (FAB) decides to proceed with such implementation. The Free Route Concept of Operations forms the basis for a common understanding for all ATM partners involved in FRA Operations implementation [10]. The Concept of Operations encompasses various Free Route Operations implementation scenarios that will:

 Meet the Safety Objectives;

 Be compatible with existing operations;

 Be sustainable through further development;

 Be capable of expansion/connectivity to/with adjacent airspace;

 Be capable of being exported to other regions.

EDITION 00.02.00

The implementation of FRA offers a number of efficiency benefits for the operators. There are also a number of challenges and issues but, overall, this is considered one of the most cost-effective changes to the ATS provision in Europe. The most notable benefits are:

 Reduced flight time, since most flights will be using the shortest routes possible;

 Reduced CO2 emissions, as a consequence of the reduced flight time;

 Reduced fuel waste, also a consequence of the reduced flight time and more optimal flight profiles;

 Low implementation costs for ANSPs – in most cases implementation of FRA is supported by the existing Area Control Centre (ACC) equipment;

 Fewer conflicts (in upper sectors) – since the same number of aircraft are spread over more routes;

 Weight optimisation – in general FRA reduces the difference in distance between the planned route and the actual route. This in turn reduces the amount of extra fuel that needs to be carried potentially allowing for a heavier payload.

As any new technology and procedure in aviation, FRA poses a number of challenges to the users. These do not outweigh the benefits but need to be addressed properly in order to gain the best of FRA. Such issues and challenges are:

 Conflicts may become harder to detect due to the spread and increased number of possible conflicting points.

 Changes to the separation provision methods used by ATC (e.g. direct routes are less an option for solving conflicts since most aircraft are using the most direct route available anyway).

 Vectoring aircraft that have planned their route using points with geographical coordinates can lead to issues when instructing the flight crew to resume own navigation.

 Conflicts occurring shortly after entering the area of responsibility of an ATC sector require controllers to be even more vigilant during transfer/acceptance of control.

 Need for coordinated approach to FRA implementation – the efficiency benefits will only be achieved if FRA is deployed over large areas and appropriated measures are taken so that aerodromes do not become bottlenecks.

 Need for enhanced (system supported) coordination between ANSPs in case FRA extends beyond the state borders.

 Use of odd/even levels, usually determined in the respective Aeronautical Information Publications (AIPs), may not follow the standard assignment (i.e. odd=eastbound, even=westbound).

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

 Aircraft flying along the sector boundaries – the probability of loss of separation in case of deviation from the planned route (e.g. due to weather) shall be given due consideration.

 Aircraft flying near restricted areas (danger areas, TRAs, TSAs, etc.) that have no built-in safety buffer.

 Sectorisation may need to be optimized to better accommodate the new traffic flows. This is a particularly challenging task in case of time limited FRA implementation.

To overcome the presented issues, some mitigation measures have been already developed. Indeed, the following measures can be used to mitigate the safety issues and to cope with the challenges posed by FRA implementation. The list is not to be considered exclusive:

 Large scale deployment of FRA would increase the overall efficiency benefits.

 Step-by-step deployment of FRA would reduce the safety risks. Airspace-specific safety risks could be detected more easily and addressed in a timely fashion.

 Appropriate changes to the airspace design and updates of the letters of agreement (entry and exit points, sectors, restricted areas, ATS delegation, etc.)

 Dedicated training to help controllers familiarise themselves with the new operational issues arising from FRA (e.g. new conflicts, unfamiliar traffic flows, etc.)

 As far as reasonably practicable, both the transferring and accepting controllers should make their best effort to ensure that the aircraft exiting or entering their area of responsibility are not in immediate conflict with other aircraft and be ready to initiate timely coordinated measures for solving the conflict.

 Controllers should coordinate flights flying along sector boundaries with the adjacent sector or unit.

 Restricted areas (TSAs, TRAs, danger areas, etc.) should have buffers so that aircraft can fly safely close to their borders. If a restricted area does not include a buffer airspace, controllers shall ensure that aircraft fly at a safe distance from the area boundaries.

 Re-evaluation and optimisation of existing sector definitions might be necessary; flexible ATC sector configuration management might be applied to manage controller workload in line with changes in the traffic flow and its complexity

Free route airspace initiatives support the objectives of the Flight Efficiency Plan [11] a combined initiative of International Air Transport Association (IATA), Civil Air Navigation Services Organization (CANSO) and EUROCONTROL that was signed in August 2009. Its aim is to enhance European en- route airspace design by giving priority to FRA implementation. However, the introduction of FRA in Europe is a step-by-step process rather than a single act [12]. Most states have decided to start with a limited implementation (e.g. during night hours) and then gradually expand it. Progress so far:

 09.04.2009 – Sweden (Sweden Upper Information Region (UIR))

EDITION 00.02.00

 07.05.2009 – Portugal (Lisboa FIR)  17.12.2009 – UK/Ireland FAB (Shannon Upper Traffic Area (UTA)  10.02.2011 – Maastricht  30.06.2011 – Germany (Karlsruhe Upper Area Control (UAC))  17.11.2011 – Denmark/Sweden (Kobenhavn FIR and Sweden UIR)  02.05.2013 – Czech Republic, Croatia, Poland and Serbia  15.11.2013 – Bulgaria and Romania

The target in the 2014 Network Manager Performance Plan, to implement FRA, either totally or partially, in 25 ACCs, was met in March 2014. As of that dates, a total of 26 ACCs had implemented FRA, 20 had partially implemented FRA and six had done so fully. 3.2 Automation in future scenarios

In the current ATC scenario we have become accustomed to automated systems that support the ATCOs in understanding the situation and defining the appropriate resolution strategy by which to solve it. Such technologies support the human behaviour by facilitating tasks that would be much more complicated without them. However, they do not interfere with the decision of which strategy to apply nor its actual application, as ultimately both these tasks remain under the judgement and the responsibility of the human operator(s). In the future, as anticipated, we may see the introduction of a new generation of highly automated supporting technologies that are able to autonomously (or partially autonomously) manage tasks that are currently carried out by human operators and/or to provide inputs to human decisions that the operators will hardly be in a position to question. The introduction of higher levels of automation will bring about a new task allocation between the human and the machine. Tasks previously carried out by the human, for example the provision of separation, are supposed to be partially delegated to the system. Highly automated systems are expected to take over operators’ repetitive tasks, while human role is expected to be focused on strategic planning, intervening on exceptions and monitoring the system’s behaviour. In general, rather than governing flight operations directly, pilots and controllers will likely supervise the automated systems doing the job. The new scenario will inherently change the supporting role of the technology, its interactions with the operators and the capability of the operators to judge the quality of the information provided. This, in turn, will imply the need to consider the new systemic risks and mitigations that can be associated with technological and/or organisational failures. It will also imply the need for a radical revision of the competences required to perform the tasks – as well as of how tasks, roles and responsibilities are allocated among the operators (both in the front-end and in the back-end) and between operators and machines.

3.2.1 Review on automation: frameworks and models

As postulated in research by Sheridan & Verplanck [13] automation is not all-or-nothing, that is, automation is not only a matter of either automating a task entirely or not, but to decide on what task to automate and on the extent it should be automated. In fact, different tasks involve the use of

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

different psychomotor and cognitive functions, which in turn implies the adoption of different automation solutions. For example, expanding human capabilities to monitor a certain process (e.g. a Remote Tower) is not the same as replacing the human in the execution of a certain action (e.g. the aircraft auto-braking system). Similarly supporting the analysis of a complex dataset, such as that involved in predicting the risk of a traffic conflict, is not the same as identifying the best solution to resolve the conflict. Automation then is not seen to replace operators but to empower them and to improve the overall performance of ATM as clearly defined by the research network on Higher Automation Levels in Aviation (HALA!) [14]. This approach is similar to the one followed in the early 80’s when flying crew for large civil aircraft was reduced from three to two by adding sophisticated Flight Management Systems (FMS) [15]. However, it is important to design this automation very carefully taking into account the elements of the CAS. As clearly pointed out by Lisanne Bainbridge [16] automation malfunctions end up most of the time in the hands of operators that were precisely supported with automation as their tasks were too complex or too resource consuming. Moreover, introducing higher level of automation requires (beyond design issues) an evaluation of the impact which the new technology may have on each CAS part (operator, computing system and organisation) such as tasks migration and/or functions allocation [17]. Such function allocation (as illustrated in Figure 4), concerns the ground side of the ATM System i.e. the Air Traffic Control Centres (ATCs). On the top- right side of Figure 4, Air Traffic Controllers communicate with pilots via data link or transfer aircraft interacting on the electronic labels of the aircraft on the radar screen instead of using paper strips and communicating by voice using Very High Frequency (VHF) medium. Similarly, on the airborne side (lower part of Figure 4), glass cockpits [18] provide a means for integrating information to support pilots activities while this information was previously distributed amongst multiple displays throughout the cockpit. In both cases, task migration and/or functions allocation require that humans improve their knowledge, learn how to interact and collaborate with the new technology for accomplishing tasks.

EDITION 00.02.00

Figure 4: The enhancement of integrated automation support in ATCs and cockpits over time In automated systems, function allocation [19] between human and machine has always been a point of controversy. In the context of automation, “functions allocation” means that the actor (either human or machine) that is best suited (based on some continuum of parameters) should perform the function. The basis for selection and grading of such parameters is at the heart of the issue of function allocation and has been subject to much investigation over the years. The approach proposed by Fitts with his Men are best at – Machines are best at (MABA-MABA) list [20], relied on the idea that, given a set of 44 pre-existing tasks, one should decide which ones are worth automating, considering the strengths and weaknesses of respectively humans and machines. Although this approach is now deemed outdated, there is still limited awareness of the fact that introducing automation brings qualitative shifts in the way people practice, rather than mere substitutions of pre-existing human tasks. An initial scale of levels of automation was proposed by Sheridan & Verplanck [21] representing a continuum of levels between low automation, in which the human performs the task manually, and full automation in which the computer is fully autonomous (cf. Figure 5).

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 5 - Levels of Automation of Decision and Action Selection by Sheridan and Verplanck A decisive step was made by Parasuraman et al. [22] who acknowledged the Sheridan-Verplanck 10- point scale and introduced the idea of associating levels of automation to functions (Figure 6). These functions are based on a four-stage model of human information processing and can be translated into equivalent system functions: (1) information acquisition, (2) information analysis, (3) decision and action selection and (4) action implementation. The four functions can provide an initial categorisation for types of tasks in which automation can support the human.

Figure 6 - A model of types and levels of automation by Parasuraman et al.

EDITION 00.02.00

Most of the time automation is only partial keeping the operators in the loop so that they can forecast what will happen next and interfere with automation in case of adverse events or automation malfunction. The design of this cooperation requires understanding how to balance automation and interactivity and specify how a task can be performed by assigning the generic functions to the operator and the system in terms of function allocation. “Function allocation cannot be based on a consideration of the tasks only, but must consider the total equilibrium of a work situation—corresponding to a notion of balanced work. The concept of equilibrium emphasises the fact that a change in function allocation disturbs the established equilibrium. This will have consequences for the system as a whole, and one result may be that a new equilibrium is established which differs significantly from the previous one.” [23, p. 292]) Previous work on automation can be divided according to three different perspectives: 1) the design perspective which focuses on how to engineer the computing systems (offering automation) and more precisely its user interface [24]; 2) the evaluation perspective which focuses on how to assess the operational aspects of automation including performance impact of automation on operations [25], [26]; 3) the human perspective which focuses on how to understand the role of the operators who deal with a new technology or a different level of automation [22][27]. While this research work has been mostly conducted in separate fields, as the increase of automation might come along with an increase of performance variability of the whole ATM especially in case of new automated systems, there is a need to provide an integrated view on these disjoint research activities [28]–[30].

3.2.2 Classifying automation: the Level of Automation Taxonomy

In the framework of the SESAR Programme, a Level Of Automation Taxonomy (LOAT) has been developed to classify and compare different kinds of automation support in Air Traffic Management (ATM) [27]. The LOAT (Figure 7) is based on the taxonomy of Endsley & Kaber [26] and the principles of Parasuraman et al. [22], which combines cognitive functions and levels of automation, and on ideas from activity theory and distributed cognition [31]. Its purpose is to classify automation examples in a practical way. The Taxonomy is organised as a matrix. In the horizontal direction, four functions are depicted: information acquisition, information analysis, decision and action selection, and action implementation. A consequence of having four functions – different in nature – is that each function can be automated at different levels. In line with this, vertically, each cognitive function groups a number of automation levels (between 5 and 8). All automation levels start with a default level 0, corresponding to manual task accomplishment, and increase to full automation. Automation level 1 is based on the principle that the human is accomplishing a task with primitive external support, which is not automation as such. Any non-automated means that support the human mind, e.g. using flight strips to compare parameters of different aircraft and to pre-plan future traffic, could correspond to this intermediate level.

The classification of the level of automation is provided according to the concerned cognitive function. This means that a certain technology may have different levels of automation according to whether we look at the information acquisition (A), information analysis (B), decision-making (C) or action implementation (D) fields. Examples of technologies are included in the LOAT, to facilitate the reader in understanding how to interpret and use the table for the classification of a technology.

From INFORMATION to ACTION A B C D

INFORMATION ACQUISITION INFORMATION ANALYSIS DECISION AND ACTION ACTION IMPLEMENTATION SELECTION

A0 B0 C0 D0 Manual Information Working Memory Based Human Decision Manual Action and Control Acquisition Information Analysis Making The human acquires relevant The human compares, The human generates The human executes and controls all actions information on the process combines, and analyses decision options, selects the manually.

INCREASINGAUTOMATION s/he is following without different information items appropriate ones and decides using any tool. regarding the status of the all actions to be performed. process s/he is following by way of mental elaborations. S/he does not use any tool or support external to her/his working memory. A1 B1 C1 D1 Artefact-Supported Artefact-Supported Artefact-Supported Decision Artefact-Supported Action Implementation information Acquisition Information Analysis Making The human acquires relevant The human compares, The human generates The human executes and controls actions with information on the process combines, and analyses decision options, selects the the help of mechanical non-software based s/he is following with the different information items appropriate ones and decides tools. support of low-tech non- regarding the status of the all actions to be performed digital artefacts. process s/he is following utilising paper or other non- Ex. 1) Use of a hammer or leverage to increase utilising paper or other non- digital artefacts. the kinetic energy of human gesture. digital artefacts. Ex. 1) Identification of Ex. 2) Use of a mechanical or hydraulic rudder aircraft positions on an to achieve a change in direction. aerodrome/airport according Ex. 1) Use of flight strips to to Procedural Air Traffic compare EDITION 00.02.00

Control rules and without use altitudes/levels/planning times of radar support. of different aircraft and to pre- plan future traffic. A2 B2 C2 D2 Low-Level Automation Low-Level Automation Automated Decision Support Step-by-step Action Support: Support of Information Support of Information Acquisition Analysis The system supports the Based on user’s request, the The system proposes one or The system assists the operator in performing human in acquiring system helps the human in more decision alternatives to actions by executing part of the action and/or information on the process comparing, combining and the human, leaving freedom by providing guidance for its execution. s/he is following. Filtering analysing different information to the human to generate However, each action is executed based on and/or highlighting of the items regarding the status of alternative options. The human initiative and the human keeps full most relevant information the process being followed. human can select one of the control of its execution. are up to the human. alternatives proposed by the Ex. 1) Activation by ATCOs of system or her/his own one. Ex. 1) A crane tele-operated by a human for Ex. 1) Identification of Speed Vectors for specific construction works aircraft positions in the tracks on the Controller Ex.1) AMAN visualization of Ex. 2) The aural and visual component of airspace by way of Primary Working Position (CWP), in the proposed sequence of Traffic Collision Avoidance System (TCAS) Radar working positions. order to anticipate potential aircraft. Resolution Advisory (RA) in current TCAS II Ex 2). Use of video cameras conflicts in a defined time Ex.2) Enhanced Task version 7.0 –see also Level Of Automation to monitor traffic on airport’s frame. Monitoring Load (E-TML) (LOA) C5 area not visible from the Ex 2). Colour coding of traffic used by the Supervisor to Tower flows on ATCOs request identify the best configuration of sectors. A3 B3 C3 D3 Medium-Level Automation Medium-Level Automation Rigid Automated Decision Low-Level Support of Action Sequence Support of Information Support of Information Support Execution Acquisition Analysis The system supports the Based on user’s request, the The system proposes one or The system performs automatically a 30 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

human in acquiring system helps the human in more decision alternatives to sequence of actions after activation by the information on the process comparing, combining and the human. The human can human. The human maintains full control of s/he is following. It helps the analysing different information only select one of the the sequence and can modify or interrupt the human in integrating data items regarding the status of alternatives or ask the system sequence during its execution. coming from different the process being followed. to generate new options. Ex. 1) Explicit initiation of an electronic sources and in filtering The system triggers visual coordination with adjacent sector via digital and/or highlighting the most and/or aural alerts if the input (replacing use of telephone). relevant information items, analysis produces results Ex. 2) ATCO’s input into the Controller-Pilot based on user’s settings. requiring attention by the user. Data Link Communication (CPDLC) of a Ex. 1) CWP allowing ATCOs to clearance which is than transmitted to the a/c set flight level filters to Ex. 1) En-Route Air Traffic (replacing clearance sent via R/T). display only certain traffic on Organizer (ERATO) Filtering the screen. and What-if function. Ex 2). VERA Separation and Resolution Tool to display the closest point of approach between two aircrafts. A4 B4 C4 D4 High-Level Automation High-Level Automation Low-Level Automatic High-Level Support of Action Sequence Support of Information Support of Information Decision Making Execution Acquisition Analysis The system supports the The system helps the human in The system generates options The system performs automatically a human in acquiring comparing, combining and and decides autonomously sequence of actions after activation by the information on the process analysing different information on the actions to be human. The human can monitor all the s/he is following. The system items regarding the status of performed. The human is sequence and can interrupt it during its

EDITION 00.02.00 integrates data coming from the process being followed, informed of its decision. execution. different sources and filters based on parameters pre- Ex.1) Acknowledgment by pilot of a clearance and/or highlights the defined by the user. The Ex.1) Aural and visual received trough CPDLC and automatically sent information items which are system triggers visual and/or component of TCAS RA in to FMS and autopilot. considered relevant for the aural alerts if the analysis current TCAS II version 7.0 Ex. 2) Autopilot following the FMS trajectory. user. The criteria for produces results requiring (also LOA D2) integrating, filtering and attention by the user. highlighting the relevant information are predefined Ex. 1) Medium-Term Conflict at design level but visible to Detection (MTCD) visual alerts the user. (allowing some tuning of parameters by the user) Ex.1) Airport data-link function (D-TAXI) tool (including graphical route information) A5 B5 C5 D5 Full Automation Support of Full Automation Support of High-Level Automatic Low-Level Automation of Action Sequence Information Acquisition Information Analysis Decision Making Execution The system supports the The system performs The system generates options The system initiates and executes human in acquiring comparisons and analyses of and decides autonomously automatically a sequence of actions. The information on the process data available on the status of on the action to be human can monitor all the sequence and can s/he is following. The system the process being followed performed. The human is modify or interrupt it during its execution. integrates data coming from based on parameters defined informed of its decision only Ex. 1) Implicit initiation of an electronic co- different sources and filters at design level. The system on request. ordination with adjacent sector as agreed exit and/or highlights the triggers visual and/or aural (Note that this level is always conditions (according to Letter of Agreement) information items which are alerts if the analysis produces connected to some kind of cannot be met anymore after changes to the considered relevant for the results requiring attention by ACTION IMPLEMENTATION, a/c trajectory (route or flight level) has been user. The criteria for the user. at an automation level not made. integrating, filtering and lower than D5.) 32 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

highlighting the relevant info Ex. 1) TCAS Traffic Alert (TA) are predefined at design (including visual display and level and not visible to the aural annunciation of intruding user (transparent to the user traffic) in Computer Science terms). Ex. 2) Short-Term Conflict Alert (STCA) visual and aural alerts. Ex. 3) Down-linked Selected Altitude Monitoring (DSAM) alerts. Ex. 4) Trajectory prediction to display the Estimated Time Over a fix on an aircraft trajectory C6 D6 Full Automatic Decision Medium-Level Automation of Action Making Sequence Execution The system generates options The system initiates and executes and decides autonomously automatically a sequence of actions. The on the action to be human can monitor all the sequence and can performed without informing interrupt it during its execution. the human. (Note that this level is always connected to Ex.1) TCAS Auto Pilot/Flight Director (AP/FD) some kind of ACTION mode concept during execution of a corrective IMPLEMENTATION, at an TCAS RA. automation level not lower

EDITION 00.02.00

than D5.) D7 High-Level Automation of Action Sequence Execution The system initiates and executes a sequence of actions. The human can only monitor part of it and has limited opportunities to interrupt it. D8 Full Automation of Action Sequence Execution The system initiates and executes a sequence of actions. The human cannot monitor nor interrupt it until the sequence is not terminated. Figure 7 - The LOAT

The way LOAT is designed demonstrates the following principles [32]:  An automated system cannot have one ‘overall’ level of automation as such. In other words, a statement about a level of automation for a system always refers to a specific function being supported;  One automated system can support more than one function, each having a different level of automation;  The description of each automation level follows the reasoning that automation is addressed in relation to human performance, i.e. the automation being analysed is not just a technical improvement but has an impact on how the human is supported in his/her task accomplishment. It should be kept in mind that these generic functions are a simplification of the many components of human information processing. The functions are not meant to be understood as a strict sequence, but they may temporally overlap in their processing. From a practical point of view, the human may be performing a task that involves one or several functions. However it is useful to differentiate the subtleties between the functions when one wants to identify how a specific automated system supports the human.

3.2.3 Automation support to ATCOs and pilots tasks

Pilot and Controller tasks are not automated in the same way [33]. Aircraft automation is sometimes considered to be more advanced than ATC automation. This perception is only partially true, as it seems to disregard the nature of pilot and controller activities, at least to the extent that non-pilots sometimes understand them. Pilot tasks are much more “Action Implementation” oriented than controller tasks, for which the emphasis is more on monitoring, planning and communicating. Therefore, the replacement or support of a human action – which is normally perceived as “real” automation – is inevitably more successful when pilot tasks are concerned. In the limited number of automated functionalities reported as examples in the LOAT, there is a prevalence of “Information acquisition” and “Information Analysis” functions in ATC-related automations. Examples of this were the Multi-Radar Tracking system display, the Short Term Conflict Alert (STCA) system, the Medium Term Conflict Detection (MTCD) system and the Tactical Controller Tool (TCT). On the other hand there was a clear prevalence of “Action Implementation” functionalities among aircraft automations. For instance, the Autopilot following a FMS trajectory, the Autobrake system and the ASAS-ASPA (Airborne Separation Assistance – Airborne Spacing system) capability. Finally a more balanced distribution between ground and aircraft was observed for the “Decision and Action Selection” automations, although the ATC functionalities were generally less mature and were providing a lower level of support. The Arrival Manager (AMAN), which is a good example of ATC “Decision and Action Selection” functionality, is increasingly prevalent but in most of the cases it provides just a useful reference that the controller may decide to follow or not, depending on operational circumstances. This kind of support is at a considerably lower level than that offered, for example, by the Resolution Advisory (RA) of the Traffic Collision Avoidance System (TCAS) which indicates to the pilot one single and directed action to avoid possible collision with conflicting traffic. It is interesting to note that some of the aircraft functionalities we analysed also included “Information Acquisition” and “Information Analysis” components. However these were generally

EDITION 00.02.00

acknowledged to be less sophisticated than the ATC-related ones (consider the example of the TCAS Traffic Display which is known to be of limited functionality relative to controllers' radar displays and well known to be unusable by pilots as a means of self-separation). Much more sophisticated “Information Acquisition” functionalities are beginning to be introduced for the flight deck and we looked at ATSAW-SURF (Air Traffic Situation Awareness for Surface Operations) – which uses Automatic Dependent Surveillance - Broadcast (ADS-B) capability. More than just a simple technological improvement, this will, subject to the development of operator procedures, make possible a partial delegation to pilots of tasks which have previously been an exclusive prerogative of ATC. 3.3 HF relevant concepts in future scenarios

Technological and organizational changes constantly bring about a practical need to know about the cognitive and emotional processes of the operators. The aviation research on neurophysiological indicators has mostly focused on cognitive concepts, traditionally disregarding the emotional aspects. This oversimplification is hard to justify at the light of current neurophysiological knowledge, where emotions have been shown to play a key role in “cognitive” processes like decision-making or attentional focus. Most of the proposed changes deriving from the implementation of the FRA scenario are expected to increase the pilot’s autonomy in controlling the route of their aircraft, including the requirement to maintain required separation between aircraft. Flight time, delays and fuel expenditures are all expected to profit from such changes. Despite the increase in pilot autonomy, controllers are still responsible for running the ATC system safely. In other words, the controllers role would be shifted from that of an active controller to one more like that of a monitor [34]. Assuming that the equipment and the pilots all perform correctly, the controller’s workload would be expected to decrease in nominal situations, since they are interacting with fewer aircraft. It also seems plausible that the controller’s situational awareness of the airspace would decrease as well, since they would not be focusing as much attention as they previously did on many of the aircraft. However, for an unknown period of time, certain aircraft in the system would be properly equipped to participate according to the new rules, allowing more autonomy, but others would not be and would require the controller to manage their flight path. As systems become more automated, and humans move to monitoring positions, the weight of stress is likely to grow. A typical case is the automation disruptions, when humans have to react quickly in highly stressful conditions. In these cases, stress is known to influence performance and impair attention, memory, and decision making [35]. The relevance of stress is also recognised by EASA, that in the Notice of Proposed Amendment (NPA) addresses the issue of licensing and medical certification of air traffic controllers [6], considering stress and fatigue management as an essential topic for training (AMC1 ATCO.D.045(c)(4) Human Factors training). In the following section a review of the frameworks for understanding stress, attention, mental workload and cognitive control are provided. In section 4.3, neurophysiological measures for each factor will be presented.

3.3.1 Stress

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

According to psychological theories stress is determined by the balance between the perceived demands from the environment and the individual’s resources to meet those demands [36], [37]. So stressful experiences arise as person-environment transactions. These transactions depend on the impact of the external stressor. This is mediated firstly by the person’s appraisal of the stressor and secondly on the social and cultural resources at his or her disposal[38]. When faced with a stressor, a person evaluates the potential threat (primary appraisal). Primary appraisal is a person’s judgment about the significance of an event as stressful, positive, controllable, challenging or irrelevant. Facing a stressor, the second appraisal follows, which is an assessment of people’s coping resources and options. Secondary appraisals address what one can do about the situation. Actual coping efforts aimed at regulation of the problem give rise to outcomes of the coping process. Lazarus has drawn a distinction among three kinds of stress: harm, threat, and challenge . Harm refers to psychological damage that had already been done, as for example an irrevocable loss. Threat is the anticipation of harm that has not yet taken place but may be imminent. Challenge results from difficult demands that we feel confident about overcoming by effectively mobilizing and deploying our coping resources. These different kinds of psychological stress states are presumably brought about by different antecedent conditions, both in the environment and within the person. They are relevant in the framework of our research as they can have different consequences in terms of human performance. For example, threat is an unpleasant state of mind that may seriously block mental operations and impair functioning, while challenge is exhilarating and associated with expansive, often outstanding performance. To the extent that we take these variations seriously, stress cannot be considered in terms of a single dimension such as activation. The recognition of the different dimensions of stress involves considering diverse emotional states (some negative, some positive) and different impact on performance. The model also highlights that the relation between stress and performance is not linear. In fact, stress is normal up to a point and can be optimal for certain performance related tasks, while it becomes a problem when interfering with a person’s ability to do daily life tasks over a period of a few weeks or impacting his/her health in a dangerous or risky way. It has been seen that stress can influence performance and may impair attention and memory, and can contribute to an increase of human errors and accidents. In general, stress affects how we perceive and process information, as well as what decisions we make, leading to an increase in the number of errors and mistakes. In the context of ATC, the human perception of stress is particularly important. Just consider that the Air Traffic Controller job has been classified, by the U.S. Department of Labor, the fourth most stressful job ever [39]. In fact, it entails a complex set of tasks requiring very high levels of knowledge and expertise, as well as the practical application of specific skills pertaining to cognitive domains (e.g. spatial perception, information processing, logic reasoning, decision making), communicative aspects and human relations. To have an idea of its complexity, it is sufficient to mention that, according to a job analysis of en-route controllers carried out by a group of American researchers [40], six main activities can be identified (i.e. situation monitoring, resolving aircraft conflicts, managing air traffic sequences, routing or planning flights, assessing weather impact, managing sector/position resources), which include 46 sub-activities and 348 distinct tasks. For example, the relevant cognitive/sensory attributes required for high performance levels at radar workstations are spatial scanning, movement detection, image and pattern recognition, prioritizing, visual and verbal

EDITION 00.02.00

filtering, coding and decoding, inductive and deductive reasoning, short- and long-term memory, and mathematic and probabilistic reasoning. From a physiological view, a typical stress response means that autonomic activity increases, although in certain situations and in certain individual’s the stress response might be different (even a decrease is possible). The basis of the physiological stress model has its roots in the research activity of Walter B. Cannon and later Hans Selye. Cannon [41] developed the “fight-flight” concept, which linked emotional expressions, such as fear, to physiological changes in the periphery. He emphasized the activation of the Sympathetic Adrenal Medullary (SAM) system in such situations, irrespective of whether the emergency reaction was “fight” or “flight”. The more reactive biomarkers of the fight-flight response are the catecholamines, in the form of adrenaline and noradrenaline, which increase when stress appears, and other physiological indicators associated with the Autonomic Nervous System (ANS). Selye [35] theorized the General Adaptation Syndrome (GAS) to model the dynamic of the human body adaptation to stressful environmental conditions. Therefore, the SAM system is activated when the individual is challenged in its control of the environment, or is threatened, and this defence reaction prepares the body to battle or escape. Increased adrenaline under normal levels of stress is associated with improved performance. In fact, several studies show that stress, but only up to a certain level, improve performance, e.g. on selective attention tasks [42]. In fact, the cognitive psychology literature demonstrates that activation has an ‘‘inverted U-shape’’ relationship with performance in that some levels of activation may help an individual to perform at a level that is higher than their baseline state [43]. LeBlanc et al. (2008) noted that general surgery residents had improved technical performance on training tasks while dealing with stress conditions. On the other hand, excessive activation may lead to severe stress that overpowers an individual or team, with resulting impairment in memory, attention, decision-making, and general performance, regardless of previous training [35]. Therefore, stress is a physiological response to the mental, emotional, or physical challenges that we encounter. The immediate body's “fight or flight" response causes hormones secretion into the bloodstream to intensify concentration and quicken reflexes. From a physical point of view, several reactions related to the ANS activation can be measurable, such as increased heart rate skin sweating. Under healthy conditions, the body returns to its normal state after dealing with acute stressors. Stress can be categorized into two basic forms: acute stress, relatively short in duration and is often experienced as caused by high task load; chronic stress, prolonged stress that can result from occupational or non-occupational sources. Four possible situations that may cause chronic stress: too frequent stress exposure, failure to habituate to repeated exposure of the same kind of stressor, inability to shut off the stress response, despite that stress has terminated, and situations that cause regulatory disturbances of the stress system. Continuous monitoring of an individual's stress levels is essential for understanding and managing personal stress.

3.3.2 Attention

Attention is the ability to attend to information in the environment [44]. In particular, attention allows us to process selectively the vast amount of information whom we have to face, prioritizing some aspects of information while ignoring others by focusing on a certain

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

location or aspect of the visual scene [45]. Attention can be classified into three main components [22]:  Selective attention: The ability to process or focus on one message in the presence of distracting information.  Divided attention: The ability to process more than one message at a time.  Visual attention: The mechanism determining what information is or is not extracted from our visual field. One important concern for a notification system in ATM field is that seemingly prominent objects in the visual field can sometimes elude attention despite their relevance and importance to the primary task (e.g. [46]). This phenomenon “inattentional blindness” [47], [48] can occur even when a stimulus is salient in terms of its colour or movement [47], thus posing a challenge for the design of safety- critical emergency alerts. The likelihood of inattentional blindness is increased with the attentional demands of the task [49], working memory load [50], the need to maintain information in visuo- spatial memory [51], low expectancy of events [52] and during periods of high tempo activity with competing visual demands [53]. An object is more likely to be detected if it is near the focus of visuo- spatial attention [54], but proximity is not sufficient for detection and it can still be missed [49], [55]. In demanding tasks, operators may experience attentional narrowing or ‘tunnelling’ [56] and become fixated on a particular facet of their task, to the exclusion of other equally – or perhaps more – important aspects of the environment [57]. Neuroimaging studies related to attention have revealed three networks related to different aspects of attention: alerting, orienting, and executive control [58].  Alerting is defined as maintaining a state of high sensitivity to incoming stimuli, and is associated with the frontal and parietal regions of the right hemisphere [59].  Orienting is the selection of information from sensory input, and it is associated with posterior brain areas including the superior parietal lobe (related to the lateral intraparietal area in monkeys), the temporal parietal junction and the frontal eye fields [58], [60].  Executive control is defined as involving the mechanisms for resolving conflict among possible responses. It activates the anterior cingulate and the lateral prefrontal cortex [61]. This attention network affects visual processing, which is one of the most efficient ways to enhance the stimulus representation for the purpose of selection. Shift in attention need not entail an overt shift of the eyes [56]. Nevertheless, spatial attention and the eyes often move about the environment in “tandem”. For example, an abrupt onset in the visual periphery can reflexively “capture” both attention and the eyes [62], [63]. In this way, attention can be allocated in overt or covert modality: overt, when an observer moves his/her eyes to a relevant location and the focus of attention coincides with the movement of the eyes, or covert [64], when attention is deployed to relevant locations without accompanying eye movements. The deployment of covert attention aids us in monitoring the environment and can inform subsequent eye movements. Humans deploy covert attention routinely in many everyday situations, such as searching for objects, driving, crossing the street, playing sports and dancing. Covert attention allows us to monitor the environment and guides our eye movements (overt attention) to locations of the

EDITION 00.02.00

visual field where salient and/or relevant information is. Moreover, covert attention plays an important role in social situations, for example, in competitive situations (such as sports activities). Moving the eyes also provides a cue to intentions that the individual wishes to conceal, a predicament solved by covert attention. Spatial resolution, our ability to discriminate fine patterns, is not uniform across locations in the visual field. It decreases with eccentricity. Correspondingly, signals from the central parts of the visual field are processed with greater accuracy and faster reaction times (e.g., [41], [65], [66]). Lack of attention/vigilance or distraction usually affects human performance by causing the omission of procedural steps, forgetfulness to complete tasks, and taking shortcuts that may not be for the better. A performance decrement can be noticed when attention/vigilance, workload and task difficulty increase; the reaction time and number of errors increase as well, while accuracy and number of completed tasks decrease. Reduction of the performance in monitoring, tracking, auditory discrimination, and reduction of visual field can be observed too.

3.3.3 Mental workload

Mental workload is a hypothetical construct that describes the extent to which the cognitive resources required to perform a task have been actively engaged by the operator. The term is used to describe aspects of the interaction between an operator and an assigned task. Tasks are specified in terms of their structural properties; a set of stimuli and responses are specified with a set of rules that map responses to stimuli. On the other side, there are expectations regarding the quality of the performance, which derive from knowledge of the relation between the structure of the task and the nature of human capacities and skills. Workload is invoked to account for those aspects of the interaction between a person and a task that cause task demands to exceed the person’s capacity to deliver. Mental workload is clearly an attribute of the information processing and control systems that mediate between stimuli, rules and responses. Increase of workload and task difficulty lead to a performance decrement that reflects in a decrease of accuracy and number of completed task, while reaction times and number of errors increase. The increase of mental workload could lead to a Situation Awareness decrease which, in turn, could lead to worse performances. However, the adoption of compensation strategy can result in the lack of visible effects of workload variations on subject performance.

3.3.4 Cognitive control

The efficiency of humans in coping with complex situations is largely due to the availability of a large repertoire of different mental representations of the environment from which rules to control behaviour can be generated ad-hoc [67]–[69]. Nowadays, several models defining the different types of cognitive human behaviour are available [46], [70], [71]. For this study, we have selected the Skill, Rule and Knowledge (SRK) model, proposed by Rasmussen in 1983 [72]. Accordingly with basic different ways of representing the constraints in the cognitive human behaviour, three typical levels emerge: skill-, rule-, and knowledge-based level [72], [73]. At the skill-based level, the types of activity are usually routine and automated, such as annotating flight strips. This type of behaviour tends to encourage errors which are associated with attentional or memory failures.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

At the rule-based level of activity, an individual uses certain types of response to known and often rehearsed scenarios. For instance, in the ATC environment, standard operating procedures would be classed as rule-based behaviour. The use of separation procedures and weather limits would also be classed as a similar type of activity. Rule-based mistakes involve the application of already known but inappropriate solutions to problems that have been encountered many times before or which have been highly trained. Rule based mistakes can be divided into two types: the misapplication of good rules and the application of bad rules. Knowledge-based behaviour is the result of skill, ability, observation, training and experience. These variables enable the individual to tackle novel, difficult or even dangerous situations with adequate reliability and in most cases the likelihood of a successful outcome. Knowledge-based mistakes occur when a person is attempting to solve a novel problem, namely searching for a solution to a problem which has not been previously encountered in training or experience. The main problems with this situation are that the individual is forced into a position of active reasoning and retrieval from long- term memory, which has an influence on the working (mid-term) memory capacity. Table 1 provides an overview of the levels of cognitive control. Table 1 - Levels of cognitive control

Conscious & Mainly conscious type Mainly automatic type Automatic type of of control of control control

Routine- expected SKILL-BASED situation

Familiar or trained-for RULE-BASED problems

Novel, difficult or KNOWLEDGE-BASED dangerous situations

3.4 Human Performance

Controllers’ performance is recognised to be impacted by all the aforementioned factors, namely stress, emotions, attentional resources available, attention focus and so on. In the recent years the concept of Human Performance Envelope (HPE) [74] has been introduced as new paradigm in Human Factors. Rather than focusing on one or two individual factors (e.g. fatigue, situation awareness, etc.), it considers their full range, mapping how they work alone or in combination to lead to a performance decrement that could affect safety.

3.4.1 Automation impact on Human Performance

The future ATCO role will share some commonalities with what happened to pilots due to cockpit technological evolution. The HP envelope currently being considered by the Future Sky Safety project

EDITION 00.02.00

is a first good approximation of what may be expected also for controllers: a prominence of workload and stress, then attention, communication, vigilance, and fatigue.

Figure 8: The HP envelope being considered by the Future Sky Safety project for pilots

It is reasonable to expect that the future ATCO HP envelope will be different than the one we would use today. It will have different underlying HF concepts, or at least a different weight among them. For instance, ATCOs are expected to move to a monitoring position of highly automated systems, with very few tactical interventions, strategic planning by exception (only when automation cannot find a solution), need to intervene rapidly to recover disruptions or unexpected events. As compared to pilots, workload may be even less primary, but with sudden bursts when recovery actions are needed. Stress will be indeed a major factor, both in normal conditions (when ATCO will need to rely on automation without having the possibility of controlling it) and in disruptions. Such a monitoring role will probably require even more attention that pilots today. ATCOs will need to deal with very complex system, with many interacting elements, of different typologies (e.g. RPAS), moving in 4D trajectories across space. A full understanding of the future HP envelope and of the underlying HF factors will allow the optimization of Human Performance, by identifying human bottlenecks and limitations. This concept is aligned with the ACARE Strategic Research Agenda [75]. This concept applies to nominal conditions, when the HP envelope analysis would determine the human bottlenecks to higher automation levels, e.g. lack of trust, too high stress level, or lack of adequate attentional patterns. But it is also relevant in degraded conditions, where it will be possible to track the temporal variations of the HP envelope. As a practical example, this could mean forecasting what would typically happen to the stress and workload levels in case of automation disruptions and when their levels could be considered “back to normal”. In the framework of the already mentioned SESAR 16.05.01 project, a method to identify design principles linking the level of automation with HP issues, has been generated. It has to be noted that the results of this investigation are highly related to the knowledge on HFs currently available, and to experiments conducted with currently available automated tools. Table 2 Example of design principles and HP impact of different automation levels

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Automation Example of design principle for Cognitive support selecting a specific automation Impact on HP Functions example level (current):  Increase task demand and cognitive workload IAC-1.1: Level A2  Simultaneous tasks Information  Visualization These functions transmit competing for user Acquisition of traffic on information on the process to be attention or causing (IAC) CWP through followed without any filtering, interruptions of high Acquisition and Multi-Radar highlighting and integration of workload activities, registration of Tracking the available data. These reducing efficiency and multiple sources System. functions should be preferred increasing the risk of when the two following human error of information.  Advanced - conditions are met:  Excessive ‘head down’ Initial pre- Surface time, with potential processing of Movements -the concerned activity requires negative impact on data prior to full Guidance and the processing of a limited human performance perception and Control system number of information items by selective (A-SMGCS) the human operator  Reduce accessibility to attention. airport moving -the information can be accessed relevant information, with map through a limited number of negative impact on information sources. decision making processes and situation awareness.

 STCA- IAN-1.2: Levels B4-B5 Minimum Safe Altitude These functions perform Information Warning systematic information analysis Analysis (IAN) (MSAW)- and trigger automatic alerts as Conscious Approach Path appropriate. These functions should be preferred when the two perception, Monitor  Induce workarounds and following conditions are met: manipulation of (APM)-Area higher workload in human information in Proximity - for operational reasons the user operators. working Warning should be informed as soon as  Induce misuse, disuse memory. (APW) visual possible of critical results of the or abuse of automation Cognitive and aural information analysis  Distrust in automation operations alerts and increase workload including - the internal logic of the  A-SMGCS rehearsal, automated function has the route-planning integration, capability to process the most function inference. relevant operational constraints  MTCD and/or dynamic elements of the conflict concerned operational alerting environment with an impact on

EDITION 00.02.00

function the performance of the function, thus reducing the risk of  Monitoring triggering nuisance alerts to an Aids (MONA) effective minimum. alerts  A-SMGCS alerts DAS-1.3: Levels C5-C6 Decision and Action Selection These functions do not inform the (DAS) human of the selected option or inform her/him only on request. Selection among decision  AMAN These functions should be alternatives, visualization of preferred when it is possible to based on proposed safely and efficiently isolate a  Increase task demand previous sequence of limited set of parameters and and cognitive workload information aircraft variables to be managed  Increase task load and analysis.  AMAN autonomously by the automation. reduced acceptance. Deciding on a speed In such cases the best way of particular advisories generating decisions is delegating (‘optimal’) them to the automated function, option or leaving to the human operator strategy. only the possibility to interrupt the following action

implementation.

AIS-1.1: Levels D2-D3-D4  Reduced vigilance and Action loss of situation Implementation This level, which perform action awareness Support  Automatic implementation only after human  Loss of skills and Functions (AIS) Flight Plan initiation, should be preferred proficiency Implementation Correlation when it is not possible to safely  Impact recovery from of a response or function and efficiently isolate a limited set system failure of parameters and variables to be action  Automatic  Reduce situation managed autonomously by the consistent with Special Service awareness. automation. In such cases the the decision Request (SSR)  Induce workarounds and best strategy for action previously code higher workload in human implementation is based on a made. Carrying assignment operators. cooperation between the human out the chosen function  Reduce the human operator and the automated option. potential to adapt to function, maintaining the human normal and abnormal fully in control of the situation. situations

A brand new contribution will be the study of the transitions among levels, providing information about the impact of transition on HP and generating guidelines to cope with them minimizing system performances degradation.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

4 Scenarios selection

In this chapter, the method for selecting the scenarios to be reproduced in the STRESS validation activities is described. 4.1 Method for STRESS scenarios selection

With the aim of selecting the most appropriate scenarios for STRESS, it is needed to identify a set of criteria. In particular, these selection criteria are:  Scenarios should be compliant with the project objectives as reported in the proposal and in the successive documents or publications (i.e. poster submitted at SESAR Innovation Days 2016);  Scenarios should be compliant with the lessons learned derived from the extensive literature review presented in Chapter 3;  Scenarios should be compliant with the initial operational needs collected during the preliminary focus group. Furthermore, the method for STRESS scenarios selection includes some essential and unavoidable assumptions and limitations. Both are often necessary to provide a frame for the validation process [76], but, they can also have a powerful effect on the conclusions of the analysis that should not be underestimated. In STRESS, assumptions and limitations derive from:  Platform capabilities which could impact on the scenarios implementation;  Capability to measure the neurophysiological indicators identified to measure human performance. In this chapter we further explain the selection criteria and the assumptions and limitations.

4.1.1 Compliance with STRESS objectives

STRESS has three main objectives (as explained in Chapter 2): 1. Define the factors composing the Human Performance envelope in future SESAR scenarios, mapping the relevance of Human Factors concepts on the characteristics of the scenarios; 2. Identify neurophysiological indicators for monitoring in real-time the controllers’ mental status;

EDITION 00.02.00

3. Develop automation design guidelines to support Human Performance during safe transitions from the higher levels of automation to the lower levels of automation, and vice versa. Accordingly to these objectives, scenarios shall: I. Elicit the relevant Human Factors such as stress, workload, attention, level of control in the context of future SESAR framework in which are foreseen less tactical interventions, high automation support, multi-sector operations, and so on ; II. Allow to monitor in real-time the controllers’ mental state in order to measure the identified neurophysiological indicators; III. Offer the possibility to deal with different levels of automation in order to develop specific guidelines for supporting tasks migration or new tasks allocation and so on.

4.1.2 Compliance with the literature review

To identify the most appropriate scenarios for our project, we browsed several documents and repositories. In particular:  With regard to automation and future changes in ATM, we reviewed several SESAR, EUROCONTROL, ICAO and EASA reports. Furthermore, we took into account the latest global demonstration activities, Skybrary Repository and research projects in order to obtain an appropriate and, as much as possible, updated picture of these aspects in relation to HP (see Chapter 3 Sections 3.1 and 3.2)  With regard to HP and HF concepts relevant in future ATM scenarios, we started from the role and significance of stress, attention, mental workload and cognitive control recognised by EASA. Then we investigated each individual factor taking into account the psychological studies and how the specific factor may impact on the HP and vice versa (see Chapter 3 Section 3.3) Furthermore we started a fruitful collaboration with different H2020 project dealing with similar topics such as AUTOPACE (Facilitating the AUTOmation PACE) project [77] dealing with psychological modelling to predict how future automation would impact on ATCOs’ performance and to identify competences and training to cope with the effects of automation on humans and MINIMA (Mitigating Negative Impacts of Monitoring high levels of Automation) project [78] aiming at understanding and mitigating “out of the loop” phenomena of ATCOs in highly automated environments especially Terminal Manoeuvring Areas (TMA).

4.1.3 Compliance with the initial operational needs

To select the most appropriate scenarios for the project, we preliminary discussed feasible scenarios with operational personnel. In particular, Anadolu (AU) contacted Turkish Air Traffic Controllers’ Association (TATCA) for sharing STRESS project objectives and research plans. It is vital to have real inputs from operational environment and ATC expert’s view and experiences about automation and its effects on human performance when developing realistic scenarios for the project purposes. Focus group meetings provide advantage to the researcher for reaching more people about the research [79]. Focus group meeting is defined as pre-planned meetings and discussions in the

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

positive environment [80]. Additionally, the selected and collected together participants or experts by the researchers provide opinions from their experiences [81]. Meetings are performed for 1-2 hours with the group of 6-10 people [82]. During the meeting, researcher can ask the questions participants individually or create a discussion about the subject. In this respect, researcher can have the qualitative information about participant’s experiences, views, trends, ideas, feelings, attitudes and behaviours. The important point is to provide an opportunity for the participants giving their own ideas and opinions without any effect from others. The opinions, expectations and experiences of participants are evaluated and used to create patterns helping the research [83]. AU and TATCA organized an online focus group workshop meeting at 10:30 CEST 23rd of September 2016, since controllers are working in busy Istanbul Atatürk ATC centre and tower. The 7 participant controllers from Atatürk tower and radar control environment where is busiest ATC sector in Turkey serving between Europe and Asia air traffic breaking records every year. This year Atatürk airport traffic reached up to 1500 traffic in a day with limited airport environment. AU defined semi-structured questions about ATC automation performance and controller performance. Before the focus group study, participants were informed about the STRESS project and objectives by the moderator Dr. Turhan. Then participant controllers were asked to answer questions and discussions were managed by AU experts. Controller’s involvement and motivation were very positive considering the STRESS project is focusing on their needs in operations and technology. The focus group study shows that participant controllers who are working Istanbul ATC facilities are facing some problems with automated tools/systems. They reported that automation positively affects air traffic safety and capacity when it works correctly. On the contrary, it affects their workload and situational awareness negatively. They get stressed and face with complexity and high workload (see Appendix 1: AU _TATCA Focus Group). The STCA failures, radar data failures, label failures and relatively identification difficulties are reported (see Chapter 5 Section 5.3.2). Beside these failures, heavy air traffic and complexity levels in their environment are another contributing factor for their performance on the job related to stress. This study shows that; the controller’s experiences are very important to create realistic experimental scenarios (see Chapter 7).

4.1.4 Assumptions and limitations

Assumptions are an essential and unavoidable element of each research activity. They are often necessary to provide a frame for the validation process, but, they can also have a powerful effect on the conclusions of the analysis that should not be underestimated. The introduction of unmotivated assumptions in analysis and evaluation is a widely recognised issue in the scientific literature [84]. During the project lifecycle and in the planned simulations, we will adopt some assumptions. The most important are listed in the following together with a brief discussion about their impact on the validation (this will be further explain in the dedicated deliverable [76]).

EDITION 00.02.00

 The selection, implementation and analysis of the use case and scenarios can be limited to the portion of the ATM system compliant with the STRESS objectives and the platform capabilities (see Chapter 4 Section 4.2).  The selection, measurement and analysis of neurophysiological indicators can be limited to the ones compliant with the STRESS objectives and the tools availability (see Chapter 4 Section 4.3). Both are essential to limit the extent of the analysis at a manageable size. These assumptions have no influence on the validation as long as any comparison (e.g. comparison of the STRESS validation activities outcomes with Subject Matter Experts (SME) or ATCOs feedback) is done with several simulations considering the same portion of the ATM world and involving the ATCOs personally. Besides, the opinion of the SMEs is always used to redefine and evaluate the project activities. 4.2 Environments and platform capabilities

In the framework of STRESS, experimental environments to perform the validation activities are provided by Anadolu University and ENAC. Below detailed explanation of their platforms capabilities.

4.2.1 At Anadolu University

Air Traffic Control Department of The Faculty of Aeronautics and Astronautics, Anadolu University has been teaching future air traffic controllers by advanced simulation technology. System provides realistic ATC environment in which students can develop their capabilities by experienced instructors who are academic researchers holding ATC licences. Faculty has two simulator environments which are 3D tower (aerodrome) simulator and radar simulator. Both systems are operated by the same software named BEST. Radar simulator system is used also for the non-radar approach and area control procedures. Tower and radar simulators have been established in the same building that can be operated closely. Both systems plan can be seen below.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 9: AU simulator laboratoires plan

4.2.1.1 Available environments AU System consists of:

 BEST Simulation System

 BEST/IMAGINE Image Generator

 BEST Fast Airport Builder

 BARCO Visual Display System

EDITION 00.02.00

Figure 10: Architecture of AU Simulators Radar simulator consists of:

 BEST System Manager – hosted on Microsoft Server 2003

 12 x Controller positions

 12 x Pilot positions

 1 x Supervisor position

 1 x Data Preparation position

Radar simulator provides training on approach and en-route control procedures for both radar and non-radar control. The controller positions are multi-functional to be used for different exercises simultaneously. While some of the controller positions are used for student training purposes, the remaining controller positions can be used for running different exercises such as research purposes. Wind and other meteorological conditions can be modified at any time by the supervisor while exercise program is running. User-friendly interface for the creation of user-defined airspaces and traffic exercises.

Radar simulator environment layout can be seen below:

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 11: AU radar simulator layout

Figure 12: AU radar simulator working positions

EDITION 00.02.00

Figure 13: AU radar simulator CWP Tower simulator consists of:

 BEST System Manager – hosted on Microsoft Windows XP

 4 x Controller positions

 2 x Pilot positions

 1 x Supervisor position

 1 x Data Preparation position

 1 x FAB position

Tower system provides realistic aerodrome image with 360 degree and 3D view. System has realistic aircraft and operational performances by aircraft library. All weather conditions can be simulated with visual perspective. Emergency conditions can be simulated during the exercise planning and running system has 6 different airport layouts including all busiest airports of Turkey such as Istanbul- Atatürk (third European airport in 2015, before Frankfurt and Schiphol Amsterdam), Ankara- Esenboğa, Antalya, Izmir-Adnan Menderes, Dalaman and Anadolu airport. Airport layouts are created by FAB (Fast Airport Builder) provides modifications and new development of any type of airport. Pseudo pilot position can be extended by assigning position from radar pilot positions. Very effective airport and air traffic scenarios can be created as well as testing even emergency for purposes and dangerous situations in the air and on the ground.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Tower Visual System comprises:

 10 Image Generator PCs

 10 BARCO Projectors

 360-degree screen

 Space Navigator

 Joystick

Tower simulator system layout can be seen below:

Figure 14: AU tower simulator layout

EDITION 00.02.00

Figure 15: AU tower simulator working positions Other generic features of the system can be listed as:

 Controlled by BEST System Manager can configure the system to assign airspace, exercise, weather and functions to positions. Data and System Backup/Restore is available. All controller/pilot positions run the same software. Working Positions can be configured to be any function as controller, pilot and supervisor. System can be configured to run one exercise group (circuit) or many. Each exercise controlled by the Supervisor position.

 BEST System Manager has facilities to backup and restore: Data files (Access database and other general files); Airspace data files (text files); Recording archives; Configuration files; Exercises; System.

 Back up ‘Data Files’ archives: Microsoft Access database; Communication system configuration; Strip format data; Other general (non airspace specific) data

 Stored in two formats within Data Preparation tool, data is stored as Access relational database; Data is exported to text file format prior to running in simulation to: Improve speed of loading; Allow data to be archived efficiently; to assist error processing.

4.2.1.2 Possible scenarios Scenario components of Simulator system.

Running a scenario requires three database parts to be selected for use:

 Airspace

o The Airspace database holds all permanent data for a scenario

o Airports, Runways, Approaches, Standard Instrument Departure Routes (SIDs), Standard Arrival Routes (STARs) and Missed Approaches

o Fixes

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

o Airways

o Aircraft Performance Data

 Exercise

o An Exercise contains the traffic sample (flight plans) to be used for a scenario

. i.e. a series of flights which will become active at the start, or, during the exercise

o and includes the definition of

. the start conditions for the flight

. the start time of the flight

. the flight plan data for the flight

. automated commands for the flight (Scripts)

 Weather

o The weather conditions that are to be applied to the traffic sample

. the initial weather conditions

 Temperature / Dew point

 Wind Data, both surface and upper layers

 Cloud

 Visibility

 the initial landing and departure runway configurations for each airport

 any pre-programmed changes for the weather or runways (Scripts)

Script Management and Pseudo Pilots

Exercise Scripting is the practice of predefining events that will automatically take place during the running of an exercise, with no need for Pilot or Supervisor command input. Such events can range from radar site serviceability changes to changes in the weather or the routing of an individual aircraft. These scripts are saved as text files that form a component of BEST airspace and exercise data and are automatically reapplied each time a simulation is run or may be invoked dynamically whilst an exercise is running.

EDITION 00.02.00

Script files may be edited whilst running an exercise, via the Data Preparation application or independently of BEST using a Windows text editor.

The BEST scripting facility is text file based. These text files are read when any exercise is run and all relevant script items are then applied. Script files can be created and used for airspace, exercise, weather and flight plan events – each being defined as a separate file.

Supervisor and Pseudo pilot positions can pretend by giving verbal information to the controller and manage flights showing any failure. Controller can observe automated system failure on his display depending on the scenario. For instance, intentionally flights can be managed by fake flight levels, directions, breaching of boundaries, reporting ext.

4.2.1.3 Implementable Automation AU radar simulator system provides configurable automation alerts to help controllers to detect any collisions between obstacles and air traffics. These are:

 APW

 STCA Alerts

 MSAW Alerts

 MTCD Alerts

 Audible Alerts

4.2.1.3.1 Area Proximity Warnings

All correlated flights have their paths projected to establish whether or not they are going to enter a restricted area. Correlated flights are those flights where the identifier squawked by the aircraft matches that held in a valid flight plan. In addition, the aircraft must be in one of the following Flight Data Processing (FDP) sector states: assumed, coordination, transfer initiated, or redundant. As relatively short time periods are used to establish whether or not the flight will breach the restricted area, radar returns are used rather than the flight plan data.

The distance projected forward is dependent on two factors: the speed at which the aircraft is travelling and the time period required for the projection. This time period relates to the amount of warning required prior to an aircraft entering a restricted area.

For example, consider an aircraft travelling at 450 knots and the APW probe time has been set to 120 seconds.

The distance projected in front of the aircraft will be: (450/3600)×120=15 Nm

The flight profile of the aircraft is taken into account when probing ahead of the flight and this can affect when an APW alert is generated. When the radar sweep occurs both aircraft are at the same point in the airspace. It can be seen in the above diagram that the aircraft in level flight will breach the restricted area within 120 seconds, this will generate an alert. However, the aircraft on a descent profile remains outside the restricted area in the next 120 seconds and an APW alert will not be generated until the next radar sweep occurs. 56 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

The aircraft label differentiates between an aircraft that is about to enter a restricted area, and one that has entered the restricted area.

Label Format

Aircraft about to enter a restricted area

Aircraft has entered a restricted area By default, the APW alert is displayed in red text or yellow text (as appropriate) in line 0 of the aircraft label. These colours can be changed as required within a configurable file.

4.2.1.3.2 STCA Alerts

The STCA is used to predict if the minimum lateral and vertical separation criteria will be breached in the near future. As this is a relatively short time period, the radar returns as used rather than the flight plan data. An STCA area has specific separation values and all flights within an STCA area are tested according to these values. Where the two flights in conflict span two STCA areas, the larger separation values are used for processing. Where one of the flights is in an STCA inhibited area and the other is not, STCA processing is still performed.

Each correlated flight in the STCA area is compared against all other Mode C or Mode S equipped flights. A correlated flight is one where the identifier squawked by the aircraft corresponds to a valid flight plan for the exercise that has an ICAO route defined. In addition, the aircraft must be in one of the following FDP sector states: assumed, coordinated, ongoing coordinated, transfer initiated or redundant.

First, a test is performed to see if a lateral conflict is detected. Where this is not the case, no further processing is required. Where a lateral conflict is established, the system then checks the vertical component of the two flights. For each STCA volume created within the airspace lateral and vertical separation parameters are configured. The lateral parameter defines an aircraft separation distance in nautical miles. Where the distance between two aircraft is below this value and the vertical separation parameter is also breached, an STCA alert is generated.

When the shortest distance between two aircraft is less than the separation distance configured for the STCA area that the aircraft are within, the flights pass the lateral separation test for an STCA alert.

Once the system has detected that the minimum lateral separation distance has been breached, the vertical component is analysed. The STCA area is divided into three level bands. A vertical separation

EDITION 00.02.00

distance is assigned to each band. Where the minimum vertical separation distance is breached, an STCA alert is generated.

Label Format

4.2.1.3.3 Minimum Safe Altitude Warning Alerts

An MSAW is a minimum safe altitude warning. This MSAW Area Editor dialog allows the configuration of a volume that extends up from the ground to the minimum safe altitude. A background MSAW exists for the entire airspace which is set to 3000 feet by default. The background value is configured in the Airspace Editor dialog within the Data Preparation application.

The aircraft label indicates when an aircraft is predicted to breach or has breached the minimum safe altitude. The probe time ahead of a flight with regard to MSAW detection is configured set to 2 minutes by default.

Label Format Aircraft has fallen below, or is predicted to fall below the minimum safe altitude

By default, the MSAW alert is displayed in red text in line 0 of the aircraft label. This colour can be changed as required within a configurable file.

4.2.1.3.4 MTCD Alerts

MTCD interrogates the projected routes of those aircraft currently within the sector and those that will enter the sector to establish whether or not the flights will come into conflict. Only correlated flights are subject to MTCD processing. In the event of a conflict being identified, the MTCD List will be displayed automatically to alert the controller to the conflict.

MTCD is an iterative process that determines whether or not an aircraft’s route conflicts with other aircraft flying through the same MTCD region. The process has six stages that are carried out in the order described below:

1. MTCD is only performed on valid flights. To qualify as a valid flight, the following conditions must be met:

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

a. The flight must not be squawking an SSR code of 2000 (by default, the system excludes aircraft squawking 2000 from MTCD processing).

b. The flight must have a full flight plan, that is to say the ICAO route field is completed.

c. The flight must be airborne.

d. Must not be in manual mode.

e. Must not be under Visual Flight Rules (VFR) Only flight rules.

2. The times at which the aircraft are present in the MTCD region are considered. Those flights whose times in the MTCD region do not overlap are not processed against one another.

3. Once an aircraft enters an MTCD region, the entire route through the MTCD region is compared against the routes of all other flights within the MTCD region. Aircraft outside the MTCD region are not considered in any way.

4. Where it is determined those two flights may potentially conflict, both routes are broken down into 1-minute sections and the routes are reanalysed. If the two flights conflict within the same 1-minute section, further processing is required.

5. The 1-minute section in which the conflict occurs is further broken down into 30-second sections and the routes are reanalysed. If the two flights conflict within the same 30-second section a horizontal conflict has been established.

6. The FDP sectors that the horizontal conflict occurs in are checked to establish whether or not a vertical conflict is also present. The following checks are performed to establish if a vertical conflict is also present.

a. If both flights are assumed in the same sector, the Actual Flight Level (AFL)–Cleared Flight Level (CFL) bands are compared for both flights. Where the bands overlap or breach the vertical separation limit, a conflict alert is generated.

b. Where one flight is assumed, its AFL–CFL band is compared against the Planned Entry Level (PEL)– AFL band for the flight in the other sector state1. Where the bands overlap or breach the vertical separation limit, a conflict alert is generated. If any of the stages 3–5 detailed above fail to detect a conflict, the stages that follow are not carried out. When both a horizontal and vertical conflict occur an MTCD alert is issued.

c. The other sector state that is considered is: coordination. Where a flight has the sector state ‘none’ it will not be compared with any other flight to determine a horizontal conflict.

4.2.1.3.5 Audible Alerts

The system supports Route Adherence Monitoring (RAM). This feature generates an alert when the flight deviates from the ICAO route detailed in the flight plan. A number of parameters are

EDITION 00.02.00

configured within the ANADOLUFDP.ini file that determine the deviation tolerance that must be exceeded to generate an alert.

Label Format The RAM alert is displayed line 0 of the aircraft track label.

The system supports Cleared Level Adherence Monitoring (CLAM). This feature generates an alert when the flight deviates from the specified cleared level for the current route leg as detailed in the flight plan. A parameter must be configured within the ANADOLUFDP.ini file to determine the deviation tolerance that must be exceeded to generate an alert.

Label Format

The CLAM alert is displayed line 0 of the aircraft track label. Audible Alerts

An audible alert will sound when a flight enters any of the following alert conditions: MSAW, STCA, MTCD or APW, or when the aircraft is squawking any of the following emergency codes 7500 (hijack), 7600 (Radio Transmission (RT) failure) and 7700 (emergency). When any alert condition is active, an audible alert will sound at the controller position that has the flight assumed and in the case of an emergency condition at all other controller positions. The audible alert can be silenced via the Silence Alerts button on the toolbar.

Silencing an Audible Alert

Click the Silence Alerts button on the toolbar. All audible alerts are silenced but the aircraft remain in an alert condition. Any alerts that are generated subsequent to this action will result in an audible alert.

Temporary Segregated Airspace

A Temporary Segregated Airspace (TSA) is a no fly zone that can be active 24 hours a day or during a specified time period. The active period is defined by a script (see Exercise Scripting manual). Usually TSAs are created over military areas (e.g. bases, weapons ranges), prisons, nuclear power stations, oil refineries, etc. The TSAs are linked to the APW functionality which generates a warning when the aircraft is approaching a TSA.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

4.2.1.4 Subject availability Validation studies will be performed in the AU simulators. Anadolu University has been recruiting 15 students every year to be involved in training as future air traffic controllers hence they can recruit participants amongst their students. The participant students will be at different training levels such as tower, radar or non-radar experienced and all positions.

Recruitment of operational controllers is also a possibility. AU will invite controllers from Turkish ANSP (DHMI), from different ATC centres and units. Also Turkish Air Traffic Controller’s Association will be asked to support the validation experiments. The experiments will be performed in AU’s ATC simulator facilities.

Table 3: Subjects Availability Subjects Availability

Session TBD

Delivery mode On Faculty class and simulators

Simulator positions TBD

Supervisor 1 AU expert

Instructors AU Instructors_3

Experts/Observers AU+UniSap+ENAC

Pseudo Pilots AU_2

Participants 15 students_15 controllers !!!!

4.2.2 At École Nationale de l'Aviation Civile

Beside of its training simulators (considered as operational), ENAC possessed a complete range of ATC research simulators, from ground to en-route. ENAC research facility can support Technology Readiness Level (TRL) from level 4 to 6 (from serious games in laboratory settings to high fidelity research simulators). These simulators allow controller pilot communication with headphones and usual hardware. Simulations can be run with standalone tools that will only require an air traffic controller or in a complex and realistic setup with pseudo-pilots and adjacent sectors operated by pseudo-air traffic controllers. Traffic and simulation environment such as airspace, controlled sectors or airport are designed with the assistance of seniors air traffic controllers with a high level of realism.

En-route simulations ENAC’s en-route simulator working position (WP) is stripless and composed of two screens per role (executive and coordinator controllers). The upper screen displays radar visualization and lists showing assumed aircraft and incoming aircraft filtered by entry waypoint. There is neither mouse

EDITION 00.02.00

nor keyboard to interact with the radar visualization, all the required interaction like flight designation and clearance processing are made using an interaction area displayed on the lower screen (a Wacom touch input device and screen). This lower screen is also used to display additional information like an extended electronic flight strip if an aircraft is selected on the radar visualization. We can run simulations in a simplified setup with only the executive controller position (left part in the Figure 16) or with both positions but an additional pseudo air traffic controller is required to process the aircraft in adjacent sectors.

Figure 16: Prototyping platform at ENAC. The standard WP is composed of two screens, one 30" (2560x1600 pixels) for the radar visualization and a 21" Wacom screen (Wacom 1600x1200 pixels) including a stylus pointing device. A new position is currently developed using a 32” 4K radar display and a 27” (2560x1600) interaction display.

A screenshot of the radar screen of the prototyping platform’s Human Machine Interface (HMI) is shown in Figure 17. The radar screen displays the sector (light green), routes, waypoints and flights according to their coordination state (white are assumed). Information about incoming flights is displayed in list. Interactions on the radar visualization are done with the Wacom stylus on the black area displayed on the lower screen which acts as a touchpad for the upper screen.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 17: RADAR screen Selecting an aircraft on the radar visualization triggers a pie menu. This pie menu allows the air traffic controller to display information on the selected flight or enter clearances.

Figure 18: WACOM screen

EDITION 00.02.00

The interaction screen (Wacom represented Figure 18) displays the flight plan information and allows displaying multiflight information straight on the flight on the radar screen or changing zoom/position/speedvector on the radar visualization. The large black pad is the indirection area that is used to interact with the radar image.

The pseudo-pilot position for en-route working positions will allow us to have only one pilot for each simulation. This will not be limitative as the pilot will be able to control up to thirty aircraft. S/he will use the interface presented in Figure 19. The main view shows the selected aircraft navigation display. On the left side, there is all aircraft managed or that will be managed by pilot with the selected aircraft highlighted. The bottom of the screen is dedicated to scheduled actions (define at scenario creation) requested from pilot for the different aircraft.

Figure 19 Pseudo-pilot HMI display Tower environment The Ground Control Working Position New automated techniques are being developed, aiming at saving fuel during the ground taxiing phase. Although the environmental benefit would be interesting on its own, technologies such as the TaxiBot© (autonomous tugs) system may also increase the number of ground movements, or the throughput. A previous project, MoTa (Modern Taxiing) project [85], deals with providing ground ATCOs a tool that will help with managing increased traffic and taking advantage of modern aircraft taxiing techniques when available. 64 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

The tool consists of an integrated ground control interface featuring the latest progress in modern taxiing methods and multi-agent algorithms for enhanced ground automation while still supporting current and conventional ground control procedures during the transition period.

The setting of the ground controller working position is composed of an external view representing the view from the tower. Our simulator is set to work on the south ground sector at Roissy Charles de Gaulle airport (CDG).

Figure 20: Ground controller working position The control interface is based on the AVISO (the ground radar image currently in use at CDG) but it was enriched to include information from the paper flight strips, thus capable of replacing the paper strips entirely in ours simulations. Together, these two technologies provide the minimum information required to manage today’s ground taxiing operations.

EDITION 00.02.00

Figure 21: MoTa ground controller interface prototype, as in use for the South ground sector at CDG.

We will briefly describe the HMI presented in Figure 21, flight information is displayed on the aircraft label and a concealable side panel presents a flight list. The standard path suggestion (calculated from airport's traffic flow rules) for an aircraft can be retrieved by selecting its icon or label (i.e. a stylus touch). As seen in the Figure 1 inset, ACA1609 is departing on runway 26R and the ATCO can validate the suggested path (marked in yellow) by clicking on one of the 3 holding points to the runway (represented by the large green zones).

The pseudo pilot interface is very similar to the one used for controller. The number of pseudo pilot position requested for the simulation depends on choices made for simulation.

4.2.2.1 Possible scenarios This paragraph will detail the possible scenarios of the two environments.

Equipment for en-route scenarios is already available, an upgrade of this hardware could be available later and the project will be able to use it. The position will allow us to measure two persons at the same time. Pseudo-pilots availability must be planned at least one month ahead of the simulation. One pseudo pilot will be necessary for each controller. So if we record two controllers at the same time we will also require two pilots.

Equipment for tour scenarios is already available. If we take as a constraint to have a very large external visualization we will have only one simulation each time. But this is not mandatory, we could use a smaller one and this way we could provide at least two positions for simultaneous recording. Pseudo-pilots requirement will be strongly linked to scenario’s traffic and automation. It could go from no pseudo-pilot up to three per simulated position.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

4.2.2.2 Implementable Automation As stated above, our interests concern highest level of automation. There are different automation for each environment, some of them are easier than others to implement.

En-route automations that could be used are conflict detection. The presentation of such result is presented as a vertical time line on the side of the radar image. Each label represents a potential conflict, as time goes on the label goes down and the conflict is happening when the label reach the bottom of the timeline.

Figure 22: Vertical Time line with conflict detection displayed A previous project was using an automation tool called Traffic Control using Speed Adjustment (TCSA). This tool is able to slow or speed up some aircraft in order to make the traffic easier to flow. The speed adjustment is sent to the aircraft. The controller is informed of those speed adjustments and will still be able to manage the traffic the way he wants. S/he will be asked to avoid as much as possible to give clearances to those specific aircraft, but he still decides.

Two different high automation tools can be used for Tour Scenarios, the first one displays conflict detection between aircraft (or other vehicle). The second tool is a multi agent system which will pilots tugs once they are detached from aircraft.

In the right side of the figure below, two aircraft are highlighted because of a potential crossover. AF626BV has been instructed to turn right while ACA1609 is going straight ahead and neither of them has been told to give way to the other.

EDITION 00.02.00

Figure 23: Conflict detection on left side, autonomous tug on right side. The tugs used for aircraft pushback could be used to continue towing the aircraft along the taxiways until the runway holding point. In that manner, a departure aircraft would be handled as usual by ground control, but when the tug is detached from the aircraft after depositing it at the runway, automation takes place as the empty tug would return to the parking areas without controller clearances.

4.2.2.3 Subject availability In both environments, subjects can be recruited in controller students; we have to inform the head of trainers’ team as soon as possible. We could also try to recruit instructors, but there will be fewer subjects available. 4.3 Neurophysiological Indicators

As anticipated in section 3.3, stress level and emotional engagement, workload and level of cognitive control, attention focus and supervisory attention are human factors issues of relevance in the future ATC scenarios. The STRESS project is planning to develop neurophysiological indicators for each of them. This entails two synergic steps: the customization of existing neurophysiological indicators to future ATM, in order to measure the HP envelope with neurophysiological indicators.

4.3.1.1 Stress A number of physiological markers are widely used for stress assessment, including: hormonal secretion, galvanic skin response, blood pressure, several features of heart beat patterns, and respiration activity. The classical biochemical markers for stress are the cortisol and adrenaline levels, which will increase rapidly to stress exposure. Certainly, to monitor such biomarkers in operational environments would be unfeasible, because of several issues related to the invasiveness and real-time applicability of such measure. Skin sweating, measured in terms of Galvanic Skin Response (GSR), is a well-accepted indicator of reticular activation and, therefore, of emotion and cognition. A transient increase in skin conductance is proportional to sweat secretion. When an individual is under mental stress, sweat gland activity is activated and increases skin conductance. Since the sweat glands are also controlled by the sympathetic nervous system (SNS), skin conductance acts as an indicator for sympathetic activation due to the stress reaction [86].

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Another interesting stress measure is the fluctuation of heart rate over time. Heart rate variability (HRV) measures frequency fluctuations across time and reflects the autonomic balance, i.e. whether it is the SNS or the Parasympathetic Nervous System (PNS) that is dominating. Spectral analysis in the frequency domain enables a separation between vagal (PNS related) and sympathetic cardiac control to be made. Stress usually causes a decrease in HRV. It sounds not strange that a decrease in HRV, in particular the 0.1 Hz component, is associated also with increased mental effort [87]. The higher frequency components (HF, 0.15-0.4 Hz) are mainly related to respiratory influences and solely controlled by PNS. There is also a low frequency component (LF, 0.04-0.15 Hz) and a very low frequency component (VLF, ≤0.04 Hz), which are controlled by both SNS and PNS. Some studies have also analysed the LF/HF ratio and observed that an increase is associated with mental stress [88]. The validity of using Electrocardiography (ECG) and GSR measurements in mental stress monitoring has been demonstrated in both psychophysiology and bio-engineering. HRV analysis based on ECG measurement is commonly used as a quantitative marker describing the activity of the Autonomous Nervous System (ANS) during stress. For example, Sloten et al [89] conclude that the mean R-R (distance between two R-peaks) is significantly lower (i.e. the heart rate is higher) with a mental task than in the control condition. Also, conventional short-term HRV features (e.g., a 5-minute sample window) may not capture the onset of acute mental stress for a mobile subject. Salahuddin et al. [90] noted that HR and RR-intervals within 10 sec, high frequency band (HF: 0.15 to 0.4 Hz) within 40 sec, LF/HF, normalized low frequency band (LF: 0.04 to 0.15 Hz), and normalized HF within 50 sec can be reliably used for monitoring mental stress in mobile settings [90]. Hence, mental stress can be recognized with most HRV features calculated within one minute. Boucsein [91] provided an extensive coverage of early research of GSR related to stress. There are two major components for GSR analysis. Skin conductance level (SCL – phasic component) is a slowly changing part of the GSR signal, and it can be computed as the mean value of skin conductance over a window of data. A fast changing part of the GSR signal is called skin conductance response (SCR – tonic component), which occurs in relation to a single stimulus. He showed that slowly changing SCL and SCR aroused by specific stimulus are sensitive and valid indicators for the course of a stress reaction. However, to measure physiological signals and to use them as biomarker of cognitive processes in ecological environments, and in general during everyday activity, is more difficult than in a rigorous laboratory environment. First, the physiological responses caused by mental stress can be masked by variations due to physical activity [92]. For example, people may have higher heart rate when standing than when sitting. Hence, using heart rate alone as an indicator to detect mental stress may lead to misclassifications. Second, signal artefacts caused by motion, electrode placement, or respiratory movement affect the accuracy of measured recordings. In general, it is crucial to control subjective variables, such as physical activity, posture, breathing and speech, as well as environmental variables, such as temperature and air humidity, which will affect the biosignals variability.

4.3.1.2 Attention Neuroimaging has yielded information on the integrated brain activity underlying visual attention in humans. Studies documenting the neural correlates of attention have used several techniques, among them functional magnetic resonance imaging (fMRI), electroencephalogram (EEG), and magnetoencephalography (MEG, [45]). In particular, we will take into account the EEG technique, that allows the use in realistic environments with respect to the other techniques, because its high

EDITION 00.02.00

temporal resolution and portability, compared to other neuroimaging techniques (fMRI, MEG, etc.). There are literature evidences demonstrating that it could be possible to measure variations of few attentional correlates by using EEG activity, in particular the P300 Event Related Potential (ERP) and EEG frequency bands.

4.3.1.2.1 Event Related Potentials (ERPs): P300

Event-related potentials (ERPs) represent the voltage fluctuations that are associated in time with some physical or mental occurrence [93]. Specifically, the event related potential (ERP) P300 is a positive deflection of the EEG signal elicited in the process of decision-making [94]. More specifically, the P300 potential arises when higher-order cognitive operations related to selective attention and resource allocation are engaged [95]. In other words, the P300 potential could be considered as a measure of the attentional resources of the subject. In particular, the amplitude of the P300 is proportional to the amount of attentional resources engaged in processing a given stimulus [96] and it is not influenced by factors related to response selection or execution [97]. Gray et al. [98], reported that the P300 amplitude therefore served as our covert measure of attention that arises independently of behavioral responding. Further, P300 latency is thought to reflect stimulus classification speed, such that it serves as a temporal measure of neural activity underlying attention allocation and immediate memory operations [99]–[101]. Kutas and colleagues [102] first suggested that the P300 latency can be used to index the duration of specific subcomponent processes that underlie attentional resource allocation.

The ERPs serve as important adjuncts to studies of human information processing, a fundamental problem with this method is the signal-noise ratio. The magnitude of the ERP signal is around 5-10 µV, which is far smaller than the amplitude of the background EEG (0-100 µV; [103]). Therefore, the classic approach for ERP extraction is to average the signal over a number of trials in order to obtain a stable response with a sufficient signal-to-noise ratio (SNR). In this regard, the University La Sapienza (UNISAP) group have been developed a method able to improve the SNR and to extract the P300 potential at a single stimulus timing [104].

4.3.1.2.2 EEG frequency bands

During a cognitive process, the electrical activity of neurons populations of a specific cortical area is synchronized, causing a measurable increasing of that brain specific rhythm over the scalp. Therefore, the analysis of the increase and/or decrease of the EEG frequency spectral content of specific cortical areas, for instance in our case those ones involved in attentional processes, could provide crucial information to monitor brain activity. In particular, literature evidences showed that an increase of EEG activity, especially over the frontal cortex in the theta band, has been observed when the demand of executive control, reflected by stronger levels of attention and working memory, is high [105]. Concerning the alpha EEG rhythm, many evidences demonstrated how the structures of the thalamus are involved in exerting attentional bias [106]. A measure of support for this notion comes from imaging studies reporting attentional modulations in the pulvinar nucleus and in the lateral geniculate nucleus [107]. In the work of Neuper and Pfurtscheller [108], it has been shown that the Event-Related Desynchronization (ERD) of EEG activity in the alpha band reflected an increased excitability level of neurons in the involved cortical areas, which could be related to an enhanced information transfer in thalamo-cortical circuits. In contrast, Event-Related Synchronization (ERS) of alpha activity (i.e. increases in alpha activity) is thought to reflect a reduced state of active information processing in the underlying neuronal networks [109]. Therefore, alpha 70 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

power decreases when individuals become engaged in the task and increase their attentional level, as compared to the simple retention of information [105]. As a result, we hypothesize that the alpha EEG band, especially on the frontal sites, could be used as a high-informative feature to discriminate the different attentional levels adopted by the operator during its operational activity.

4.3.1.3 Mental workload Many neurophysiological measures have been used for the mental workload assessment, including Electroencephalography (EEG), functional Near-InfraRed (fNIR) imaging, functional Magnetic Resonance Imaging (fMRI), Magnetoencephalography (MEG), and other types of biosignals such as Electrocardiography (ECG), Electrooculography (EOG) and Galvanic Skin Response (GSR) [110]–[112]. The size, weight, and power constraints outlined above limit the types of neurofeedback that can be used to realize applications usable in real contexts. For example, fMRI [113] and MEG techniques require room-size equipment, thus they would not be portable. EOG, ECG and GSR activities highlighted correlations with some mental states (stress, mental fatigue, drowsiness), but they were demonstrated to be useful only in combination with other neuroimaging techniques directly linked to the Central Nervous System (CNS), i.e. the brain [114]–[116]. Consequently, the EEG and fNIR are the most likely candidates that can be straightforwardly employed to realize applications usable in operational environments.

Regarding the EEG measurements, most part of the studies showed that the brain electrical activities mainly considered for the mental workload analysis are the theta and alpha brain rhythms typically gathered from the Pre-Frontal Cortex (PFC) and the Posterior Parietal Cortex (PPC) regions. Previous studies demonstrated as the EEG theta rhythm over the PFC present a positive correlation with the mental workload [117], [118]. Moreover, published literature stressed the inverse correlation between the EEG power in the alpha frequency band over the PPC and the mental workload [114], [119]–[123]. Only few studies have reported significant results about the modulation of the EEG power in other frequency bands, i.e. the delta, beta and gamma [114], [120], [124]. More specifically, most of the studies are focalized on the EEG power modulation occurring in theta (4 – 8 Hz) and alpha (8 – 12 Hz) frequency bands, usually associated with cognitive processes such as working memory and attention, typically involved in mental workload. Onton [125] reported that the frontal midline theta rhythm increases with memory load, confirming previous results about the correlation between the frontal theta EEG activity and mental effort [120], [124]. Mental workload is also known to suppress EEG alpha rhythm and to increase theta rhythm during activity of information encoding and retrieval [122], [126], [127].

According to the idea that the higher the mental workload level is, the greater the brain blood oxygenation will be, the functional Near-Infrared spectroscopy (fNIRs) has been demonstrated to be another reliable mental workload measurement technique [128], [129]. FNIR spectroscopy is safe, highly portable, user-friendly and relatively inexpensive, with rapid application times and near-zero run-time costs, so it could be a potential portable system for measuring mental workload in realistic settings. The most common fNIR system uses infra-red light introduced in the scalp to measure changes in blood oxygenation. Oxy-hemoglobin (HbO2) converts to deoxy-hemoglobin (HbR) during neural activity, that is the cerebral hemodynamic response. This phenomenon is called Blood-oxygen- level dependent (BOLD) signal. fNIRs has been shown to compare favorably with other functional imaging methods [130] and demonstrates solid test - retest reliability for task-specific brain activation [131], [132]. Thus, the primary hypothesis was that blood oxygenation in the prefrontal

EDITION 00.02.00

cortex, as assessed by fNIR, would rise with increasing task load and would demonstrate a positive correlation with the mental workload. In fact, Izzetoglu et al. [133] indicated clearly that the rate of changes in blood oxygenation was significantly sensitive to task load variations.

Several researches in the ATM domain treated the neurophysiological measurements of ATCOs’ mental workload in realistic settings with the aim of developing HMI systems, by using both EEG and fNIRS techniques. In the following examples, it has been discussed how each technique was able to provide reliable estimations of mental workload, especially for applications in operational environments. For example, both EEG and Fast Optical Signal (FOS)-based fNIR have similar bandwidth and sample rate requirements, as the FOS appears to directly reflect aggregated neural spike activity in real-time and can be used as a high-bandwidth signal akin to EEG [134]. However, EEG and fNIRs systems have different physical interfaces, sizes, weights and power budgets, thus different wearability and usability in real operational contexts. Specifically, the physical interface merits scrutiny as it is non-trivial to maintain a good contact between the sensors (i.e. electrodes or optodes) and the brain scalp in freely-moving tasks. It is worth noticing that fNIRs is not affected by motion artifacts and does not require both scalp abrasion and conductive gel. In addition, there is not the necessity to wear a cap but only a headband. Furthermore, unlike EEG, fNIRs recordings are not affected by electroculographic and environmental electrical noise, and less sensitive to facial muscular activity, which are undoubtedly ubiquitous in human-computer interactions. Thus, fNIRs technology could appear more suitable in realistic environments [133], [135]–[137].

However, in a recent study, Harrison et al. [138], reported how the BOLD signal showed a lower resolution than the subjective measures (ISA, [139]) to evaluate the mental workload of ATCOs involved in the experiment. In particular, while the task was becoming more difficult, the subjective measure was still increasing, and the BOLD signal (neurophysiological index) reached its maximum, lingering on this value. Furthermore, the BOLD signal, used as workload index, was shown to be not reliable over time since the workload measurements performed in different days were significantly different and in discordance with the subjective measures.

In addition, since the presence of hair may impact on both photon absorption [140] and the coupling of the probes with the underlying scalp, the fNIRs technique is very reliable only on those un-hairy brain areas, like the PFC. As quoted above, the parietal brain sites also play a key role in the mental workload evaluation, and Derosière et al. [128] pointed out how some fNIRs-measured hemodynamic variables were relatively insensitive to certain changes in mental workload and attentional states.

Due to its higher temporal resolution and usability, in comparison with the fNIRs technique, the EEG technique overcomes such kind of issues. In addition, there are several studies in ATM domain that highlighted the high reliability of EEG-based mental workload indexes [119]. The results showed that the effects of the task demand were evident on the EEG rhythms variations. EEG power spectra increased in the theta band, while significantly decreased in the alpha band as the task difficulty increased, over central, parietal, frontal and temporal brain sites. In a recent study, Shou et al. [141] evaluated the mental workload during an ATC experiment using a new time-frequency Independent Component Analysis (tfICA) method for the analysis of the EEG signal. They found that “the frontal theta EEG activity was a sensitive and reliable metric to assess workload and time-on-task effect during an ATC task at the resolution of minute(s)”. In other recent studies involving professional and trainees ATCOs [142]–[144], it was demonstrated how it was possible to compute an EEG-based Workload Index able to significantly discriminate the workload demands of the ATM task by using

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

machine-learning techniques and frontal-parietal brain features. In those studies, the ATM tasks were developed with a continuously varying difficulty levels in order to ensure realistic ATC conditions, i.e. starting form an easy level, then increasing up to a hard one and finishing with an easy one again. The EEG-based mental workload indices showed to be directly and significantly correlated with the actual mental demand experienced by the ATCOs during the entire task.

The same EEG-based Workload Index was also used to evaluate and compare the impact of different avionic technologies on the mental workload of professional helicopter pilots [145]. Furthermore, the machine-learning techniques have been successfully used in other real environments for the evaluation of mental states [146] and mental workload [147], [148]. Another interesting application of the neurophysiological workload evaluation was proposed by Borghini et al. [149], where a neuroelectrical metric was defined and used for the training assessment of subjects while learning to execute correctly a new task.

Even if the main limitation of the EEG is its wearability, technology improvements [150] are being developed and tested in terms of dry electrodes (no gel and impedances adaptation issues), comfort, ergonomic and wireless communications (no cables between EEG sensors and the recording system).

In conclusion, the EEG technique seems to be the appropriate solution to evaluate the mental workload in realistic and operational settings. Such systems will support the operator during his/her working activity in order to improve the works wellness and, most of all, the safety standards of the whole environment.

4.3.1.4 Cognitive Control Behaviours: Skill, Rule and Knowledge model. One of the objectives of the STRESS project is to assess whether it is possible to differentiate the three degrees of cognitive controls proposed by the SRK model by the analysis of brain activity during the execution of proposed tasks. In particular, we aimed to characterize the three cognitive control behaviours in terms of “pure” cognitive engagement by considering only brain features linked to cognitive processes, such as information processing, decision making, working memory, and attention. In fact, we have not considered motor brain features (e.g. sensory-premotor cortex) in order to do not differentiate the SRK behaviours because of the different amount of movements (hands or feet). Also, neither the occipital lobe has been considered to avoid possible muscle artefacts due to neck contractions during the execution of the task, as they are more likely happen on the occipital channels. Brain activations are generally correlated to specific cognitive phenomena [151]. In fact, literature evidences show that an increase of EEG activity, especially over the frontal cortex in the theta band has been observed when the demand of executive control (attention and working memory) [152]–[154], the activation of decision-making processes (like resolution of conflicts and error detection [155]), problem solving demand [156], mental workload request [157]– [159], and the task complexity are high [160]–[162]. Additionally, the potential role of theta rhythms engaged in the hippocampal/PFC (prefrontal cortex) interplays in consolidation of memory and reports empirical results on the effect of enhanced theta oscillations on memory consolidation [163]–[165], and induction of long-term plasticity [166]–[168]. Therefore, we hypothesized that enhanced parietal theta activation supports exchange of information between the hippocampus and neocortical areas, hence it could be considered as indicator of memory consolidation [169], [170]. Concerning the alpha EEG rhythm, many evidences demonstrated how the structures of the thalamus are involved in exerting attentional bias [171], [172]. A measure of support for this notion

EDITION 00.02.00

comes from imaging studies reporting attentional modulations in the pulvinar nucleus [173]–[175] and in the lateral geniculate nucleus [176], [177]. In the work of Neuper and Pfurtscheller [178], it has been shown that the ERD of EEG activity in the alpha band reflected an increased excitability level of neurons in the involved cortical areas, which could be related to an enhanced information transfer in thalamo-cortical circuits. In contrast, ERS of alpha activity (i.e. increases in alpha activity) is thought to reflect a reduced state of active information processing in the underlying neuronal networks [179]. Therefore, alpha power decreases when individuals become engaged in the task and during manipulation of memory content as compared to the simple retention of information [153], [168], [180]. As a result, we hypnotized to discriminate the SRK behaviours by evaluating the degree of information processing, working memory (frontal theta EEG rhythm) [157]–[159], [161], procedural memory (parietal theta EEG rhythm) [155], [163], and attention (frontal alpha EEG rhythm) [172], [176], [180], [181] along the execution of the SRK events. In other words, we expected to characterize the cognitive control behaviours by the brain activations summarized in the following table.

Table 4: Hypothesized cognitive control behaviours characterization by means of the selected brain features. The SRK behaviours were described in terms of EEG rhythms activation: synch – synchronization; desynch – desynchronization EEG RHYTHM

COGNITIVE CONTROL BEHAVIOUR Frontal Theta Parietal Theta Frontal Alpha

Skill Lowest Synch Lowest Synch De-synch

Rule Synch Synch Synch

Knowledge Highest Synch Highest Synch Synch

4.3.1.5 Summary Along the previous paragraphs, all the Human Factors to investigate during the STRESS project have been introduced from a physiological point of view, in order to better understand the correlation with the neurophysiological evidences described in literature. Despite the several techniques investigated and adopted in the scientific literature, thanks to the expertise of the STRESS consortium partners specific technologies have been chosen, because of previous studies of the partners themselves but also of their reliability and usability in operational environment. The Table II summarizes the state of the art and the technologies that will be adopted within the STRESS project. Of course, this preliminary choice will be not mandatory, but it has to be considered as the first outcome of a thoughtful work planning activity. However, thanks to the scientific literature evidences discussed before, the partners feel confident to be able to obtain the desired neurophysiological indicators, even by using technologies different from those one already expected to be adopted.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Table 5: Summary of the chapter about neurophysiological indicators. For each Human Factor it is indicated the current state of the art, in terms of technologies able to obtain features of that specific process/state, and the technologies that will be used in the project

Human Factors State of the art STRESS project will use: Hormonal secretion GSR GSR ECG/HR STRESS ECG/HR Blood pressure Respiration activity Skin temperature EEG EEG EOG Eye-tracker ATTENTION Eye-tracker MEG fMRI EEG EEG fNIRS MENTAL WORKLOAD MEG fMRI ECG/HR EOG COGNITIVE CONTROL BEHAVIOUR EEG EEG

EDITION 00.02.00

5 Results

This chapter gives an overview of the potential operational scenarios that will be simulated in the framework of the STRESS validation activities. The overview is based on the review of the SESAR future scenarios and impacted HF issues (presented in section 3), and also takes into account compliance criteria and limitations (presented in section 4). The consolidated set of scenarios will be presented in deliverable D4.1 Validation Plan, which will describe in detail all the aspects related to the preparation of the STRESS validation exercises. 5.1 STRESS validation activities

Two validations activities have been planned. The first one will validate the links among HF concepts and neurophysiologic indexes. For each condition tested, the overall human response to the proposed scenarios will be collected, finding HF concepts configuration associated to specific levels of automation and/or to non-nominal events.

Figure 24: conceptual representation (no actual data) of the link between configurations of HF concepts and indexes to be investigated in different scenarios conditions Once the neurophysiological indexes will be validated, also the link between HP and System performance will be investigated. This will be the scope of the second validation. In order to do that, for each scenario, a performance index will be generated, based on SESAR performance Key Performance Areas (KPAs) (e.g. Predictability, Punctuality, Capacity). For example, within an approach scenario, with a highly automated supporting tools (e.g. an enhanced version of AMAN), parameters such as inter-arrival time variability, capacity, adherence to the planned and actual landing time, can be used to generate a system performance index. In this way a complete view on automations impact on future operations can be achieved (see Figure 6).

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Figure 25: impact of the generated scenarios on human and system performance 5.2 Operational environments

Both ENAC and Anadolu have the potential to simulate both radar and tower environment. According to the compliance criterion related to STRESS objectives, the scenarios should allow monitoring in real time the controller’s mental status in order to measure the identified neurophysiological indicators. This entails that:  As physical activity can mask physiological response to stress (see section 4.3.1.1), we will consider only environments which minimize it;  As continuous motion can impact on the reliability of EEG1 measurements, we will only consider environments which minimize it (for example tower operations may be excluded as they do not guarantee that controllers are seated for all the time). In line with this, we will choose an environment where the position of ATCOs is seated. Anadolu Tower environment allows controller to move freely in opposition to ENAC tower environment where the position is seated. In ENAC, both radar and en-route environments are complying with this requirement. The table below shows sums up the ENAC and Anadolu environments capabilities to measure relevant human factors through neurophysiological indicators measurement tools. Table 6: ENAC and Anadolu environments capabilities to measure neurophysiological indicators of stress, workload, attention and control level Relevant STRESS Human Factors

Site Environments Stress Workload Attention Level of control

Tower No No No No Anadolu Radar Yes Yes Yes Yes

Tower Yes Yes Yes Yes ENAC En-Route Yes Yes Yes Yes

1 The use of EEG equipment requires to maintain a good contact between the sensors (i.e. electrodes or optodes) and the brain scalp. For this reason, in freely-moving tasks the EEG measurements reliability may be affected (see also section 4.3.1.3).

EDITION 00.02.00

5.3 Potential Scenarios

Both ENAC and Anadolu have the potential to simulate the FRA concept, although in ENAC this may be limited to direct routing. FRA has been described in the literature review of future SESAR concepts (see 3.1), so its implementation would satisfy the concerned compliance criterion (see 4.1.2). It would also meet the compliance with the initial operational needs concerning questions about the reliability of automation and its impact on workload and situation awareness (see 4.1.3).

5.3.1 Automation

Both ENAC and Anadolu have already available automations to be integrated into the scenarios, and new automations, simulating future SESAR concepts can be simulated in ENAC and Anadolu facilities. Examples are provided in the table below.

Table 7. Examples of automation already simulated

The SESAR’s Extended It refers to preparing further in advance the sequencing of air traffic Arrival Management (E- destined for a particular airport. This allows controllers upstream to give AMAN) early instructions to pilots to adjust their speed before initiating descent towards the destination airport.

The National It automates portions of the Tactical Departure Scheduling process. Aeronautics and Space Specifically, PDRC's surface automation component predicts the earliest Administration (NASA)'s achievable takeoff times for departure flights and the departure runway. Precision Departure The Federal Aviation Administration (FAA) will use PDRC to help integrate Release Capability tower and en route decision support tools being developed for the next (PDRC) generation air transportation system (NextGen).

The Spot and Runway SARDA is NASA's contribution to improving the efficiency of airport Departure Advisor surface operations. It is the centerpiece of a partnership among airlines, (SARDA) airports, and air traffic controllers to improve operations at the nation’s busiest airports, providing an optimal sequence and times for runway usage (takeoff times for departures and runway crossing times for arrivals) and the times to release aircraft from gates or spots.

ERATO (En-Route Air It is a modern Mid- Term Conflict Detection tool for air traffic controllers Traffic Organizer) designed by the French ANSP (DSNA) providing assistance for the detection and resolution of conflicts and facilitating the cooperation between the executive and the planning controller on a control suite.

En Route Automation ERAM provides core functionality for air traffic controllers, and the FAA Modernization (ERAM), designed it to support satellite-based systems such as ADS-B and data communication technologies. ERAM increases flexible routing around congestion, weather, and other restrictions. Real-time air traffic management and information-sharing on flight restrictions improves airlines' ability to plan flights with minimal changes.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

iFACTS (interim Future iFACTS uses several sources of information to construct an accurate Area Control Tools) prediction (up to 18 minutes ahead) of the flight trajectories of aircraft in its area of interest. If it calculates that your aircraft will come into conflict with another aircraft or that separation might be compromised, it alerts the controller. It also records the clearances issued by controllers.

Time-based Separations TBS enables to minimise the impact of strong headwinds on landing rates, (TBS) thereby reducing delays and cancellations. The application of "Time- based separation" minima will not lead to any increase in runway capacity, but is expected to help prevent the loss of runway arrival capacity that typically occurs under strong headwind conditions. Note that as a first step, the Time-Based Separations project will only address operations dedicated to single runway arrivals, which are typical of most major European airports.

5.3.2 Non-nominal scenarios examples

AU developed primitive non-nominal scenarios below derived from AU_TACTA focus group results and simulator capabilities. The scenarios are listed below including aerodrome and terminal air space which can be created in synthetic or real/existing environment simulations.

Scenario 1-Tower: Safety net failure

Scenario description:

This scenario is related to the runway incursion, which is created the safety barrier failure in low visibility weather conditions. Safety barriers fail when an aircraft holding for line up and the other is in the final approach. The pilot of aircraft waiting for the holding point sees the safety barrier lights green and misunderstands the controller clearance as line up approved. Traffic moves and enters the runway when the other is very close for landing. Controllers have to take immediate actions to solve the confliction.

Impact:

 ATCOs: o Individual: Controller gets stressed about the situation and has to take immediate actions to solve the problem. He/she faces workload for related aircraft and also for the other aircraft in their responsibility. o Team: Situation should be solved with local, ground and approach controllers. Other controllers will be affected for runway incursion and going around procedures as coordination issues with neighbour sectors. They have to help to solve problem with team member faced. Their attention will be focus on the event and extra workload appears for them.  System:

o All aircraft will be affected with new sequence and planning created by the situation.

EDITION 00.02.00

o Terminal airspace complexity increases, o Capacity will be affected negatively, o Risk factor increases related to safety of other stakeholders… Countermeasures:

 Automated system validation analysis,  Advanced risk management analysis,  Procedural and operational training,  Re-analysing of the designed automated tools in the cockpit and ATCO’s working position

Scenario 2- Sequencing failure at CTA fix-in Terminal Area, related i4D issue.

Scenario description:

Traffics are sequenced in the terminal airspace through the Controlled Time of Arrival (CTA)/merged/meter fix by automated support. A pilot of an approaching airline enters wrong inputs into the FMS and airplane behaves different from the expected by the controllers. Then it creates complexity in a peak time period resulting extra workload and stress for the controllers.

Impact:

 ATCOs: o Individual: Controller meets the complex situation from he/she expected. His attention focuses on the mis-routing airplane. He has to reorganize the traffic sequence depending on the new aspects of traffic situation. o Team: Other controllers will be affected as coordination issues with neighbour sectors. They have to help to solve problem with team member faced. Their attention will be focus on the event and extra workload appears for them.  System: o All aircraft will be affected with new sequence and planning created by the situation. o Terminal airspace complexity increases, o Capacity will be effected negatively, o Risk factor increases related to safety of other stakeholders…

Countermeasures:

Scenario 3-Label transfer failure

Scenario description: The label on the traffics which are just landed and departing can be exchanged because of system failure. The situation creates communication and coordination failure and complexity between pilots and controllers when departure controller identifies the departure traffic handed over automatically from tower.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Impact:

 ATCOs: o Individual: Controller meets the complex situation from he/she expected. His attention focuses on the re-identification of the traffics. He has to reorganize the traffic sequence depending on the new aspects of traffic situation. o Team: Other controllers will be affected as coordination issues with neighbour sectors. They have to help to solve problem with team member faced. Their attention will be focus on the event and extra workload appears for them.  System: o All aircraft will be affected with new sequence and planning created by the situation. o Terminal airspace complexity increases, o Capacity will be effected negatively, o Risk factor increases related to safety of other stakeholders…

Countermeasures:

Scenario 4-Wrong info on the label

Scenario description: The direction of the flight is shown wrongly because of FDP failure. Controllers assume traffic should go through north but in the reality it has to be directed to the south. Situation creates complexity and discussion between controllers and pilots.

Impact:

 ATCOs: o Individual: Controller meets the complex situation from he/she expected. His attention focuses on the re-direction of the traffics. He has to reorganize the traffic sequence depending on the new aspects of traffic situation. o Team: Other controllers will be affected as coordination issues with neighbour sectors. They have to help to solve problem with team member faced. Their attention will be focus on the event and extra workload appears for them.  System: o All aircraft will be affected with new sequence and planning created by the situation. o Terminal airspace complexity increases, o Capacity will be effected negatively, o Risk factor increases related to safety of other stakeholders…

Countermeasures:

 Automated system validation analysis,

EDITION 00.02.00

 Advanced risk management analysis,  Procedural and operational training,  Re-analysing of the designed automated tools in the cockpit and ATCO’s working positions.

Scenario 5-STCA detection failure

Scenario description: The automatic STCA does not appears for the conflicted traffics. Controllers just realises that two traffics are very close each other informed by pilots having TCAS TA alert. The situation requires sudden makeovers affecting other traffics. Controller gets stress and high workload.

Impact:

 ATCOs: o Individual: Controller meets the complex situation from he/she expected. His attention focuses on the resolution of the traffics. He has to reorganize the traffic sequence depending on the new aspects of traffic situation. o Team: Other controllers will be affected as coordination issues with neighbour sectors. They have to help to solve problem with team member faced. Their attention will be focus on the event and extra workload appears for them.  System: o All aircraft will be affected with new sequence and planning created by the situation. o Terminal airspace complexity increases, o Capacity will be effected negatively, o Risk factor increases related to safety of other stakeholders…

Countermeasures:

 Automated system validation analysis,  Advanced risk management analysis,  Procedural and operational training, Re-analysing of the designed automated tools in the cockpit and ATCO’s working positions.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

6 References

[1] C. Martinie, P. Palanque, M. Ragosta, M. A. Sujan, D. Navarre, and A. Pasquini, “Understanding functional resonance through a federation of models: Preliminary findings of an avionics case study,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2013, vol. 8153 LNCS, pp. 216–227.

[2] E. Rigaud et al., “A Framework for Modelling the Consequences of the Propagation of Automation Degradation: Application to Air Traffic Control Systems,” in SESAR Innovation Days (SID 2012), 2012.

[3] O. Consortium, “OPTICS project,” 2015. .

[4] SESAR JU, “SESAR SWIM Global Demonstration.” 2016.

[5] International Civil Aviation Organization (ICAO), “Annual Report of the Council,” 2015.

[6] European Aviation Safety Agency (EASA), “Annual Safety Review,” 2016.

[7] SESAR JU, “European ATM Master Plan - Edition 2015 - Executive View,” 2015.

[8] European Organisation for the Safety of Air Navigation (EUROCONTROL) and SESAR JU, “European ATM Master Plan.” .

[9] SESAR JU, “SESAR Solutions,” 2014. .

[10] European Organisation for Safety of Air Navigation (EUROCONTROL), “European Free Route Airspace Developments,” 2012.

[11] IATA, EUROCONTROL, and CANSO, “Flight Efficiency Plan - Fuel and emissions savings,” 2008.

[12] European Organisation for the Safety of Air Navigation (EUROCONTROL), “Flight plan processing enters Free Route Airspace era.” .

[13] T. B. Sheridan and W. L. Verplank, “Human and Computer Control of Undersea Teleoperators,” Jul. 1978.

[14] HALA! SESAR Research Network, “Towards Higher Level of Automation in ATM - A

EDITION 00.02.00

Position Paper,” 2015.

[15] J. A. Wise, V. D. Hopkin, and D. J. Garland, Handbook of aviation human factors. CRC Press. CMC Press, 2009.

[16] L. Bainbridge, “Ironies of automation,” Automatica, vol. 19, no. 6, pp. 775–779, 1983.

[17] N. Jordan, “Allocation of functions between man and machines in automated systems,” J. Appl. Psychol., vol. 47, pp. 161–165, 1963.

[18] W. Sweet, “The glass cockpit [flight deck automation],” Spectrum, IEEE, vol. 32, no. 9, pp. 30–38, 1995.

[19] M. Older, P. Waterson, and C. Clegg, “A critical assessment of task allocation methods and their applicability,” Ergonomics, vol. 40, no. 2, pp. 151–171, 1997.

[20] P. M. Fitts, “The information capacity of the human motor system in controlling the amplitude of movement,” J. Exp. Psychol., pp. 381–391, 1954.

[21] T. B. Sheridan and W. Verplank, Human and Computer Control of Undersea Teleoperators. Cambridge, MA, USA: Man-Machine Systems Laboratory, Department of Mechanical Engineering, MIT., 1978.

[22] R. Parasuraman, T. Sheridan, and C. Wickens, “A model for types and levels of human interaction with automation,” IEEE Trans. Syst. Mans Cybern. Part A Syst. Humans, vol. 30, no. 3, pp. 286–297, 2000.

[23] A. Bye, E. Hollnagel, and T. S. Brendeford, “Human–machine function allocation: a functional modelling approach,” Reliab. Eng. Syst. Saf., vol. 64, pp. 291–300, 1999.

[24] J. M. Bindewald, M. E. Miller, and G. L. Peterson, “A function-to-task process model for adaptive automation system design,” Int. J. Hum. Comput. Stud., vol. 72, no. 12, pp. 822–834, 2014.

[25] R. Parasuraman and V. Riley, “Humans and automation: Use, misuse, disuse, abuse.,” Hum. Factors J. Hum. Factors Ergon. Soc., vol. 39, no. 2, pp. 230–253, 1997.

[26] D. B. Kaber and M. R. Endsley, “The effects of level of automation and adaptive automation on human performance, situation awareness and workload in a dynamic control task,” Theor. Issues Ergon. Sci., vol. 5, no. 2, pp. 113–153, 2004.

[27] B. Save, L., and Feuerberg, “Designing human-automation interaction: a new level of automation taxonomy,” in Proc. Human Factors of Systems and Technology, 2012.

[28] C. Martinie, P. Palanque, and M. Ragosta, “Some Issues with Interaction Design and Implementation in the Context of Autonomous Interactive Critical Systems,” in In CHI 2012 workshop: End-user interactions with intelligent and autonomous systems, 2012.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

[29] M. Ragosta, “Effective Automation in ATM : a new perspective on management and control of automated system,” in International Conference on Application and Theory of Automation in Command and Control Systems, 2012, no. Figure 2, pp. 215–218.

[30] M. Ragosta, “A multi-modelS based approach for the modelling and the analysis of usable and resilient partly autonomous interactive systems,” PhD Thesis - Toulouse 3 – Paul Sabatier University, 2015.

[31] B. A. Nardi, “Activity theory and human-computer interaction.,” in Context and consciousness: activity theory and human-computer interaction, B. A. Nardi, Ed. Cambridge and London: MIT Press, 1996, pp. 69–103.

[32] L. Save and B. Feuerberg, “Designing Human-Automation Interaction: a new level of Automation Taxonomy,” Hum. Factors Syst. Technol. (on Occas. Hum. Factors Ergon. Soc. Eur. Chapter Annu. Meet., 2012.

[33] L. Save, “Not all or nothing, not all the same: classifying automation in practice,” Hindsight, vol. 20, no. winter, 2014.

[34] E. Salas and D. Maurino, Human Factors in Aviation. Printbook, 2010.

[35] A. Angeli, M. Minetto, A. Dovio, and P. Paccotti, “The overtraining syndrome in athletes: a stress-related disorder.,” J. Endocrinol. Invest., vol. 27, no. 6, pp. 603–12, Jun. 2004.

[36] M. Frankenhaeuser and B. Gardell, “Underload and overload in working life: outline of a multidisciplinary approach,” J. Human Stress, vol. 2, no. 3, pp. 35–46, Sep. 1976.

[37] S. Alehagen, K. Wijma, U. Lundberg, B. Melin, and B. Wijma, “Catecholamine and cortisol reaction to childbirth,” Int. J. Behav. Med., vol. 8, no. 1, pp. 50–65, Mar. 2001.

[38] J. B. Sexton, E. J. Thomas, and R. L. Helmreich, “Error, stress, and teamwork in medicine and aviation: cross sectional surveys,” BMJ Br. Med. J., vol. 320, no. 7237, pp. 745–749, Mar. 2000.

[39] “Home : Occupational Outlook Handbook: : U.S. Bureau of Labor Statistics.” [Online]. Available: http://www.bls.gov/ooh/. [Accessed: 24-Jun-2015].

[40] H. B. Ammennan, “FAA air traffic control operations concepts,” vol. 6, no. AKTCC/HOST En route controllers, 1987.

[41] M. W. Cannon, “Perceived contrast in the fovea and periphery,” J. Opt. Soc. Am. A., vol. 2, no. 10, pp. 1760–1768, Oct. 1985.

[42] E. Chajut and D. Algom, “Selective attention improves under stress: implications for theories of social cognition.,” J. Pers. Soc. Psychol., vol. 85, no. 2, pp. 231–48, Aug.

EDITION 00.02.00

2003.

[43] E. J. Calabrese, “Neuroscience and hormesis: overview and general findings,” Crit. Rev. Toxicol., vol. 38, no. 4, pp. 249–252, 2008.

[44] D. A. Norman and T. Shallice, “Attention to Action,” R. J. Davidson, G. E. Schwartz, and D. Shapiro, Eds. Boston, MA: Springer US, 1986, pp. 1–18.

[45] M. Carrasco, “Visual attention: the past 25 years.,” Vision Res., vol. 51, no. 13, pp. 1484–525, Jul. 2011.

[46] L. Giraudet, J.-P. Imbert, M. Bérenger, S. Tremblay, and M. Causse, “The neuroergonomic evaluation of human machine interface design in air traffic control using behavioral and EEG/ERP measures,” Behav. Brain Res., vol. 294, pp. 246–253, Nov. 2015.

[47] A. Mack and I. Rock, Inattentional blindness. MIT Press, 2000.

[48] S. B. Most, B. J. Scholl, E. R. Clifford, and D. J. Simons, “What you see is what you set: sustained inattentional blindness and the capture of awareness,” Psychol. Rev., vol. 112, no. 1, pp. 217–242, Jan. 2005.

[49] D. J. Simons and C. F. Chabris, “Gorillas in our midst: sustained inattentional blindness for dynamic events.,” Perception, vol. 28, no. 9, pp. 1059–74, 1999.

[50] D. Fougnie and R. Marois, “Executive working memory load induces inattentional blindness.,” Psychon. Bull. Rev., vol. 14, no. 1, pp. 142–7, Feb. 2007.

[51] J. J. Todd, D. Fougnie, and R. Marois, “Visual short-term memory load suppresses temporo-parietal junction activity and induces inattentional blindness.,” Psychol. Sci., vol. 16, no. 12, pp. 965–72, Dec. 2005.

[52] K. S. Steelman, J. S. McCarley, and C. D. Wickens, “Great expectations: top-down attention modulates the costs of clutter and eccentricity,” J. Exp. Psychol. Appl., vol. 19, no. 4, pp. 403–419, Dec. 2013.

[53] M. I. Nikolic and N. B. Sarter, “Peripheral visual feedback: a powerful means of supporting effective attention allocation in event-driven, data-rich environments,” Hum. Factors, vol. 43, no. 1, pp. 30–38, 2001.

[54] S. B. Most, B. J. Scholl, E. R. Clifford, and D. J. Simons, “What you see is what you set: sustained inattentional blindness and the capture of awareness.,” Psychol. Rev., vol. 112, no. 1, pp. 217–42, Jan. 2005.

[55] E. A. Newby and I. Rock, “Inattentional blindness as a function of proximity to the focus of attention.,” Perception, vol. 27, no. 9, pp. 1025–40, 1998.

[56] H. S. Chan and A. J. Courtney, “Effects of cognitive foveal load on a peripheral single- 86 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

target detection task,” Percept. Mot. Skills, vol. 77, no. 2, pp. 515–533, Oct. 1993.

[57] F. Dehais, C. Tessier, L. Christophe, and F. Reuzeau, “The Perseveration Syndrome in the Pilot’s Activity: Guidelines and Cognitive Countermeasures,” Springer Berlin Heidelberg, 2010, pp. 68–80.

[58] M. I. Posner and S. E. Petersen, “The Attention System of the Human Brain,” Annu. Rev. Neurosci., vol. 13, no. 1, pp. 25–42, 1990.

[59] R. Parasuraman, P. M. Greenwood, and T. Sunderland, “The apolipoprotein E gene, attention, and brain function.,” Neuropsychology, vol. 16, no. 2, pp. 254–74, Apr. 2002.

[60] M. Corbetta, J. M. Kincade, J. M. Ollinger, M. P. McAvoy, and G. L. Shulman, “Voluntary orienting is dissociated from target detection in human posterior parietal cortex,” Nat. Neurosci., vol. 3, no. 3, pp. 292–297, Mar. 2000.

[61] M. M. Botvinick, T. S. Braver, D. M. Barch, C. S. Carter, and J. D. Cohen, “Conflict monitoring and cognitive control,” Psychol. Rev., vol. 108, no. 3, pp. 624–652, Jul. 2001.

[62] J. Theeuwes and R. Burger, “Attentional control during visual search: the effect of irrelevant singletons,” J. Exp. Psychol. Hum. Percept. Perform., vol. 24, no. 5, pp. 1342–1353, Oct. 1998.

[63] S. Yantis and J. Jonides, “Abrupt visual onsets and selective attention: evidence from visual search,” J. Exp. Psychol. Hum. Percept. Perform., vol. 10, no. 5, pp. 601–621, Oct. 1984.

[64] S. Suzuki and P. Cavanagh, “Focused attention distorts visual space: an attentional repulsion effect,” J. Exp. Psychol. Hum. Percept. Perform., vol. 23, no. 2, pp. 443–463, Apr. 1997.

[65] M. Carrasco, D. L. Evert, I. Chang, and S. M. Katz, “The eccentricity effect: target eccentricity affects performance on conjunction searches,” Percept. Psychophys., vol. 57, no. 8, pp. 1241–1261, Nov. 1995.

[66] J. P. Rijsdijk, J. N. Kroon, and G. J. van der Wildt, “Contrast sensitivity as a function of position on the retina,” Vision Res., vol. 20, no. 3, pp. 235–241, 1980.

[67] J. Rasmussen, “The definition of human error and a taxonomy for technical system design,” pp. 23–30, 1987.

[68] J. Rasmussen and K. Vicente, “Coping with Human Errors Through System Design: Implications for Ecological Interface Design,” Int. J. Man-Mach. Stud., vol. 31, no. 5,

EDITION 00.02.00

pp. 517–534, Nov. 1989.

[69] K. J. Vicente and J. Rasmussen, “Ecological interface design: theoretical foundations,” IEEE Trans. Syst. Man. Cybern., vol. 22, no. 4, pp. 589–606, Jul. 1992.

[70] K. A. Gluck and R. W. Pew, Modeling Human Behavior With Integrated Cognitive Architectures: Comparison, Evaluation, and Validation. Psychology Press, 2006.

[71] D. Kahneman, Thinking, Fast and Slow. Macmillan, 2011.

[72] J. Rasmussen, “Skills, rules, and knowledge; signals, signs, and symbols, and other distinctions in human performance models,” IEEE Trans. Syst. Man. Cybern., vol. SMC- 13, no. 3, pp. 257–266, 1983.

[73] J. Rasmussen, Information Processing and Human-Machine Interaction: An Approach to Cognitive Engineering. New York, NY, USA: Elsevier Science Inc., 1986.

[74] T. Edwards, “Human performance in air traffic control,” Nottingham, 2013.

[75] Advisory Council for Aviation Research and innovation in Europe, “ACARE,” 2012. .

[76] STRESS Consortium, “D4.1 - Validation Plan,” 2017.

[77] AUTOPACE project, “AUTOPACE website.” [Online]. Available: http://autopace.eu/.

[78] MINIMA project, “MINIMA website.” [Online]. Available: http://www.minima- project.eu/wp/.

[79] F. K. Punch, Introduction to Social Research–Quantitative & Qualitative Approaches. London, UK: Sage Publications, 2005.

[80] R. A. Krueger and M. A. Casey, Focus Groups:A Pratical Guide For Applied Research. Thousand Oaks, CA: Sage Publications, 2000.

[81] Y. A. and H. Şimşek, Sosyal Bilimlerde Nitel Araştırma Yöntemleri. Ankara: Seçkin Yayıncılık, 2013.

[82] T. Baş and U. Akturan, Nitel Araştırma Yöntemleri. Ankara: Seçkin Yayıncılık, 2013.

[83] D. L. Morgan and R. A. Kreuger, “When to use focus groups and why,” in Successful Focus Groups, D. L. Morgan, Ed. London, UK: Sage Publications, 1993.

[84] A. Tversky and D. Kahneman, “Judgment under Uncertainty: Heuristics and Biases,” Science (80-. )., vol. 185, 1974.

[85] Chua, Cousy, Causse, and Lancelot, “Initial Impact of Modern Taxiing Techniques on Airport Ground Conbtrol,” in HCI Aero, 2016.

[86] H. Sequeira, P. Hot, L. Silvert, and S. Delplanque, “Electrical autonomic correlates of 88 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

emotion.,” Int. J. Psychophysiol., vol. 71, no. 1, pp. 50–6, Jan. 2009.

[87] L. J. Mulder, “Measurement and analysis methods of heart rate and respiration for use in applied environments.,” Biol. Psychol., vol. 34, no. 2–3, pp. 205–36, Nov. 1992.

[88] R. P. Sloan et al., “Effect of mental stress throughout the day on cardiac autonomic control.,” Biol. Psychol., vol. 37, no. 2, pp. 89–99, Mar. 1994.

[89] J. Vander Sloten, P. Verdonck, M. Nyssen, and J. Haueisen, Eds., 4th European Conference of the International Federation for Medical and Biological Engineering, vol. 22. Berlin, Heidelberg: Springer Berlin Heidelberg, 2009.

[90] L. Salahuddin, J. Cho, M. G. Jeong, and D. Kim, “Ultra short term analysis of heart rate variability for monitoring mental stress in mobile settings.,” Conf. Proc. ... Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf., vol. 2007, pp. 4656–9, 2007.

[91] W. Boucsein, “Principles of Electrodermal Phenomena,” in Electrodermal Activity, Boston, MA: Springer US, 2012, pp. 1–86.

[92] L. Bernardi et al., “Effects of controlled breathing, mental activity and mental stress with or without verbalization on heart rate variability.,” J. Am. Coll. Cardiol., vol. 35, no. 6, pp. 1462–9, May 2000.

[93] T. W. Picton et al., “Guidelines for using human event-related potentials to study cognition: recording standards and publication criteria.,” Psychophysiology, vol. 37, no. 2, pp. 127–52, Mar. 2000.

[94] M. Fabiani, G. Gratton, D. Karis, and E. Donchin, “Definition, identification, and reliability of measurement of the P300 component of the event-related brain potential,” Adv. Psychophysiol., 1987.

[95] J. Polich, “Updating P300: an integrative theory of P3a and P3b.,” Clin. Neurophysiol., vol. 118, no. 10, pp. 2128–48, Oct. 2007.

[96] R. Johnson, “A triarchic model of P300 amplitude.,” Psychophysiology, vol. 23, no. 4, pp. 367–84, Jul. 1986.

[97] S. L. Crites, J. T. Cacioppo, W. L. Gardner, and G. G. Berntson, “Bioelectrical echoes from evaluative categorization: II. A late positive brain potential that varies as a function of attitude registration rather than attitude report.,” J. Pers. Soc. Psychol., vol. 68, no. 6, pp. 997–1013, Jun. 1995.

[98] H. M. Gray, H. M. Gray, N. Ambady, W. T. Lowenthal, and P. Deldin, “P300 as an index of attention to self-relevant stimuli,” J. Exp. Soc. Psychol., pp. 216--224, 2004.

EDITION 00.02.00

[99] C. C. Duncan-Johnson and E. Donchin, “On Quantifying Surprise: The Variation of Event-Related Potentials With Subjective Probability,” Psychophysiology, vol. 14, no. 5, pp. 456–467, Sep. 1977.

[100] A. Magliero, T. R. Bashore, M. G. Coles, and E. Donchin, “On the dependence of P300 latency on stimulus evaluation processes.,” Psychophysiology, vol. 21, no. 2, pp. 171– 86, Mar. 1984.

[101] J. Polich, “Attention, probability, and task demands as determinants of P300 latency from auditory stimuli.,” Electroencephalogr. Clin. Neurophysiol., vol. 63, no. 3, pp. 251–9, Mar. 1986.

[102] M. Kutas, G. McCarthy, and E. Donchin, “Augmenting mental chronometry: the P300 as a measure of stimulus evaluation time.,” Science, vol. 197, no. 4305, pp. 792–5, Aug. 1977.

[103] K. Hagemann, “The alpha band as an electrophysiological indicator for internalized attention and high mental workload in real traffic driving.”

[104] P. Aricò, F. Aloise, F. Schettini, S. Salinari, D. Mattia, and F. Cincotti, “Influence of P300 latency jitter on event related potential-based brain–computer interface performance,” J. Neural Eng., vol. 11, no. 3, p. 35008, Jun. 2014.

[105] O. Jensen, J. Kaiser, and J.-P. Lachaux, “Human gamma-frequency oscillations associated with attention and memory.,” Trends Neurosci., vol. 30, no. 7, pp. 317–24, Jul. 2007.

[106] D. LaBerge, “Attention, consciousness, and electrical wave activity within the cortical column.,” Int. J. Psychophysiol., vol. 43, no. 1, pp. 5–24, Dec. 2001.

[107] S. E. Petersen and M. I. Posner, “The attention system of the human brain: 20 years after.,” Annu. Rev. Neurosci., vol. 35, pp. 73–89, 2012.

[108] C. Neuper and G. Pfurtscheller, “Event-related dynamics of cortical rhythms: frequency-specific features and functional correlates.,” Int. J. Psychophysiol., vol. 43, no. 1, pp. 41–58, Dec. 2001.

[109] C. Neuper and W. Klimesch, Event-related dynamics of brain oscillations. Elsevier, 2006.

[110] G. Borghini, L. Astolfi, G. Vecchiato, D. Mattia, and F. Babiloni, “Measuring neurophysiological signals in aircraft pilots and car drivers for the assessment of mental workload, fatigue and drowsiness,” Neuroscience and Biobehavioral Reviews, vol. 44. Elsevier Ltd, pp. 58–75, 2014.

[111] N. Ramnani and A. M. Owen, “Anterior prefrontal cortex: insights into function from anatomy and neuroimaging,” Nat. Rev. Neurosci., vol. 5, no. 3, pp. 184–194, Mar. 90 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

2004.

[112] J. N. Wood and J. Grafman, “Human prefrontal cortex: processing and representational perspectives,” Nat. Rev. Neurosci., vol. 4, no. 2, pp. 139–147, Feb. 2003.

[113] R. Cabeza and L. Nyberg, “Imaging cognition II: An empirical review of 275 PET and fMRI studies,” J. Cogn. Neurosci., vol. 12, no. 1, pp. 1–47, Jan. 2000.

[114] G. Borghini et al., “A neurophysiological training evaluation metric for air traffic management,” Conf. Proc. ... Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf., vol. 2014, pp. 3005–3008, 2014.

[115] G. Borghini et al., “Quantitative Assessment of the Training Improvement in a Motor- Cognitive Task by Using EEG, ECG and EOG Signals,” Brain Topogr., Jan. 2015.

[116] K. Ryu and R. Myung, “Evaluation of mental workload with a combined measure based on physiological indices during a dual task of tracking and mental arithmetic,” Int. J. Ind. Ergon., vol. 35, no. 11, pp. 991–1009, 2005.

[117] Gevins, A and M. E. Smith, “Neurophysiological measures of cognitive workload during human-computer interaction,” Theor. Issues Ergon. Sci., vol. 4, no. 1–2, 2003.

[118] A. S. Smit, P. a. T. M. Eling, M. T. E. Hopman, and A. M. L. Coenen, “Mental and physical effort affect vigilance differently,” 218, 2005.

[119] J. B. Brookings, G. F. Wilson, and C. R. Swain, “Psychophysiological responses to changes in workload during simulated air traffic control,” Biol. Psychol., vol. 42, no. 3, pp. 361–377, Feb. 1996.

[120] A. Gevins, M. E. Smith, L. McEvoy, and D. Yu, “High-resolution EEG mapping of cortical activation related to working memory: effects of task difficulty, type of processing, and practice,” Cereb. Cortex (New York, N.Y. 1991), vol. 7, no. 4, pp. 374–385, Jun. 1997.

[121] N. Jaušovec and K. Jaušovec, “Working memory training: Improving intelligence – Changing brain activity,” Brain Cogn., vol. 79, no. 2, pp. 96–106, 2012.

[122] W. Klimesch, M. Doppelmayr, T. Pachinger, and B. Ripper, “Brain oscillations and human memory: EEG correlates in the upper alpha and theta band,” Neurosci. Lett., vol. 238, no. 1–2, pp. 9–12, Nov. 1997.

[123] L. Venables and S. H. Fairclough, “The influence of performance feedback on goal- setting and mental effort regulation,” Motiv. Emot., vol. 33, no. 1, pp. 63–74, Jan. 2009.

EDITION 00.02.00

[124] M. E. Smith and A. Gevins, “Neurophysiologic monitoring of mental workload and fatigue during operation of a flight simulator,” 2005, vol. 5797, pp. 116–126.

[125] J. Onton, A. Delorme, and S. Makeig, “Frontal midline EEG dynamics during working memory,” Neuroimage, vol. 27, no. 2, pp. 341–356, Aug. 2005.

[126] W. Klimesch, “EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis,” Brain Res. Brain Res. Rev., vol. 29, no. 2–3, pp. 169–195, Apr. 1999.

[127] G. Vecchiato et al., “Neuroelectrical correlates of trustworthiness and dominance judgments related to the observation of political candidates,” Comput. Math. Methods Med., vol. 2014, p. 434296, 2014.

[128] G. Derosière, K. Mandrick, G. Dray, T. E. Ward, and S. Perrey, “NIRS-measured prefrontal cortex activity in neuroergonomics: strengths and weaknesses,” Front. Hum. Neurosci., vol. 7, Sep. 2013.

[129] X. Cui, S. Bray, D. M. Bryant, G. H. Glover, and A. L. Reiss, “A quantitative comparison of NIRS and fMRI across multiple cognitive tasks,” Neuroimage, vol. 54, no. 4, pp. 2808–2821, Feb. 2011.

[130] T. J. Huppert, R. D. Hoge, S. G. Diamond, M. A. Franceschini, and D. A. Boas, “A temporal comparison of BOLD, ASL, and NIRS hemodynamic responses to motor stimuli in adult humans,” Neuroimage, vol. 29, no. 2, pp. 368–382, Jan. 2006.

[131] C. Herff, D. Heger, O. Fortmann, J. Hennrich, F. Putze, and T. Schultz, “Mental workload during n-back task-quantified in the prefrontal cortex using fNIRS,” Front. Hum. Neurosci., vol. 7, p. 935, 2013.

[132] M. M. Plichta et al., “Event-related functional near-infrared spectroscopy (fNIRS): are the measurements reliable?,” Neuroimage, vol. 31, no. 1, pp. 116–124, May 2006.

[133] K. Izzetoglu, S. Bunce, B. Onaral, K. Pourrezaei, and B. Chance, “Functional Optical Brain Imaging Using Near-Infrared During Cognitive Tasks,” Int. J. Hum. Comput. Interact., vol. 17, no. 2, pp. 211–227, Jun. 2004.

[134] A. V. Medvedev, J. Kainerstorfer, S. V. Borisov, R. L. Barbour, and J. VanMeter, “Event- related fast optical signal in a rapid object recognition task: improving detection by the independent component analysis,” Brain Res., vol. 1236, pp. 145–158, Oct. 2008.

[135] D. Hutchison et al., “Toward a Wearable, Neurally-Enhanced Augmented Reality System,” vol. 6780, D. D. Schmorrow and C. M. Fidopiastis, Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2011, pp. 493–499.

[136] G. Durantin, J.-F. Gagnon, S. Tremblay, and F. Dehais, “Using near infrared spectroscopy and heart rate variability to detect mental overload,” Behav. Brain Res., 92 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

vol. 259, pp. 16–23, Feb. 2014.

[137] A. M. Owen, K. M. McMillan, A. R. Laird, and E. Bullmore, “N-back working memory paradigm: a meta-analysis of normative functional neuroimaging studies,” Hum. Brain Mapp., vol. 25, no. 1, pp. 46–59, May 2005.

[138] J. Harrison et al., “Cognitive Workload and Learning Assessment During the Implementation of a Next-Generation Air Traffic Control Technology Using Functional Near-Infrared Spectroscopy,” IEEE Trans. Human-Machine Syst., vol. 44, no. 4, pp. 429–440, Aug. 2014.

[139] B. Kirwan, R. Scaife, and R. Kennedy, “Investigating complexity factors in UK air traffic management,” Hum. Factors Aerosp. Saf., vol. 1, no. 2, 2001.

[140] J. M. Murkin and M. Arango, “Near-infrared spectroscopy as an index of brain and tissue oxygenation,” Br. J. Anaesth., vol. 103 Suppl, pp. i3-13, Dec. 2009.

[141] G. Shou, L. Ding, and D. Dasari, “Probing neural activations from continuous EEG in a real-world task: time-frequency independent component analysis,” J. Neurosci. Methods, vol. 209, no. 1, pp. 22–34, Jul. 2012.

[142] P. Aricò, F. Aloise, F. Schettini, S. Salinari, D. Mattia, and F. Cincotti, “Influence of P300 latency jitter on event related potential-based brain-computer interface performance,” J. Neural Eng., vol. 11, no. 3, p. 35008, Jun. 2014.

[143] P. Aricò et al., “Reliability over time of EEG-based mental workload evaluation during Air Traffic Management (ATM) tasks,” Conf. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf., vol. 2015, pp. 7242–7245, 2015.

[144] G. Di Flumeri et al., “On the Use of Cognitive Neurometric Indexes in Aeronautic and Air Traffic Management Environments,” vol. 9359, B. Blankertz, G. Jacucci, L. Gamberini, A. Spagnolli, and J. Freeman, Eds. Cham: Springer International Publishing, 2015, pp. 45–56.

[145] P. Aricò et al., “ATCO: Neurophysiological Analysis Of The Training And Of The Workload,” Ital. J. Aerosp. Med., vol. 1, no. 12, Jan. 2015.

[146] K.-R. Müller, M. Tangermann, G. Dornhege, M. Krauledat, G. Curio, and B. Blankertz, “Machine learning for real-time single-trial EEG-analysis: from brain-computer interfacing to mental state monitoring,” J. Neurosci. Methods, vol. 167, no. 1, pp. 82– 90, Jan. 2008.

[147] C. Berka et al., “Real-Time Analysis of EEG Indexes of Alertness, Cognition, and Memory Acquired With a Wireless EEG Headset,” Int. J. Hum. Comput. Interact., vol. 17, no. 2, pp. 151–170, Jun. 2004.

EDITION 00.02.00

[148] C. Berka et al., “EEG correlates of task engagement and mental workload in vigilance, learning, and memory tasks,” Aviat. Space. Environ. Med., vol. 78, no. 5 Suppl, pp. B231-244, May 2007.

[149] G. B. Pietro Arico, “Towards a multimodal bioelectrical framework for the online mental workload evaluation,” 2014.

[150] L.-D. Liao et al., “Gaming control using a wearable and wireless EEG-based brain- computer interface device with novel dry foam-based sensors,” J. Neuroeng. Rehabil., vol. 9, p. 5, 2012.

[151] “Brain Mapping,” MIT Technology Review. .

[152] A. Gevins and M. E. Smith, “Neurophysiological measures of cognitive workload during human-computer interaction,” Theor. Issues Ergon. Sci., vol. 4, no. 1–2, pp. 113–131, Jan. 2003.

[153] O. Jensen, J. Kaiser, and J.-P. Lachaux, “Human gamma-frequency oscillations associated with attention and memory,” Trends Neurosci., vol. 30, no. 7, pp. 317–324, Jul. 2007.

[154] M. B. Sterman, C. A. Mann, D. A. Kaiser, and B. Y. Suyenobu, “Multiband topographic EEG analysis of a simulated visuomotor aviation task,” Int. J. Psychophysiol. Off. J. Int. Organ. Psychophysiol., vol. 16, no. 1, pp. 49–56, Feb. 1994.

[155] M. A. Beulen, “The role of theta oscillations in memory and decision making,” Jul- 2011. .

[156] S. Sandkühler and J. Bhattacharya, “Deconstructing insight: EEG correlates of insightful problem solving,” PLoS One, vol. 3, no. 1, p. e1459, 2008.

[157] G. Borghini et al., “Frontal EEG theta changes assess the training improvements of novices in flight simulation tasks,” Conf. Proc. ... Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf., vol. 2013, pp. 6619–6622, 2013.

[158] G. Borghini, L. Astolfi, G. Vecchiato, D. Mattia, and F. Babiloni, “Measuring neurophysiological signals in aircraft pilots and car drivers for the assessment of mental workload, fatigue and drowsiness,” Neurosci. Biobehav. Rev., vol. 44, pp. 58– 75, Jul. 2014.

[159] G. Borghini et al., “Assessment of mental fatigue during car driving by using high resolution EEG activity and neurophysiologic indices,” Conf. Proc. ... Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Conf., vol. 2012, pp. 6442–6445, 2012.

[160] A. Gevins et al., “Monitoring working memory load during computer-based tasks with EEG pattern recognition methods,” Hum. Factors, vol. 40, no. 1, pp. 79–91, Mar. 1998. 94 © – 2016 – Deep Blue, University La Sapienza, ENAC, Anadolu University, Eurocontrol. All rights reserved. Licensed to the SESAR Joint Undertaking under conditions

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

[161] J. Toppi et al., “Investigating Cooperative Behavior in Ecological Settings: An EEG Hyperscanning Study,” PLoS One, vol. 11, no. 4, p. e0154236, Apr. 2016.

[162] G. Vecchiato et al., “Investigation of the effect of EEG-BCI on the simultaneous execution of flight simulation and attentional tasks,” Med. Biol. Eng. Comput., vol. 54, no. 10, pp. 1503–1513, Oct. 2016.

[163] G. Albouy et al., “Both the hippocampus and striatum are involved in consolidation of motor sequence memory,” Neuron, vol. 58, no. 2, pp. 261–272, Apr. 2008.

[164] G. Borghini et al., “Quantitative Assessment of the Training Improvement in a Motor- Cognitive Task by Using EEG, ECG and EOG Signals,” Brain Topogr., Jan. 2015.

[165] G. Borghini et al., “Neurophysiological Measures for Users’ Training Objective Assessment During Simulated Robot-Assisted Laparoscopic Surgery,” 38th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., Aug. 2016.

[166] S. Chauvette, J. Seigneur, and I. Timofeev, “Sleep oscillations in the thalamocortical system induce long-term neuronal plasticity,” Neuron, vol. 75, no. 6, pp. 1105–1113, Sep. 2012.

[167] J. D. Kropotov, Quantitative EEG, Event-Related Potentials and Neurotherapy. Academic Press, 2010.

[168] P. Sauseng, W. Klimesch, M. Schabus, and M. Doppelmayr, “Fronto-parietal EEG coherence in theta and upper alpha reflect central executive functions of working memory,” Int. J. Psychophysiol. Off. J. Int. Organ. Psychophysiol., vol. 57, no. 2, pp. 97– 103, Aug. 2005.

[169] J. Gruzelier, “A theory of alpha/theta neurofeedback, creative performance enhancement, long distance functional connectivity and psychological integration,” Cogn. Process., vol. 10 Suppl 1, pp. S101-109, Feb. 2009.

[170] J. Gruzelier, T. Egner, and D. Vernon, “Validating the efficacy of neurofeedback for optimising performance,” Prog. Brain Res., vol. 159, pp. 421–431, 2006.

[171] F. Crick, “Function of the thalamic reticular complex: the searchlight hypothesis,” Proc. Natl. Acad. Sci. U. S. A., vol. 81, no. 14, pp. 4586–4590, Jul. 1984.

[172] D. LaBerge, “Attention, consciousness, and electrical wave activity within the cortical column,” Int. J. Psychophysiol. Off. J. Int. Organ. Psychophysiol., vol. 43, no. 1, pp. 5– 24, Dec. 2001.

[173] D. LaBerge and M. S. Buchsbaum, “Positron emission tomographic measurements of pulvinar activity during an attention task,” J. Neurosci. Off. J. Soc. Neurosci., vol. 10,

EDITION 00.02.00

no. 2, pp. 613–619, Feb. 1990.

[174] S. E. Petersen and M. I. Posner, “The Attention System of the Human Brain: 20 Years After,” Annu. Rev. Neurosci., vol. 35, pp. 73–89, Jul. 2012.

[175] S. E. Petersen, D. L. Robinson, and J. D. Morris, “Contributions of the pulvinar to visual spatial attention,” Neuropsychologia, vol. 25, no. 1A, pp. 97–105, 1987.

[176] D. H. O’Connor, M. M. Fukui, M. A. Pinsk, and S. Kastner, “Attention modulates responses in the human lateral geniculate nucleus,” Nat. Neurosci., vol. 5, no. 11, pp. 1203–1209, Nov. 2002.

[177] W. Vanduffel, R. B. Tootell, and G. A. Orban, “Attention-dependent suppression of metabolic activity in the early stages of the macaque visual system,” Cereb. Cortex (New York, N.Y. 1991), vol. 10, no. 2, pp. 109–126, Feb. 2000.

[178] C. Neuper and G. Pfurtscheller, “Event-related dynamics of cortical rhythms: frequency-specific features and functional correlates,” Int. J. Psychophysiol. Off. J. Int. Organ. Psychophysiol., vol. 43, no. 1, pp. 41–58, Dec. 2001.

[179] C. Neuper and W. Klimesch, Event-Related Dynamics of Brain Oscillations. Elsevier, 2006.

[180] J. Toppi et al., “Time varying effective connectivity for describing brain network changes induced by a memory rehabilitation treatment,” in 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2014, pp. 6786–6789.

[181] W. Klimesch, “α-band oscillations, attention, and controlled access to stored information,” Trends Cogn. Sci., vol. 16, no. 12, pp. 606–617, Dec. 2012.

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

7 Appendix 1: AU _TATCA Focus Group

This study is performed for the Project of STRESS in which Anadolu University is the consortium member, under the European Commission H2020 SESAR calls. Your opinions and experiences will be supporting value added scenario developments of the Project which is related to ATM Automation and Controller’s performance. Research study is managed by Anadolu Team including Dr. Birsen Açıkel and Dr. Ali Ozan Canarslanlar with European partners of DeepBlue, University of Sapienza, ENAC and Eurocontrol

Thank you very much for your kind participation and contributions.

Dr. Uğur Turhan

STRESS Project Anadolu University Coordinator

Table 8: Semi-structured questions form Gender: Age:

Working Position: Work Experience:

What are your automated systems/tools in your working environment?

Do you trust to automated systems/tools in your working environment?

How automated tools effect your workload??

How automated tools effect flight safety?

EDITION 00.02.00

How automated tools effect ATM capacity?

How is the working performance of automated tools? Examples?

Do you experience stress when you face with some failures of automated tolls? Examples?

How automated tools effect your situational awareness? Examples?

Could you give information about the training and orientation of new automated tools usage.

How is your knowledge about future automated systems in ATM and aviation?

The following table reports participant demographics information.

Table 9: Focus group participant demographics

Gender Age Working Position Work Experience 1 Male 32 Approach 7 2 Female 31 Approach 7 3 Male 26 Approach 2 4 Male 29 Approach 6 5 Male 34 Tower 10 6 Male 43 Approach 20 7 Male 37 Approach 14

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

The answers of air traffic controllers, results, themes and codes from focus group meeting have been summarised in the table below.

Table 10: Air traffic controller’s answers, results, themes and codes from focus group meeting What are your automated systems/tools in your working environment?

Radar, FDP, VCS, AMAN, DMAN, STCA, Digital ATIS, AWOS, BirdRad, ASMGCS,

Do you trust to automated systems/tools in your working environment?

I do not trust totally. Sub Themes I trust 80%. I do not trust (2) Yes. Except A-MAN, there is general un-trust situation. I trust [5] Yes. I do not. Unacceptable experiences occur Yes. Codes Partly. AMAN

How automated tools effect your workload?

Workload decreases when automated tool works correctly Sub Themes Decreases Decreases (5) Decreases Increases (2)

Workload decreases when automated tool works correctly Decreases Codes Increases, but I need automated support Careful working Requires more careful working. Data failures

How automated tools effect flight safety? Increases Sub Themes Increases Increases (7) Increases Useful when automated tool works correctly. Failure factors should be considered. Increases Codes Increases safety Increases safety potentially. We work with increased safety minima. failure factors

How automated tools effect ATM capacity?

Helps to increase. Sub Themes Increases Increases (6) Helps to increase Decreases (1) Positive for planning and decision making, but we consider all time error rates and failure factors. Codes

EDITION 00.02.00

Capacity increases partly. Error rates Increases Failures Decreases

How is the working performance of automated tools? Examples?

System performance is over 95%. Failures are lost data, label jump, wrong Sub Themes trajectory STCA failures The identification of traffics can be challenging. Radar failures Identification problems sometimes. Aircraft label failures STCA failure, info update delay on Radar, lost traffic label or identification difficulty. Performs good generally, sometimes fails. Codes Radar and frequency weaknesses. Lost label Fake STCA alert. Serious weaknesses. Info updating Not corresponding to real traffic Lost data

Do you experience stress when you face with some failures of automated tolls? Examples?

Yes. I get stressed and panic if the STCA works correctly. Sub Themes Traffic identification can be challenging during the heavy traffic. I experience (5) No. I do not experience (2) Sometimes stressing. Lost radar data and lost frequency Causes stress, especially in loss of radar display. Codes I experience, I get angry. STCA I do not have any experience Heavy traffic Radar display How automated tools effect your situational awareness? Examples?

When get used to systems trust increases and awareness decreases The Sub Themes unidentified traffics can be challenging in heavy traffic. Decreases (4) I feel positive contribution Increases (3) Fake errors affect my attention negatively. System should be updated regularly. My workload reduces and I can concentrate my job Codes Increases my workload and I get bored. Workload Not affecting. I take my countermeasures. System reliability

Could you give information about the training and orientation of new automated tools usage

We are informed and improve while using Sub Themes Enough we learn on the job more. enough (2) We learn in the courses Not enough (5) We are informed then we improve our selves Not enough, we learn on the job. Codes Not enough Just information provided Verbal information Individual efforts How is your knowledge about future automated systems in ATM and aviation?

D1.1 FUTURE SCENARIOS AND RELEVANT HF CONCEPTS

Medium Sub Themes Very little. Very little (2) Very little Medium level(2) Medium. No (3) Not officially, I learn from internet Codes Not enough Individual efforts Not much Virtual environment