Annex To ED Decision 2014 035 R

Annex to ED Decision 2014/035/R

I General

GM1 SKPI General A. Purpose This Annex contains acceptable means of compliance (AMC) and guidance material (GM) for measuring the safety Key Performance Indicators (KPIs) and Performance Indicators (PIs) in accordance with the performance scheme Regulation which should be understood as Commission Regulation (EU) No 691/2010 as amended by Commission Implementing Regulation (EU) No 1216/2011 for the first reference period and Commission Implementing Regulation (EU) No 390/2013 for the second reference period.

AMCs are non-binding standards adopted by the European Aviation Safety Agency (EASA) (hereinafter referred to as the ‘Agency’) to illustrate means to establish compliance with the performance scheme Regulation.

When this AMC is complied with, the obligations on measurement of the safety KPIs in the performance scheme Regulation are considered as met.

However, the AMC contained in this Annex provide means, for the measurement of the safety KPIs. Should a Member State or an Air Navigation Service Provider (ANSP) wish to use different means to measure the SKPIs, they should: — inform the Agency thereof, and — be able to demonstrate, by means of evidence, that the outcome of the application of any alternative means maintains the level of compliance with the performance scheme Regulation and reaches a result that is comparable with the result of using the AMC in this Annex.

B. Objective The objective of this Annex is to establish the methodology for the measurement and verification of the following safety KPIs under the performance scheme Regulation: (a) Effectiveness of Safety Management (EoSM) and Just Culture (JC), which should be measured through a periodic answering of the questionnaires the content of which is provided in Appendices 1 to AMC2 SKPI, 1 to AMC3 SKPI, 1 to AMC9 SKPI and 1 to AMC10 SKPI. The filled in questionnaires by the entity subject to evaluation, and distributed in accordance with the performance scheme Regulation, should be verified as guided in AMC3 and 9 SKPI.

(b) Methodology for severity classification of reported safety-related occurrences. This should be done for each occurrence subject to the application of the methodology and should be verified as guided in AMC4, 5, 6, 7 and 8 SKPI.

Commission Regulation (EU) No 691/2010 of 29 July 2010 laying down a performance scheme for air navigation services and network functions and amending Regulation (EC) No 2096/2005 laying down common requirements for the provisions of air navigation services (OJ L 201, 3.8.2010, p. 1).

Commission Implementing Regulation (EU) No 390/2013 of 3 May 2013 laying down a performance scheme for air navigation services and network functions (OJ L121, 9.5.2013).

Page 4 of 64 Annex to ED Decision 2014/035/R C. Definitions and acronyms/initialisms Definitions ‘Airspace infringement’ is a flight into notified airspace without prev iously requesting and obtaining approval from the controlling authority of that airspace in accordance with international and national regulations.

Notified airspace includes controlled airspace, restricted airspaces and transponder mandatory zones or radio mandatory zones as implemented by the Member States.

‘ATM - specific occurrences’ are events or situations where a provider’ s ability to provide ATM, ATS, ASM or ATFM services is diminished or ceases. .

‘ATM/ANS system security’ is a s ituation in which the ATM/ANS services are lost or disrupted as a result of breach of system security.

‘Best (good) practice’ is a method, initiative, process, approach, technique or activity that is believed to be more effective at delivering a particular outcome than other means. It implies accumulating and applying knowledge about what is working and what is not working, including lessons learned and the continuing process of learning, feedback, reflection and analysis.

‘Major incident’ is an incident associated with the ope ration of an aircraft, in which safety of aircraft may have been compromised, having led to a near collision between aircraft, with ground or obstacles (i.e. safety margins not respected which is not the result of an ATC instruction).

‘Not determined’ means that insufficient information was available to determine the risk involved or inconclusive or conflicting evidence precluded such determination.

‘Occurrence with no safety effect’ is an occurrence which has no safety significance.

‘Reliability factor’ is the level of confidence in the assessment (scoring) undertaken, based on the data available.

‘Runway i ncursion’ is any occurrence at an aerodrome involving the incorrect presence of an aircraft, vehicle or person on the protected area of a surface designated for the landing and take-off of aircraft.

‘Safety culture’ means the shared beliefs, assumptions and values of an organisation.

‘Safety plan’ is a high -level safety issues assessment and related action plan. The safety plan is a key element of the safety programme.

‘Safety programme’ is an integrated set of regulations and activities aimed at improving safety.

‘Separation minima infringement’ means a situation in which prescribed separation minima were not maintained between aircraft.

‘Serious incident’ is an incident involving circumstances indicating that there was a high probability of an accident and is associated with the operation of an aircraft, which in the case of a manned aircraft, takes place between the time any person boards the aircraft with the intention of flight until such time as all such persons have disembarked, or in the case of an unmanned aircraft, takes place between the time the aircraft is ready to move with the purpose of flight until such time it comes to rest at the end of the flight and the primary propulsion system is shut down.

Page 5 of 64 Annex to ED Decision 2014/035/R ‘Significant incident’ is an incident involving circumstances indicating that an accident, a serious or major incident could have occurred if the risk had not been managed within safety margins, or if another aircraft had been in the vicinity.

Acronyms/Initialisms ACC Area Control Centre A/D MAN Arrival/Departure Manager AMC Acceptable Means of Compliance AI Airspace Infringement ANS Air Navigation Service ANSP Air Navigation Service Provider APP Approach Control Unit A-SMGCS Advanced Surface Movement Guidance & Control System AST Annual Summary Template ATC Air Traffic Control ATCO Air Traffic Control Officer ATM Air Traffic Management ATS Air Traffic Services CA Competent Authority CISM Critical Incident Stress Management CWP Controller Working Position ECR European Central Repository EoSM Effectiveness of Safety Management FAB Functional Airspace Block JC Just Culture IFR Instrument Flight Rules GM Guidance Material KPI Key Performance Indicator MO Management Objective MS Member State MTCD Medium-Term Conflict Detection NSA National Supervisory Authority PI Performance Indicator PRB Performance Review Body Page 6 of 64 Annex to ED Decision 2014/035/R QMS Quality Management System RAT Risk Analysis Tool RF Reliability Factor RI Runway Incursion RP Reference Period RMZ Radio Mandatory Zone SA Study Area SFMS Safety Framework Maturity Survey SI Standardisation Inspection SIA civil aviation Safety Investigation Authority SKPI Safety Key Performance Indicator SLA Service Level Agreement SMI Separation Minima Infringement SMS Safety Management System SPI Safety Performance Indicator SSP State Safety Programme STCA Short - Term Conflict Alert TCAS RA Traffic Collision Avoidance System Resolution Advisory TMA Terminal Manoeuvring Area, also known as Terminal Control Area TMZ Transponder Mandatory Zone TWR Tower Control Unit UAC Upper Area Control Centre VFR Visual Flight Rules Page 7 of 64

Annex to ED Decision 2014/035/R

II Effectiveness of Safety Management KPI

AMC1 SKPI Measurement of Effectiveness of Safety Management KPI — General GENERAL DESCRIPTION The Effectiveness of Safety Management (EoSM) indicator should be measured by verified responses to questionnaires at State/competent authority and service provision level, as contained in this Annex. For each question, the response should indicate the level of implementation, characterising the level of performance of the respective organisation.

EFFECTIVENESS LEVELS AND EFFECTIVENESS SCORE When answering the questions, one of the following levels of implementation should be selected: — Level A which is defined as ‘Initiating’ — processes are usually ad hoc and chaotic; — Level B which is defined as ‘Planning/Initial Implementation’ — activities, processes and services are managed; — Level C which is defined as ‘Implementing’ — defined and standard processes are used for managing; — Level D which is defined as ‘Managing & Measuring’ — objectives are used to manage processes and performance is measured; and — Level E which is defined as ‘Continuous Improvement’ — continuous improvement of processes and process performance.

An effectiveness level should be selected only if all the elements described in the questionnaire are fully observed by an ANSP or Member State/competent authority. If an ANSP or a Member State/competent authority has identified elements in various adjacent effectiveness levels, then they should take a conservative approach and select the lower effectiveness level for which all elements are covered.

Based on the responses, the following scores should be derived: — The overall effectiveness score should be derived from the combination of the effectiveness levels selected by the relevant entity (ANSPs or Member State/competent authority) against each question with the weightings as described in Appendix 2 to AMC2 SKPI and Appendix 2 to AMC3 SKPI; and — An effectiveness score for each Management Objective for the State/competent authority and for each Study Area (SA) for the ANSP.

GM2 SKPI Measurement of Effectiveness of Safety Management KPI — General A Management Objective (MO) has been derived and adapted for each of the elements of the ICAO State Safety Programme (SSP) and Safety Management System (SMS) as described in ICAO Annex 19.

For each Management Objective, a question (or questions) has been derived and the levels of effectiveness have been described.

For both State and ANSP levels, the Agency and the PRB will monitor the performance regarding this indicator based on the received answers and on the results of the verification process by the States/competent authority (CA) and by the Agency as presented in Figure 2 in AMC5 SKPI, section D.

Page 8 of 64 Annex to ED Decision 2014/035/R The questionnaires’ sole intent is to monitor the performance (effectiveness) of Member States/competent authorities and ANSPs regarding ATM/ANS safety management.

In order to facilitate this process for stakeholders, the Agency has developed an online tool which may be used by respondents, in place of the paper questionnaire, in order to complete and submit their responses to the questionnaires.

Member States/competent authorities and ANSPs are expected to provide evidence-based answers to these questionnaires as far as is practicable. The response levels assessed in the completed EoSM questionnaires should be used with the sole purpose of generating recommendations and associated plans for improvement of safety management. These response levels should not be used to generate findings in the context of standardisation or oversight inspections audits.

In accordance with Regulation (EU) No 628/2013 , if during a standardisation inspection a finding is raised by the Standardisation Team, in relation to the NSA/CA responses to the EoSM questionnaire, corrective action by the NSA/CA is required. Further, where a finding identifies that any of the questions in the EoSM questionnaire is scored higher than it should be, the score should be corrected and lowered to the appropriate level of implementation. A similar approach should be applied when the NSA/competent authorities raise findings to the ANSPs.

The outcome of standardisation inspections/oversight is not designed to be used for corrections of the scores towards a higher level of implementation.

AMC2 SKPI Measurement of Effectiveness of Safety Management KPI — State level The answers to the State-level questionnaire should be used to measure the level of effectiveness in achieving the Management Objectives defined in this Annex.

For each question, States should provide the Agency with information on the level of effectiveness (or level of implementation) and evidence to justify their answer.

Section A, below, defines which should be the corresponding Management Objectives for each component and element of the SSP framework.

The questionnaire which should be filled in by the Member States/competent authority is in Appendix 1 to AMC2 SKPI — Questionnaire for measurement of Effectiveness of Safety Management KPI — State Level.

A. Components, elements and management objectives Component 1 — State safety policy and objectives Element 1.1 State safety legislative framework: Management objective 1.1 — Implement the EU safety legislative and regulatory framework including, where necessary, the alignment of the national framework.

Element 1.2 State safety responsibilities and accountabilities Commission Implementing Regulation (EU) No 628/2013 of 28 June 2013 on working methods of the European Aviation Safety Agency for conducting standardisation inspections and for monitoring the application of the rules of Regulation (EC) No 216/2008 of the European Parliament and of the Council and repealing Commission Regulation (EC) No 736/2006 (OJ L 179 29.6.2013, p. 46) Page 9 of 64 Annex to ED Decision 2014/035/R Management objective 1.2 — Establish national safety responsibilities and maintain the national safety plan in line with the European Aviation Safety Plan, where applicable. The national safety plan should include the state policy to ensure the necessary resources.

Element 1.3 Accident and incident investigation Management objective 1.3a — Establish and maintain the independence of the civil aviation safety investigation authorities, including necessary resources.

1.3b — Establish means to ensure that appropriate safety measures are taken after safety recommendations have been issued by a civil aviation safety investigation authority.

1.3c — Ensure that civil aviation safety investigation authorities involve subject matter expertise from the ATM/A NS domain.

Element 1.4 Enforcement policy Management objective 1.4 — Establish appropriate, transparent and proportionate enforcement procedures, including the suspension, limitation and revocation of licences and certificates and the application of other effective penalties.

Element 1.5 Management of related interfaces Management objective 1.5a — Ensure adequate management of the internal interfaces within the NSA.

1.5b — Ensure adequate management of the external interfaces with relevant stakeholders.

Component 2 — Safety risk management Element 2.1 Safety requirements for the air navigation service provider’s SMS Management objective 2.1 — Establish controls which govern how service providers’ safety management systems (SMS) will identify hazards and manage safety risks.

Element 2.2 Agreement on the service provider’s safety performance Management objective 2.2 — Agree on safety performance of an individual, national or FAB service provider.

Page 10 of 64 Annex to ED Decision 2014/035/R Component 3 — Safety assurance Element 3.1 Safety oversight Management objective 3.1a — Attribution of powers to the NSA responsible for safety oversight of air navigation service providers.

3.1b — Establishment of a national safety oversight system and programme to ensure effective monitoring of the air navigation service provider’s (ANSP) compliance with the applicable regulations and monitoring of the safety oversight function.

Element 3.2 Safety data collection, analysis and exchange Management objective 3.2 — Establishment of mechanisms to ensure the capture and storage of data on hazards and safety risks and analysis of that data at ANSP and State level as well as its dissemination and exchange.

Element 3.3 Safety-data-driven targeting of oversight of areas of greater concern or need Management objective 3.3 — Establishment of procedures to prioritise inspections, audits and surveys towards the areas of greater safety concern or need or in accordance with the identified safety risks.

Component 4 — Safety promotion Element 4.1 Internal training, communication and dissemination of safety information Management objective 4.1a — Training of NSA personnel on applicable legislative and regulatory framework.

4.1b — Promotion of awareness of safety in formation and communication and dissemination of safety - related information amongst the aviation authorities within a State.

Element 4.2 External training, communication and dissemination of safety information Management objective 4.2a — Education/training of ANSP personnel and air traffic controllers (ATCO) training organisations on applicable legislative and regulatory framework.

4.2b — Promotion of awareness of safety information and communication and dissemination of safety - related information with external stakeholders.

Component 5 — Safety culture Element 5.1 Establishment and promotion of safety culture Page 11 of 64 Annex to ED Decision 2014/035/R Management objective 5.1 — Establishment and promotion of safety culture within the competent authority/NSA.

Element 5.2 Measurement and improvement of Safety Culture Management objective 5.2 — Establishment of procedures to measure and improve safety culture within the competent authority/NSA.

B. Scoring and numerical analysis When scoring the EoSM at State level, each response provided by the State or the competent authority in their questionnaire should be assigned a numerical value from 0 to 4, corresponding to levels A to E.

In addition, each question should be weighted from 0 to 1 according to its relevance to each Management Objective. The list of weighting factors for each question and MO can be found in Appendix 2 to AMC2 SKPI — List of weightings for evaluation of Effectiveness of Safety Management Questionnaire — State level.

Mathematically, the effectiveness score for each Management Objective is calculated from the questionnaire responses and weighting factors as follows: Where: — S is the effectiveness score for the State in management objective j; j — r is the numeric value of the response of State to question k in management objective j (value 0 to 4); kj — w is the weight factor of question k to management objective j (value 0 to 1); and kj — n is the number of questions in management objective j for which responses were provided by the j State.

In order to measure the effectiveness of safety management for the State, the following scores should be evaluated and monitored: — Overall effectiveness score: the overall score for each State estimated by taking the average of the scores over all Management Objectives.

— An effectiveness score for each Management Objective: scores over each Management Objectives, calculated with the use of the weightings from Appendix 2 to AMC2 SKPI — List of weightings for evaluation of Effectiveness of Safety Management Questionnaire — State level.

C. Mechanism for verification The results of States’ questionnaires are to be verified by means of Agency standardisation inspections.

Page 12 of 64 Annex to ED Decision 2014/035/R The coordination between the Agency and the competent authority should take place through the national ATM/ANS standardisation coordinator appointed by the State. The process is described in Figure 1 below.

The national coordinator should be responsible for coordination within the State authorities and for coordination with the ANSPs to provide the Agency with the responses to the questionnaires.

Figure 1 — Visualisation of the mechanism for verification at State level GM3 SKPI Effectiveness of Safety Management — Justifications for selected levels of implementation This GM provides some general principles for providing justifications and a worked example for the levels selected.

General principles It is anticipated that during a reference period there will be no changes, other than clarifications, to the Effectiveness of Safety Management questionnaire. This not only enables the progress of States to be monitored during a reference period, it also means that States ’ responses to the questionnaire only need to be updated within a reference period, instead of being completely revised. It should, therefore, be anticipated that for some questions (but not the whole questionnaire) the response from a State will be the same as in previous years.

The verification process performed by the Agency uses the justifications and evidence provided in the answers to the questionnaire, alongside pre-audit questionnaires, standardisation visits and information from the State NPP and USOAP audits. Where insufficient justification has been provided, the verification relies on alternative Page 13 of 64 Annex to ED Decision 2014/035/R information such as additional requests for clarification from the NSA point of contact. Therefore, in the interest of efficiency, States are encouraged to provide the necessary justifications in the first instance.

Extensive justification, when levels of implementation A or B are selected, is not necessary. A simple statement of the fact or of when the work was, or will be, initiated is sufficient. Justifications for levels C, D, and E are required and the general principles of what formulates a good answer from the perspective of verification are shown below.

(a) Justifications should be inclusive and explanatory, they should cover all relevant information and explain how the State achieved the level selected. Answers should not simply re-state the question.

(b) Answers should clearly explain why a State is at the level selected and should avoid explaining why they are not at the level above the one selected.

(c) In many of the questions, if the State selects level D or above, it must meet the requirements of both the level selected and the levels below. Where this is the case, the justification should cover all applicable levels; although a degree of consolidation is both acceptable and advisable.

(d) The questionnaire often refers to ‘ a mechanism ’, however, it should be recognised that the differing organisational structures and project management styles between NSAs may mean that, instead of a single mechanism, there could be a series of processes, projects or initiatives that deliver the desired end results. Such a description of the processes, projects or initiatives and their interaction, provided that they are coordinated, is equally acceptable.

(e) Justifications should contain specific information such as: (1) names or titles of the processes, documents, legislation or entities being described; (2) the job roles of the people responsible for the development, implementation or review of the item being described; (3) the intended purpose of the item being described; (4) when it was developed and implemented and how often it is reviewed; (5) an outline of the means or method used for development, implementation or review (such as meetings, project teams. etc.); and (6) the applicability of the item, for example whether it currently includes all the aspects intended or whether there are exceptions.

(f) Where evidence can be easily provided, such as hyperlinks to documents that are published online, these should have been provided, regardless of the language in use.

(g) Where references are made to evidence in published documents, the reference should describe where the evidence can be found in the document and where the document itself can be found. For example, hyperlinks may be provided to documents published online, but where the document is very long, a reference to the chapter or page number would be helpful.

(h) Where reference is made to internal documents, these should be cross-referenced with evidence from previous standardisation visits (if applicable). The reference should include sufficient detail for the verification team to be able to ask for the document, or the section of the document referred to, in a follow-up question to the State.

Page 14 of 64 Annex to ED Decision 2014/035/R Example response An example of a well-structured answer is shown below and the principles shown are applicable to any question at any level. In the example provided, the response shows that the State has achieved all of levels C and D, and even some of level E, but because it has not achieved all of level, E it must select level D. In the answer, it can be seen that the information provided is concise but describes the processes by providing references, naming the entities or job functions responsible for the work (but not naming individuals), and by providing additional information that allows the verification team to understand the quality of the work being done.

Page 15 of 64 Annex to ED Decision 2014/035/R The justification describes the way in which the requirements at level C are met, providing a By providing more information regarding the reference and, because in this policy, more confidence can be placed in the case it is available, a hyperlink answer and the verification team has a better Element 2.2 Agreement on the ser vice provider’s safety performance to the document onlin e. States idea of the way in which the NSA manages the should ensure that referenced policy in question. The extra information also MO2.2: Agree on safety performance of an individual, national or FAB service provider.

documents really do contain indicates that the NSA is already moving the information described and towards achieving level E, although not all of Q2.2 The CA/NSA has agreed with individual air navigation service providers on the safety performance (consistent with the that hyperlinks are correct.

the level E requirements are met.

ones contained in the national performance plans).

Acceptable safety levels are established through the ATM safety regulatory framework in a A Initiating limited number of areas and in an ad hoc manner.

Planning/ Initial B There is a plan in place to establish and formalise acceptable safety levels for the ATM system Implementation through the ATM safety regulatory framework. Implementation activities have commenced.

C Implementing Formalised acceptable safety levels have been established for the ATM system through the implementation of the State Safety Programme.

All of Implementing plus: D Managing & Measuring An evaluation of the acceptable safety levels is carried out on a regular basis and changes are introduced when necessary.

All of Managing & Measuring plus: Continuous The acceptable safety level review process is proactively incorporated within the overall aviation E Improvement safety system. Based on proactive recommendations, acceptable safety levels are linked to potential safety - cri tical hazards and events through the State Safety Programme.

Please provide justification for selected answer D: The national competent authority has developed an acceptable level of safety policy document (ref ALS2, first published in Ju ly 2011) which has been promulgated externally via an ANS NOTICE (available from the NSA website at www.NSA.gov.xx/ANSNOTICE7 - 2011 ) . The policy identifies a number of national level ANS safety targets. Further work is currently being undertaken by the NSA to broaden this activity t o derive individual unit level safety targets for those units where the level of activity makes this appr oach practicable. An evaluation of safety performance is undertaken by the ANS and Safety Analysis Departments on a 6 monthly basis. In addition, prior to conducting on - site audits of major units, safety performance trends for a selected number of safety i ndicators is reviewed. In addition, a summary of annual national ANS safety performance is reported upon formally in the Annual Safety Oversight Report, which can be found online at www.NSA.gov.xx/AnnualSafetyOversightReport2012 By providing the timescales (every six months) and the names of the departments involved, the justification describes succinctly By providing an example via the that the evaluation is carried out on a regular basis. By describing hyperlink, the verification team can check the review process prior to major audits , the justification shows the quality of the work to understand that the criteria are met in more than one way, providing more how well the requirements are being met.

Page 16 of 64 confidence in the answer.

Annex to ED Decision 2014/035/R AMC3 SKPI Measurement of Effectiveness of Safety Management KPI — ANSP level The answers to the ANSP-level questionnaire should be used to measure the level of effectiveness in achieving the management objectives defined in this AMC.

For each question, ANSPs should provide their NSA/competent authority with information on the level of effectiveness (or level of implementation) and evidence to justify their answer as indicated below.

Section A defines for each component and element of the ICAO Safety Management Framework the corresponding Management Objectives.

The questionnaire which should be filled in by the ANSPs is in Appendix 1 to AMC3 SKPI — Questionnaire for measurement of Effectiveness of Safety Management KPI — ANSP level.

A. Components, Elements and Management Objectives Component 1 — ANSP safety policy and objectives Element 1.1 Management commitment and responsibility Management objective 1.1 — Define the ANSPs’ safety policy in accordance with Regulation (EU) No 1035/2011 (Common Requirements).

Element 1.2 Safety accountabilities — Safety responsibilities Management objective 1.2 — Define the responsibilities of all staff involved in the safety aspects of service provision and responsibility of managers for safety performance.

Element 1.3 Appointment of key safety personnel Management objective 1.3 — Define the safety management function to be the responsible for the implementation and maintenance of SMS.

Element 1.4 Coordination of emergency response planning/contingency plan Management objective 1.4 — Define a contingency plan properly coordinated with the Network Manager, other interfacing ANSPs, other relevant stakeholders and FABs.

Element 1.5 SMS documentation Page 17 of 64 Annex to ED Decision 2014/035/R Management objective 1.5 — Develop and maintain the relevant SMS documentation that defines the ANSP’s approach to the management of safety.

Element 1.6 Management of related interfaces Management objective 1.6a — Ensure adequate management of the internal interfaces.

1.6b — Ensure adequate management of the external interfaces which may influence directly the safety of their services.

Component 2 — Safety risk management Element 2.1 Safety risk assessment and mitigation Management objective 2.1 — Develop and maintain a formal process that ensures the management of safety risks.

Component 3 — Safety assurance Element 3.1 Safety performance monitoring and measurement Management objective 3.1 — Establish means to verify the safety performance of the ANSP and the effectiveness of safety risk management.

Element 3.2 The management of change Management objective 3.2 — Establish a formal pro cess to identify changes and to ensure that safety risk assessment and mitigation are systematically conducted for identified changes.

Element 3.3 Continuous improvement of the SMS Management objective 3.3 — Establish a formal process to systematically identify safety improvements.

Element 3.4 Occurrence reporting, investigation and improvement Page 18 of 64 Annex to ED Decision 2014/035/R Management objective 3.4 — Ensure that ATM operational and/or technical occurrences are reported and those which are considered to have safety implications are investigated immediately, and any necessary corrective action is taken.

Component 4 — Safety promotion Element 4.1 Training and education Management objective 4.1 — Establish a safety training prog ramme that ensures that personnel are trained and competent to perform SMS - related duties.

Element 4.2 Safety communication Management objective 4.2 — Establish formal means for safety promotion and safety communication.

Component 5 — Safety culture Element 5.1 Establishment and promotion of safety culture Management objective 5.1 — Establish and promote safety culture within the ANSP.

Element 5.2 Measurement and improvement of safety culture Management objective 5.2 — Establish procedures to measure and improve safety culture within the ANSP.

Page 19 of 64 Annex to ED Decision 2014/035/R B. Mapping between Management Objectives, Study Areas and Questions The following table contains the mapping between the Management Objectives, Study Areas and the questions: MO SA — Q SA — Q MO Safety policy and Safety culture objectives SA1 - 1 5.1 1.1 SA2 - 3 SA1 - 2 5.2 SA1 - 3 3.4 1.2 SA2 - 1, SA2 - 4 Safety Responsibilities 1.3 SA2 - 2 SA2 - 1 1.2 1.4 SA4 - 3 SA2 - 2 1.3 1.5 SA4 - 1 SA2 - 3 1.1 1.6a SA7 - 1 SA2 - 4 1.2 1.6b SA7 - 2 Compliance with international Safety risk obligations management SA3 - 1 3.3 2.1 SA6 - 1 SA3 - 2 3.3 Safety assurance Safety standards and 3.1 SA9 - 1, SA9 - 2 procedures 3.2 SA6 - 1 SA4 - 1 1.5 3.3 SA3 - 1, SA3 - 2, SA10 - 1, SA4 - 2 4.2 SA11 - 2 SA4 - 3 1.4 3.4 SA1 - 3, SA8 - 1 Competency Safety promotion SA5 - 1 4.1 4.1 SA5 - 1 Risk management 4.2 SA4 - 2, SA8 - 2, SA8 - 3, SA6 - 1 2.1, 3.2 SA9 - 3, SA11 - 1, SA11 - 3 Safety interfaces Safety culture SA7 - 1 1.6a 5.1 SA1 - 1 SA7 - 2 1.6b 5.2 SA1 - 2 Safety reporting, investigation Table 1 : Mapping Management Objectives to and improvement Study Area questions SA8 - 1 3.4 SA8 - 2 4.2 SA8 - 3 4.2 Safety performance monitoring SA9 - 1 3.1 SA9 - 2 3.1 SA9 - 3 4.2 Operational safety surveys and SMS audits SA10 - 1 3.3 Adoption and sharing of best practises SA11 - 1 4.2 SA11 - 2 3.3 SA11 - 3 4.2 Table 2 : Mapping Study Area questions to Management Objectives Page 20 of 64 Annex to ED Decision 2014/035/R Given this mapping, at any point an interpretation from Management Objective to Study Area and vice versa should be possible.

C. Scoring and numerical analysis In order to be able to measure the effectiveness of safety management of the ANSP, the answers to the questions should be quantified and weighting factors which link the questions, Study Areas and the Management Objectives should be applied.

The responses provided by the ANSP on their questionnaires are assigned a numerical value from 0 to 4, corresponding to levels A to E.

In addition, each question should be weighted: — from 0 to 5 according to its relevance to each Study Area; and — from 0 to 1 according to its relevance to each Management Objective.

The list of weighting factors for each question, Study Area and Management Objective can be found in Appendix 2 to AMC3 SKPI — List of weightings for evaluation of Effectiveness of Safety Management Questionnaire — ANSP level.

Mathematically, the effectiveness score is calculated from the questionnaire responses and weighting factors as follows: Where: — S is the effectiveness score for ANSP in Study Area/Management Objective j; j — r is the numeric value of the response of ANSP to question k in Study Area/Management Objective kj j; — w is the weight factor of question k to Study Area/Management Objective j; and kj — n is the number of questions in Study Area/Management Objective j for which non-nil responses j were provided by the ANSP.

In order to measure the effectiveness of safety management for the ANSP, the following scores should be evaluated and monitored: — Overall effectiveness score: the overall score for each ANSP estimated by taking the average score over all Study Areas, using the weighting factors in Appendix 2 to AMC3 SKPI — List of weightings for evaluation of Effectiveness of Safety Management Questionnaire — ANSP level, section 2.1.

— An effectiveness score for each Management Objective: scores for each Management Objective calculated with the use of the weighting of questions described in Appendix 2 to AMC3 SKPI — List of weightings for evaluation of Effectiveness of Safety Management Questionnaire — ANSP level, section 2.2.

Page 21 of 64 Annex to ED Decision 2014/035/R D. Mechanism for verification The verification of the ANSP questionnaires by the NSA/competent authority should take place before the questionnaires and their results are submitted to the Agency. The verification mechanism is presented in Figure 2.

ANSPs should assign a focal point for the purpose of the verification process.

EASA + PRB Verified results Verified results Verified results … NSA 1 NSA 2 NSA n Results Results Results … … … ANSP 1 ANSP 2 ANSP n ANSP 1 ANSP 2 ANSP n ANSP 1 ANSP 2 ANSP n Figure 2 — Representation of verification mechanism of the ANSPs (normal procedure) The competent authority/NSA may allocate the detailed verification task to a qualified entity or other entity.

GM4 SKPI Measurement of Effectiveness of Safety Management KPI — ANSP level — Scoring and numerical analysis EXAMPLE FOR EoSM MEASUREMENT AT ANSP LEVEL The EoSM KPI is based on the EUROCONTROL Safety Framework Maturity Survey (SFMS) which has been implemented for several years at ANSP level. The numerical analysis at ANSP level has been validated during the implementation of the SFMS by EUROCONTROL and is based on Study Areas (SA). This is the reason why in section B of AMC5 SKPI the mapping is provided in order to match the Study Areas to the Management Objectives. The overall score of EoSM is using the weightings of the SA as established in SFMS and the scoring of each MO is based on average weightings.

Example: The following tables represent the results for calculating the scores for EoSM at ANSP level as follows: — Table 1 presents the association between the selected level of implementation and the numerical value from 0 to 4; — Table 2 illustrates the score calculated for each Study Area (SA) and the overall effectiveness score (average) of the EoSM at ANSP level; and — Table 3 presents the effectiveness score for each Management Objective.

Page 22 of 64 Annex to ED Decision 2014/035/R SAs SA score MOs MO score QUESTIONS Selected level Numerical value 1 52.7 1.1 100 2 57.4 1.2 50 SA1 - 1 A 0 3 60.3 1.3 75 SA1 - 2 E 4 4 54.7 1.4 25 SA1 - 3 E 4 5 52.7 1.5 50 SA2 - 1 B 1 1.6a 50 6 53.5 SA2 - 2 D 3 1.6b 25 7 47.7 SA2 - 3 E 4 2.1 75 8 51.4 SA2 - 4 D 3 3.1 50 9 51.1 SA3 - 1 D 3 3.2 75 10 56.0 SA3 - 2 D 3 3.3 62.5 11 54,4 SA4 - 1 C 2 3.4 50 average 53,8 SA4 - 2 D 3 4.1 75 SA4 - 3 B 1 4.2 51 SA5 - 1 D 3 5.1 0 SA6 - 1 D 3 5.2 100 SA7 - 1 C 2 SA7 - 2 B 1 SA8 - 1 A 0 SA8 - 2 C 2 SA8 - 3 C 2 SA9 - 1 D 3 SA9 - 2 B 1 SA9 - 3 C 2 SA10 - 1 D 3 SA11 - 1 C 2 SA11 - 2 B 1 SA11 - 3 B 1 Table 1 Table 2 Table 3 The application of the formula for calculation of the overall effectiveness score is illustrated for the calculation of the score for SA1 as follows: S1 = 100*(0*5+4*5+4*4+1*4+3*2+4*5+3*2+3*1+3*1+2*2+3*3+1*3+3*4+3*4+2*5+1*3+0*5+2*5+2*3+3* 2+1*4+2*4+3*4+2*4+1*4+1*5)/(4*(5+5+4+4+2+5+2+1+1+2+3+3+4+4+5+3+5+5+3+2+4+4+4+4+4+5)) S1 = 52,7 In this calculation the numerical values for each question from Table 1 are multiplied by the corresponding weightings for SA1, taken from section 2.1 of Appendix 2 to AMC5 SKPI: Page 23 of 64 Annex to ED Decision 2014/035/R Then the result is divided by the sum of weights: GM5 SKPI Measurement of Effectiveness of Safety Management KPI — ANSP level — Verification mechanism VERIFICATION OF ANSP EoSM BY THE NSA/COMPETENT AUTHORITY When verifying the questionnaires completed by an ANSP for EoSM, the NSA may organise bilateral interview sessions. In these interview sessions, the NSA coordinator may ask the ANSP focal point some additional questions and request some additional evidence in order to verify the correctness of the answers provided to the questionnaires. Examples of the verification questions, together with examples of the possible outcome of the fulfilment of the objectives of EoSM for each level of implementation, are provided in Appendix 1 to GM5 SKPI — Verification of ANSP EoSM by the NSA/competent authority.

COORDINATION BETWEEN THE NSAs FOR THE VERIFICATION OF THE ANSPs The competent authorities/NSAs might need better coordination between them in the verification process in order to achieve consistent and comparable results at European level. Such coordination could be coordinated and facilitated by the Agency, supported by the PRB and EUROCONTROL. One potential solution could be the extension of the terms of reference for the NSA Coordination Platform (NCP) in the field of harmonisation of the verification mechanism of the safety KPIs at ANSP level.

Notwithstanding the above and notwithstanding the fact that NSA may delegate the verification task to another entity, the responsibility for verification of the safety KPI measurement at ANSP level stays with the overseeing competent authority/NSA.

Page 24 of 64

Annex to ED Decision 2014/035/R

III Severity classification based on the risk analysis tool methodology

AMC4 SKPI Severity classification based on the risk analysis tool methodology — General GENERAL DESCRIPTION The severity part of the risk analysis tool methodology dedicated to operational occurrences should follow the principle of evaluating several criteria and allocating a certain score to each criterion, depending on how severe each criterion is evaluated to be.

Each criterion should have a limited number of options with corresponding scores. Some criteria have an ATM Ground and an ATM Airborne component and both scores should be counted when evaluating the ATM Overall score. Other criteria should be only relevant either for ATM Ground or ATM Airborne.

The overall score for severity of an occurrence should be the sum of the scores allocated to each applicable individual criterion.

The overall score for the severity of an occurrence should be built from the sum of the score allocated to the risk of collision/proximity (itself a sum of the score allocated to the separation and the score allocated to the rate of closure) and the degree of controllability over the occurrence.

The severity of the ATM- specific occurrences should refer to the service provider’s capability to provide safe ATM/CNS services. The criteria which should be considered are: the service affected, service/function provided, operational function, type of failure, extent of the failure scope and duration.

The severity of occurrences reported by Member States should be the ATM Overall. For ATM-specific occurrences, the ATM Overall coincides with ATM Ground severity.

Member States should ensure that arrangements are in place for reporting of the ATM Overall severity score.

AMC5 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Separation Minima Infringements (SMIs) The severity of Separation Minima Infringements should be calculated as the sum of the scores totalled in each of the two main criteria: 1. Risk of collision; and 2. Controllability.

A. Risk of collision The risk of collision should be determined by the sum of the scores for the following sub-criteria: 1. Separation — based solely on the minimum distance achieved between aircraft or aircraft and obstacles. The greatest value between the horizontal and vertical in percentage of the applicable separation should be considered.

2. Rate of closure based on the relative relevant (horizontal/vertical) speed measured at the moment the separation is infringed. The greatest of the predefined intervals for each of the horizontal and vertical speeds should be considered for the evaluation if the separation is lost after the crossing point (i.e. if the aircraft are on diverging headings when the separation is lost, then the rate of closure is considered ‘none’).

Page 25 of 64 Annex to ED Decision 2014/035/R The following table should be used to determine the scores of the criteria ‘separation’ and ‘rate of closure’: Risk of collision ATM ATM ATM overall RF ground airborne weight 0 0 0 to 10 20 Minimum separation achieved ATM Ground 1 1 Separation > 75 % minimum OR ATM airborne 3 3 Separation > 50 %, < = 75 % minimum 7 7 Separation > 25 %, < = 50 % minimum Separation 10 10 Separation <= 25 % minimum 0 0 0 to 5 10 Rate of closure NONE ATM Ground 1 1 Rate of closure LOW (< = 85 knots, OR ATM < = 1 000 ft/mn) airborne 2 2 Rate of closure MEDIUM (> 85 and < = 205 knots, > 1 000 and < = 2 000 ft/mn) 4 4 Rate of closure HIGH (> 205 and Rate of closure < = 700 knots, > 2 000 and < = 4 000 ft/mn) 5 5 Rate of closure VERY HIGH (> 700 knots, > 4 000 ft/mn) For the risk of collision, either ATM Ground or ATM Airborne severity should be scored, not both. The ATM Airborne severity should be used only in cases where ATC is not responsible for providing separation (i.e.

certain classes of airspaces; e.g. close encounter between IFR and VFR flights in Class E airspace).

B. Controllability Controllability should be the second m ajor criterion of severity and describes the ‘level of control’ maintained over the situation (Air Traffic Controllers (ATCOs) and pilots supported by safety nets).

The controllability score should be defined by the following sub-criteria: 1. conflict detection; 2. planning; 3. execution; 4. ground safety nets (e.g. STCA);, 5. recovery; 6. airborne safety nets (e.g. TCAS); and 7. airborne execution of TCAS RA.

Conflict detection should refer to ATM ground detection; therefore, the ATM Overall score should have the same score as ATM Ground. ATM Airborne should not be scored here. There are three possible scenarios: — ‘Potential conflict DETECTED’ includes cases where the conflict is detected , but ATC decided to accept the situation.

— ‘Potential conflict detected LATE’ when there is not enough time to make and/or execute the plan. It should not be scored whenever separation is lost; consideration should be taken with regard to the Page 26 of 64 Annex to ED Decision 2014/035/R circumstances involved. In units with STCA with ‘loo k- ahead’ time (predictive STCA) , the conflict could be detected due to the predictive STCA. If ATCO become aware of the conflict only through the predictive STCA, then it should be scored as ‘Potential conflict detected LATE ’.

— The score ‘Potential conflict NOT detected’ is self -explanatory.

In cases such as level busts or other incidents where ATC cannot form prior plan, conflict detection should not be applicable and a zero should be scored to maintain the Reliability Factor tracked as explained in section D.

ATM ATM ATM RF ground airborne overall weight Potential conflict DETECTED 0 0 to 5 Potential conflict detected ATM 10 LATE ground Detection Potential conflict NOT detected Planning refers to the ATM Ground plan and, therefore, the ATM Overall score should have the same score as ATM Ground. ATM Airborne should not be scored here. The performance, the timing and efficiency of the ATM Ground planning should be assessed. The plan refers to the first plan developed by ATC to solve the potentially hazardous/conflict situation detected in the previous step. This plan should be referred to in the subsequent execution steps but not necessarily in the recovery step.

— When the planning is either late or does not lead to a timely and effective resolution of the conflict, then ‘Plan INADEQUATE’ should be scored.

— When ‘Conflict NOT detected’ is scored, then also ‘NO Plan’ and ‘NO Execution’ should be scored.

— Whenever conflict detection is not applicable (such as level bust cases), then the planning sub- criterion is not applicable and a zero should be scored to maintain the Reliability Factor tracked as explained in section D.

ATM ATM ATM RF ground airborne overall weight Plan CORRECT 0 0 to 5 Plan INADEQUATE 3 ATM 10 ground Planning NO plan 5 Execution refers in general to ATM Ground execution in accordance with the developed plan but it should have ATM Ground and ATM Airborne components. Execution refers to the execution of the first plan developed by ATC to solve the detected hazardous/conflict situation. When assessing the execution, the time and efficiency of that execution should be assessed. Airborne execution of the received instructions/clearances should be scored as ATM Airborne.

— ATM Ground execution should be scored as ‘Execution INADEQUATE’ when it is not timely or not effective. It refers to the same plan developed in the planning criterion, prior to the separation Page 27 of 64 Annex to ED Decision 2014/035/R infringement. It includes the cases when it is contrary to any prior good planning. The airborne execution should be scored separately as ATM Airborne.

— When no conflict is detected, ‘NO plan’ and ‘NO execution’ should apply. ‘NO execution’ also should comprise cases when there is detection and a plan but this is not implemented at all.

— Whenever conflict detection and planning are not applicable, such as deviation from ATC clearance (e.g. runway incursion due to pilot deviation from ATC clearance), then the execution criterion for ATM Ground should also not be applicable and should be scored 0.

— In case of no pilot deviation from the instructed plan by the ATCO, ATM Overall should have the same score as ATM Ground and ATM Airborne should be scored 0.

ATM ATM ATM RF ground airborne overall weight 0 to 15 Execution CORRECT 0 0 ATM Execution INADEQUATE 3 5 ground + 10 ATM Execution NO Execution 5 10 airborne Ground safety nets (STCA) (Short-Term Conflict Alert or other similar ground safety net) Only current (not-predictive) STCA should be scored here. This criterion follows the principles of TCAS, except when the STCA is a ground-based defence. Cases of false/nuisance alerts should be disregarded.

This sub-criterion should have only the ATM Ground element. ATM Airborne should not be scored here.

— If current STCA triggers and is used by the ATCO, then it served its purpose as designed and a ‘zero’ should be scored for ATM ground. As such, the units with and without STCA are scored in the same manner; — When the conflict is detected by the ATCO before the STCA triggers, then a zero should be scored; — ‘No detection’ should be scored when the conflict was not detected or detected late by the ATM Ground and STCA should have been triggered according to its implemented logic, but it failed to function. Hence, the ground safety net barrier did not work.

ATM ATM ATM RF ground airborne overall weight Current STCA triggered 0 0 or 5 ATM STCA ground Non-current STCA 5 Recovery from the actual incident is the phase requiring immediate action to restore the safety margins (e.g. separation) or at least to confine the hazard. Recovery starts from the moment the safety margins have been breached (potentially due to an inadequate or missing initial plan to solve the hazardous situation). This sub-criterion applies to both ATM Ground and ATM Airborne. Therefore, ATM Overall should be the sum of the ATM Ground and ATM Airborne values.

Page 28 of 64 Annex to ED Decision 2014/035/R From this step (recovery), the plan should be considered as a new one and as different from the first plan established in the detection/planning phase. It is seeking the performance of bringing the system back within its safety envelope (such as re-establishment of the separation minima). Recovery might include, depending on the type of occurrence (e.g. airspace in which it occurred and services to be provided), cases where traffic information or avoiding actions were issued by ATC.

— ‘Recovery CORRECT’ should be scored when the actio ns taken after the separation minima infringement were adequate and the separation was re-established within a reasonable time frame.

— Scoring ‘Recovery INADEQUATE’ indicates that the ATM reaction, after the actual incident is declared, had not improved the situation.

— When scoring ‘NO recovery’, consideration should be given as to whether a TCAS RA or pilot see-and- avoid action was triggered, as this could be the reason to not follow the ATC instructions. In this case, there should be no penalty on the ATM Airborne part.

— When the aircraft are already diverging, then recovery should be scored as not applicable and a zero value should be given.

ATM ATM ATM RF ground airborne overall weight Recovery CORRECT 0 0 0 to 25 Recovery INADEQUATE 5 6 ATM NO recovery or the ATM ground + 10 ground actions for recovery ATM Recovery have worsened the situation or 10 15 airborne ATM airborne has worsened the situation Airborne Safety Nets (TCAS) — The TCAS sub-criterion should be scored only for useful TCAS RAs (as per ICAO definitions). A similar logic applies for see-and-avoid environments where TCAS does not function.

Note: For this sub-criterion, ATM Overall should take the score of ATM Airborne. ATM Ground should be scored for the purposes of the Reliability Factor evaluation, as described in section D, and the ATM Ground severity evaluation when done separately from the ATM Overall.

— The ‘No TCAS RA’ option should be used in situations when the geometry of the encounter would require a TCAS RA (based on ICAO TCAS logic) and that did not occur.

— ‘TCAS triggered…’ should be scored as not applicable (i.e. a score of zero should be given) if adequate ATC instructions are issued before the pilot reaction due to TCAS RA.

— For cases where TCAS RA contributed significantly to the recovery and reestablishment of separation, ‘TCAS triggered...’ should be scored.

ATM ATM ATM RF ground airborne overall weight TCAS triggered or see-and- 0 or 10 avoid pilot decision (in the 10 0 ATM absence of TCAS) TCAS airborne NO TCAS RA 0 10 Page 29 of 64 Annex to ED Decision 2014/035/R Airborne execution of TCAS RA (or application of see-and-avoid in cases where TCAS is not applicable) and recovery is a criterion to gather the complementary performance to ATM ground.

— ‘ Airborne INSUFFICIENTLY followed RA’ should apply when pilot action is not reacting fully in accordance with the TCAS RA.

— ‘ Airborne INCORRECTLY followed RA (or, in the absence of RA, took other inadequate action)’ should be scored whenever the pilot actions were either missing or contradictory (e.g. did not follow the RA). A contradictory reaction or non-reaction to a TCAS RA should be considered as the worst possible case.

ATM ATM ATM RF ground airborne overall weight Airborne followed RA (or, in absence of RA, took other effective action, as a result of see-and-avoid decision) 0 to 15 Airborne INSUFFICIENTLY ATM 10 followed RA airborne Airborne INCORRECTLY followed RA (or, in the absence of RA, 15 Pilot execution of TCAS RA took other inadequate action) The score of the controllability criterion should be the sum of the scores of its components: Detection, Planning, Execution, STCA, Recovery, TCAS RA and Pilot Action.

C. Final scores Once all criteria have been evaluated and scored accordingly, the final score for severity should be the sum of the scores for ‘Risk of collision’ and ‘Controllability’.

When the overall scores have been calculated as above, the equivalence with the severity for ATM Overall should be as follows: ATM Overall Score Severity class Between 0 – 9 No safety effect (E) Between 10 – 17 Significant incident (C) Between 18 – 30 Major incident (B) Higher than 31 Serious incident (A) D. Reliability Factor Every criterion of the methodology should have its own importance for the evaluation of severity. If there is no information for the evaluation of a certain criterion or the information available is ambiguous or the scoring panel cannot agree on the choice that should be made, then these should be identified as missing elements from the methodology.

Page 30 of 64 Annex to ED Decision 2014/035/R In order to record and track the influence of the missing elements on the final severity score, an Overall Reliability Factor (RF) should be calculated in parallel with the severity score. The RF should be based purely on the amount of criteria which are considered when evaluating the severity score.

Each criterion should have its associated RF weight. The predefined RF weight per criterion is presented in the last column (RF) in the tables in sections A and B. The value of the Overall RF should be the sum of the RF weights associated with the criteria which are taken into account for the severity evaluation.

Not all criteria should be always applicable (e.g. units without safety nets, or safety nets did not trigger).

Any criterion positively known not to be applicable to the particular situation under consideration should be scored with a zero value and its associated RF weight should be added to the overall RF.

In the situation where a certain criterion is applicable but there is not enough information to make a judgement from the investigation report (due to lack of data or lack of clarity of the details), the score for that criterion should have value ‘blank’. ‘Blank’ value for a certain criterion indicates that the relevant RF weight should not take part in the calculation of the Overall RF.

If during the evaluation of two different occurrences a certain criterion is scored in the first case as zero (0) and in the second case as ‘blank’, the ATM overall severity score in both cases should have the same value but the RF should be lower in the second case.

If a score is recorded for a specific criterion, then its RF weight should be added to the overall RF value as follows: — For the Separation, Rate of closure, Conflict detection, Planning, Ground safety nets (STCA) criteria, which have only ATM Ground component, full RF value should be added if the ATM Ground value is recorded (except for Separation and Rate of closure where the ATM Ground value could be replaced by ATM Airborne).

— For the Execution, Recovery and Airborne Safety Nets (TCAS) criteria, which have both ground and airborne components, half of the RF value should be added if the ATM Ground value is recorded, and half of the weight if the ATM Airborne value is recorded.

— For the airborne execution of TCAS RA criterion, which has only an airborne component, full RF value should be added if the ATM airborne is recorded.

The RF should reach a value of 100 when all data for all criteria have been entered.

The Overall RF associated with the occurrence should be calibrated in such a way that the results of the severity assessment should be acceptable if the Overall RF has a minimum value of 70. Whenever there is not enough information (RF < 70) the occurrence should be categorised as ‘Not determined’ (D), regardless of the severity indicated after application of the methodology.

GM6 SKPI Severity classification based on the risk analysis tool methodology for Separation Minima Infringements — General description The process for evaluation of occurrences severity is presented in the following diagram: It can be contended that if the occurrence has already reached maximum possible severity, any additional data will not change the severity value. However, the occurrence is still recorded as not determined, since it is important to identify any missing data.

Page 31 of 64 Annex to ED Decision 2014/035/R

Barrier model Barrier model

ACCIDENT ACCIDENT ACCIDENT

Separation Rate of closure

ATC ATC PILOT PILOT

(V or H) (V or H) Entropy Entropy

CONTROL CONTROL CONTROL

RISK of COLLISION RISK of COLLISION CONTROLABILITY CONTROLABILITY

SEVERITY SEVERITY

Figure 3 — Visualisation of the evaluation of occurrences severity Distinction between ATM Ground and ATM Overall severity may be made in order to allow ANSPs to identify their own contribution to any occurrence, identify causes and possible mitigation plans and/or corrective actions. In order to be able to fill in all necessary fields for the ATM Overall severity, information not immediately available to ANSPs may be required, such as information on the existence or not of a TCAS RA on the causal factors on the airborne side.

Different occurrences scenarios may be considered when evaluating severity as it is done in the EUROCONTROL Risk Analysis Tool (RAT): Scenario Description 1. More than When two or more aircraft are involved in the occurrence and a standard separation is one aircraft defined — usually for incidents with airborne aircraft, e.g. usually involving separation minima infringements.

2. Aircraft — When the occurrence is an encounter between two aircraft under tower ATC. This aircraft tower includes situations where a) both aircraft are airborne; or b) both aircraft are on the ground; or c) one aircraft is airborne and one is on the ground.

In addition, this should be used for occurrences involving one aircraft and a vehicle that, at the time of occurrence, was occupying/intersecting an active runway.

3. Aircraft with When the occurrence is an encounter between an aircraft and a vehicle (includes ground towed aircraft). In this situation, the aircraft could be on the ground or it could be Movement airborne.

4. One aircraft When only one aircraft is involved in the occurrence (e.g. airspace infringement, level bust without involvement of a second aircraft, loss of separation with ground and/or obstacles). This also applies for near-CFIT occurrences.

5. ATM-specific To be applied in cases of technical occurrences influencing the capability to provide occurrence safe ATM/ANS services.

Page 32 of 64 Annex to ED Decision 2014/035/R The following link may be made between the occurrences scenarios as in RAT and the occurrence types referred to in the performance scheme Regulation: — Separation minima infringements: scenario 1; — Runway incursions: scenarios 2, 3 and 4; — ATM-specific occurrences: scenario 5.

GM7 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Separation Minima Infringements — Risk of collision — Score determination Example: If in a Separation Minima Infringement occurrence: — the minimum separation achieved was 60 % horizontally and 30 % vertically; — the rate of closure at separation loss was 160 kts and 3 000 ft/min; and — ATC was providing radar separation, then: — ATM Ground is scored 3 for separation (highest value of the two separations, i.e. the value for 60 % horizontally); — ATM Ground is scored 4 for rate of closure (highest value of the two possible marks, i.e. the value 3 000 ft/min); and — ATM Overall for Risk of collision is 7 with RF 30.

GM8 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Separation Minima Infringements — Controllability score determination The score of controllability may be used to facilitate an evaluation of the amount of hazard or entropy. If the situation is controlled, even if separation is lost, it is nevertheless recovered by the ATM system and not by chance. For this step, the typical defence barriers, as they apply chronologically, may be followed.

The ATM Ground elements may be used to evaluate whether and how ATC ( ‘ ATC ’ means not only the ATCO, but the ATCO supported by ATM system) worked the conflict situation between the aircraft later involved in the actual occurrence. The global picture should be considered and not only the two aircraft between which separation was lost. In certain cases, while trying to work an aircraft pair, ATC could generate an occurrence between another pair. All aircraft relevant to the occurrence under analysis should be considered.

When evaluating the criterion Ground Safety Nets (STCA): — ‘ Predictive STCA ’ is meant to be an STCA that triggers an alarm with sufficient time in advance of an infringement of the separation minima allowing air traffic controllers enough time to react; and — ‘ Current STCA ’ is meant to be an STCA that triggers at the time when the separation minima starts to be infringed.

When evaluating the criterion Airborne Safety Nets (TCAS) , it should be noted that this sub-criterion has an ATM Ground element, but the ATM Overall only takes the value of ATM Airborne. The purpose of the ATM Ground element here is to allow evaluating the ATM Ground value as described in GM8. When ATM Ground is scored 10, the ATM Airborne and ATM Overall for criterion Airborne Safety Nets (TCAS) should be scored zero. In such a case, it is quite possible to have ATM Ground with higher score than ATM Overall Page 33 of 64 Annex to ED Decision 2014/035/R and, when evaluating severity in accordance with the table in GM8 SKPI, this could result in a higher severity for ATM Ground than for ATM Overall. This indicates the higher contribution to the occurrence of the ATM Ground component compared to the ATM Overall.

Example of controllability score determination: Conflict detected, planning inadequate, execution inadequate by ATC, corrected by pilot, STCA not applicable, recovery corrected by ATC and pilot, TCAS RA needed but not triggered, pilot response not applicable: Ground Airborne Airborne Conflict Safety Safety execution Total Planning Execution Recovery detection Nets Nets of TCAS score (STCA) (TCAS) RA Yes Inadequate Inadequate N/A Correct N/A Ground 6 0 3 3 0 0 0 Airborn Correct Correct No N/A e 0 0 10 0 RF 10 10 5+5 10 5+5 5+5 10 70 ATM Overall Controllability = Conflict detection + Planning + Execution + Ground Safety Nets (STCA) + Recovery + Airborne Safety Nets (TCAS) + Airborne Execution of TCAS RA = 0+3+3+0+0+10+0 = 16 GM9 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Separation Minima Infringements — Final scores Example: Following the score determination in GM6 and 7 SKPI, Severity ATM Ground = Risk of collision score Ground + Controllability score Ground = 7 + 6 = 13 Severity ATM Overall = Risk of collision score Overall + Controllability score Overall = 7 + 16 = 23 When evaluating the ATM Ground value only, the table from AMC7 SKPI, D may be extended as follows: ATM Ground value Severity ATM Overall value Sev erity class class Between 0 – 9 No safety Between 0 – 9 No safety effect effect Between 10 – 17 Significant Between 10 – 17 Significant incident incident Between 18 – 30 Major Between 18 – 30 Major incident incident Higher than 31 Serious Higher than 31 Serious incident incident Page 34 of 64 Annex to ED Decision 2014/035/R Example: Severity class ATM Ground for score 13 = Significant incident Severity class ATM Overall for score 23 = Major incident GM10 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Separation Minima Infringements — Reliability Factor Example: When scoring ‘Not Applicable’ as in GM7 for the Airborne Execution of TCAS RA (because there was no TCAS RA in the example provided), the value of the score is 0. Nevertheless, the relevant value of the RF is added to the RF Overall.

Example: In the examples in GM6 and GM7, the RF for each criterion is also recorded. The overall RF based on these examples is calculated to be 100, which means that the severity in this example is evaluated with all the necessary data available. In this case, and in other cases where the overall RF is calculated to be 70 or more, the resulting severity may be considered as valid.

The same example as in GM7 may be presented with some data missing (value ‘blank’) as follows: Conflict Planning Execution Ground Recovery Airborne Airborne Total detection Safety Safety execution score Nets Nets of TCAS RA (STCA) (TCAS) Ground No data Inadequate Inadequate N/A Correct No data 6 blank 3 3 0 0 Blank Airborne No data No data No data No data 10 blank blank Blank blank RF 0 10 5+0 10 5+0 0+0 0 30 In order to evaluate the Overall RF of this example, we need to add to the RF of Controllability the RF of Risk of Collision. If we use the value of RF of the Risk of Collision as calculated in GM7 (30), the Overall RF will have a value of 60. Since the Overall RF < 70, the occurrence should be categorised as ‘Not determined’ (D).

AMC6 SKPI Severity classification based on the risk analysis tool methodology — Methodology for Runway Incursions Applying the severity classification methodology for Runway Incursions, the severity should be calculated as the sum of the total scores in each of the two main criteria: 1. Risk of collision; 2. Controllability.

A. Risk of collision The risk of collision should be determined by the sum of the scores for the following sub-criteria: 1. Separation. When evaluating the severity of runway incursion, this criterion should be interpreted as safety margin infringed. The moderation panel/investigators should, based on experts judgment, choose a score between 0 and 10, based on the perceived safety margin achieved. If there is no agreement on the safety margin, then the moderation panel/investigators will not score the criterion at all and the field should be left blank. This should be reflected in the value of the Reliability Factor by not adding the RF weight for the separation criterion.

Page 35 of 64 Annex to ED Decision 2014/035/R ATM ATM Risk of collision ATM ground RF weight airborne overall Safety margin achieved 0 0 0 to 10 Safety margin infringed minor 1 – 3 1 – 3 ATM Ground Safety margin infringed medium 4 – 6 4 – 6 OR ATM Safety margin infringed significant separation 7 – 9 7 – 9 airborne Safety margin infringed critical 10 10 2. Rate of closure — based on the vertical and horizontal speed, measured at the moment the safety margin is considered to have been lost. The greatest of the predefined intervals for each of the horizontal and vertical speeds are to be considered for the evaluation.

Depending on the situation, speed intervals should be applied as follows: — More than one aircraft — no standard separation defined; and — Aircraft with ground movement.

In cases of unauthorised entry on the runway when no other aircraft/vehicle/person was present, the rate of closure should be ‘NONE’.

Aircraft with ATM ATM ATM More than one aircraft — no ground movement ground airborne overall RF standard separation defined weight Rate of closure 0 0 0 to 5 Rate of closure NONE NONE ATM Ground Rate of closure LOW 1 1 Rate of closure LOW OR ATM (<= 20 knots) (<= 50 knots, <= 500 ft/mn) airborne Rate of closure 2 2 Rate of closure MEDIUM (>50 MEDIUM (>20 and and <= 100 knots, > 500 and <= 40 knots) <= 1 000 ft/mn) Rate of closure 4 4 Rate of closure HIGH (>100 rate of closure HIGH (>40 and and <= 250 knots, > 1 000 and <= 80 knots) <= 2 000 ft/mn) Rate of closure 5 5 Rate of closure VERY HIGH VERY HIGH (>250 knots, > 2 000 ft/mn) (>80 knots) For the risk of collision, either ATM Ground or ATM Airborne severity should be scored and not both ATM Ground and ATM Airborne. The ATM Airborne severity should be used only in cases where ATC is not responsible for providing separation.

B. Controllability The scoring for controllability should follow the same logic as in AMC5 section B, with only a few exceptions, as follows: Page 36 of 64 Annex to ED Decision 2014/035/R — STCA is not appropriate for this encounter, hence, it should be replaced by more general aerodrome ground safety nets, such as RIMCAS (Runway Incursion Monitoring and Collision Avoidance System); — Airborne Safety Nets (TCAS) is not normally available when Runway Incursions occur, therefore, only pilot see-and-avoid action should be considered. Lack of see-and-avoid should be scored in the case of low visibility and IMC conditions.

— All other sections are identical to the previous scenario, with the exception of the Safety Nets where A-SMGCS (Advanced Surface Movement Guidance & Control System) or RIMCAS should be considered, and the see-and-avoid part where driver action should also be taken into account, alongside that of the pilot.

The controllability score should be defined by the following aspects: 1. Conflict detection; 2. Planning; 3. Execution; 4. General ground safety nets, e.g. A-SMGCS; 5. Recovery; 6. Airborne Safety Nets (see-and-avoid); and 7. Pilot/driver execution of see-and-avoid.

The controllability scoring should be identical in all aspects to that in section B of AMC5 SKPI.

C. Final scores The final scoring should be identical in all aspects to that in section C of AMC5 SKPI.

D. Reliability Factor The Reliability Factor evaluation should be identical to the description in section D of AMC5 SKPI.

AMC7 SKPI Severity classification based on the risk analysis tool methodology — Methodology for ATM-specific occurrences A. Overview The ATM-specific occurrences severity evaluation should be based on a combination of criteria. For each criterion, a number of options should be available.

The combination of the chosen options for each criterion should provide the severity of an ATM-specific occurrence.

The following criteria should be considered when determining the severity of an ATM-specific occurrence: 1. Service affected; 2. Service/Function provided; 3. Operational function; 4. Type of failure; 5. Extension; Page 37 of 64 Annex to ED Decision 2014/035/R 6. Scope; and 7. Duration.

B. Options for ATM-specific occurrences The following options should be considered when evaluating each criterion in AMC7 SKPI section A: 1. Criterion ‘Service affected’ — the effect of the system failure should be assigned to one of the following services: a. (Upper) Area Control Centre — ATC service for controlled flights in a block of airspace; b. Approach Control — ATC service for arriving or departing controlled flights; c. Aerodrome Control — ATC service for aerodrome traffic; d. Oceanic Control — ATC service for controlled flights over the high seas; and e. Flight Information Service — service provided for the purpose of giving advice and information useful for the safe and efficient conduct of flights.

2. Criterion ‘Service/Function provided’ — the following options should be available for the Service/Function criterion: a. Communication — aeronautical fixed and mobile services to enable ground-to-ground, air-to- ground communications for ATC purposes; b. Navigation — those facilities and services that provide aircraft with positioning and timing information; c. Surveillance — those facilities and services used to determine the respective positions of aircraft to allow safe separation; d. Air Traffic Services — the various flight information services, alerting services, air traffic advisory services and ATC services (area, approach and aerodrome control services); e. Airspace management — a planning function with the primary objective of maximising the utilisation of available airspace by dynamic time-sharing and, at times, the segregation of airspace among various categories of airspace users on the basis of short-term needs; f. Air Traffic Flow and Capacity Management — the air traffic flow management is a function established with the objective of contributing to a safe, orderly and expeditious flow of air traffic by ensuring that ATC capacity is utilised to the maximum extent possible, and that the traffic volume is compatible with the capacities declared by the appropriate air traffic service providers; and g. Information Service — a service established within the defined area of coverage responsible for the provision of aeronautical information and data necessary for the safety, regularity and efficiency of air navigation.

3. Criterion ‘Operational function’ — the selected option for the criterion ‘Service/Function provided’ should be considered when selecting the option for the criterion ‘Operational function’. The following options should be available: a. For Communication services: Page 38 of 64 Annex to ED Decision 2014/035/R — Air/Ground communication — two-way communication between aircraft and stations or locations on the surface of the earth; and — Ground/Ground communication — two-way communication between stations or locations on the surface of the earth.

b. For Navigation service: — Navigation function.

c. For Surveillance service: — Air surveillance — those facilities and services used to determine the respective positions of aircraft in the air to ensure safe separation; — Ground surveillance — those facilities and services used to determine the respective positions of aircraft on the ground to allow the detection of conflicts; and — Surface movement guidance and control — a function providing routing, guidance and surveillance for the control of aircraft and vehicles in order to maintain the declared surface movement rate under all weather conditions within the aerodrome visibility operational level while maintaining the required level of safety.

d. For Air Traffic Services: — Flight plan information — specified information provided to air traffic service units, relative to an intended flight or portion of a flight of an aircraft; — Flight information and alert — provision of Flight Information (e.g. last position) in support to alerting services; — Operations room management capability — the functions which enable to combine/split sectors, assign roles on controllers working position; — Decision-making support tools — such as medium-term conflict detection, arrival/departure manager, collaborative decision-making; and — Safety nets — a (ground-based) safety net is a functionality within the ATM system that is assigned by the ANSP with the sole purpose of monitoring the environment of operations in order to provide timely alerts of an increased risk to flight safety which may include resolution advice.

e. For Airspace management: — Real-time airspace environment — the display on the executive air traffic controller Controllers Working Position of the entire airspace configuration at a given time (e.g.

restricted/danger areas).

f. For Air Traffic Flow and Capacity management: — Tactical & Real Time — the function that provides traffic prediction, flow monitoring and warning.

g. For Support Information Services: — Aeronautical Information — provision of aeronautical information and data necessary for the safety, regularity and efficiency of air navigation; and Page 39 of 64 Annex to ED Decision 2014/035/R — Meteorological Information — meteorological report, analysis, forecast and any other statement relating to existing or expected meteorological conditions.

4. Criterion ‘ t ype of failure’ — the following options should be available for the ‘ t ype of failure’ criterion: a. Total loss of service/function — the service/function is not available to the controller or pilot; b. Partial loss of service/function — not all of the service/function is available to ATC or pilot (e.g.

loss of one or several sub-functions); c. Redundancy reduction — loss of a technical backup. There are fewer technical ways to provide the service/function; d. Undetected corruption of service/function — data presented is incorrect but is not detected and used as being correct — if the corruption is detected, it means the function will have to be removed totally (total loss of function) or partially (partial loss of function); e. Loss of supervision — unable to control or monitor the function. If this means that the main function has to be removed, then this would be a total loss; and f. Corruption of supervision — undetected corruption of supervision. It has no impact unless a second action takes place. If left alone, there will be no impact. If an operator does something in response to an incorrect indication, then a different type of failure could occur.

5. Criterion ‘ e xtension’ — the physical extension of the failure should be categorised as one of the following options: a. Controller Working Position — one Controller Working Position (CWP); b. Sector suite — a set of CWPs which work together to control (a) sector(s); c. Multiple suites — self-explanatory; d. Unit — as applicable, the entire ACC/UAC/APP operations room, the whole tower, etc.

6. Criterion ‘ s cope’ — the operational scope of the effect should be classified as one of the following options: a. One — one frequency, one aircraft as applicable; b. Some — as applicable more than one frequency, more than one a/c, etc., and less than all; and c. All — all frequencies, all aircraft as applicable.

7. Criterion ‘ d uration’ — T1 is the time interval between the initiation of the technical event and the moment when it triggers actual or potential operational consequences either for the air traffic controller (ATCO) or the pilot.

a. Duration less than T1 — this option should be chosen when the technical failure did not last long enough to trigger actual or potential operational consequences on the air traffic controller or the pilot. In such a case, the severity of the ATM-specific occurrence should have no impact on the safe provision of air traffic services and should be classified with severity E.

Consequently, there is no need for the user to further apply the RAT methodology for this technical failure (just record the severity E); Page 40 of 64 Annex to ED Decision 2014/035/R b. Duration greater than or equal to T1 — this option should be selected when the technical failure lasted longer than or equally to T1 and triggered actual or potential operational consequences on the air traffic controller or the pilot.

C. Severity The severity of ATM-specific occurrences should be classified as follows: 1. AA — Total inability to provide safe ATM services (equivalent to ‘ s erious incident’) — an occurrence associated with the total inability to provide any degree of ATM services, where: a. there is a sudden and non-managed total loss of ATM service or situation awareness; and b. there is a totally corrupted ATM service or corrupted information provided to ATS personnel.

2. A — Serious inability to provide safe ATM services (also equivalent to ‘ s erious incident’) — an occurrence associated with a nearly total and sudden inability to provide any degree of ATM services in compliance with the applicable safety requirements. It involves circumstances indicating that the ability to provide ATM services is severely compromised and has the potential to impact many aircraft safe operations over a significant period of time.

3. B — Partial inability to provide safe ATM services (equivalent to ‘ m ajor incident’) — an occurrence associated with the sudden and partial inability to provide ATM services in compliance with the applicable safety requirements.

4. C — Ability to provide safe but degraded ATM services (equivalent to ‘ s ignificant incident’) — an occurrence involving circumstances indicating that a total, serious or partial inability to provide safe and non-degraded ATM services could have occurred if the risk had not been managed/controlled by ATS personnel within safety requirements, even if this implied limitations in the provision of ATM services.

5. D — Not determined — insufficient information was available to determine the risk involved or inconclusive or conflicting evidence precluded such determination.

6. E — No effect on ATM services — occurrences which have no effect on the ability to provide safe and non- degraded ATM services (equivalent to ‘ no safet y effect’).

The severity on an ATM-specific occurrence should be established, based on the combination of options chosen for each criterion.

GM11 SKPI Severity classification based on the risk analysis tool methodology — Methodology for ATM-specific occurrences A. Examples of some criteria for evaluating ATM-specific occurrences Criterion ‘ t ype of failure’ The following figure illustrates total loss and redundancy reduction in air – ground communication.

Page 41 of 64 Annex to ED Decision 2014/035/R Figure 4 — Total loss and redundancy reduction in air – ground communication Criterion ‘ e xtension’ The figure bellow illustrates an ATC unit with several sector suites, each of which consists of 3 Controllers Working Positions (CWP): UNIT SECTOR SUITE SECTOR SUITE CWP CWP CWP CWP CWP CWP SECTOR SUITE SECTOR SUITE CWP CWP CWP CWP CWP CWP Figure 5 — ATC unit, sectors and suites Page 42 of 64 Annex to ED Decision 2014/035/R Criterion ‘ s cope’ The table below gives an indication of what ‘ one ’ / ’ some ’ / ’ all ’ represents for different operational functions (criterion ‘ s cope’).

Scope (how many … were Services Operational functions impacted) Communication Air/Ground Communication Communication(s) ATCO/Pilot Ground/Ground Communication Communication Communication(s) ATCO/ATCO Navigation Navigation Pilot(s) Surveillance Air Surveillance Displayed Radar Track(s) Surveillance Ground Surveillance Displayed Radar Track(s) Surface Movement Surveillance Guidanc e & Control Aircraft(s)/Vehicle(s) Air Traffic Services Flight Plan Information Flight Plan(s) Air Traffic Services Flight Information & Alert Flight(s) N/A (extension should be Air Traffic Services Ops Room Management sufficient) Air Traffic Services Decision Making Support Fight(s) Air Traffic Services Safety Nets Conflict(s) Real Time Airspace Air Traffic Services Environment Route(s), Area(s), … Air Traffic Flow Capacity Management Tactical & Real Time Flight(s) Information Services Aeronautica l Information Information Type(s) Information Services Meteorological Information Information Type(s) Criterion ‘ d uration’ When criterion ‘ d uration’ is evaluated, T1 should be used for separating technical glitches with no operational consequences from failures that impact the ANSP ’s ability to provide safe ATM services.

Some of the values of T1 may be predefined, for example when they are part of the Service Level Agreement (SLA) between the technical and operational units (departments) or when they are part of the ATS unit safety case. When the value of T1 is predefined by the ANSP, it should be done based on inputs provided by the ATCOs and/or pilots. Alternatively, if a T1 is not predefined at the moment of the investigation, the evaluation of the ‘duration’ criterion may be done by determining if a particular occurrence/failure triggered actual or potential operational consequences (the criterion should be scored greater than or equal to T1).

This value cannot be established at European level as it is dependent on the functionalities of the ATM provider’s system architecture, airspace complexity, traffic load and concept of operations. When choosing the option ‘less than T1’ or ‘greater than or equal to T1’ , there is no need to know the exact duration of the event but whether it has a potential or real operational impact, i.e. is greater, or not, than the T1 value established locally.

Typical examples of operational impact where ‘ d uration’ is greater than or equal to T1: — ATC/pilot had to do something different; — ATC/pilot is presented with incorrect, reduced or no information; — Workload increase; Page 43 of 64 Annex to ED Decision 2014/035/R — Capacity reduction; — Reduced ability to provide safe services; and — ATCO can no longer cope with the situation.

The charts below illustrate the ATM system both in a steady state and failure modes, in order to ease the understanding of the role of T1.

— Steady state of the technical system (no failure) The chart below illustrates a steady state where the ATM system delivers all operational functions as expected.

— ATM-specific technical event with a potential or real operational impact The chart below provides the occurrence timeline in case of a total failure of an operational function.

In the given example, the failure has an operational impact on the ability to provide ATM services (this could be the case in a total failure of the air – ground communication function or in a total failure of surveillance function; see examples 1 and 3 below).

Page 44 of 64 Annex to ED Decision 2014/035/R T0 ATM-specific technical event commences.

T0 to T1 ATM-specific technical event has no operational impact as the ATC maintain desired traffic level.

T1 ATM-specific technical event triggers operational consequences on ATC controller or pilot.

T1 to T2 Potential safety impact on ATC or pilot.

T3 The ATM-specific technical event finishes.

T1 to T4 Business effect on ATC or Pilot, e.g. regulations applied.

T4 ATC returns to the desired traffic levels.

— Redundancy reduction The chart below illustrates the occurrence timeline in the case of a redundancy reduction with no operational impact (duration is less than T1). This case could be applied in Example 2 from section C, the failure on day D.

T3 T3 T0 Time Page 45 of 64 Annex to ED Decision 2014/035/R T0 ATM-specific technical event commences.

T1 Does not take place.

T2 Does not take place.

T0 to T3 ATM-specific technical event has no impact. ATC maintain desired traffic level.

T3 ATM-specific technical event finishes.

T4 Does not take place.

B. Look-up table Following the selection of criteria options described in this AMC9 SKPI, the severity for an ATM-specific occurrence may be determined by identifying the appropriate combination in the look-up table presented in Appendix 1 to GM11 SKPI — Look-up table for severity classification of ATM-specific occurrences and by retrieving the predetermined severity in column ‘Severity’.

The look-up table contains all the realistic combination of the criteria described in this GM. An occurrence code is uniquely assigned to each combination.

It is to be noted that in case of combination of criteria that are not realistic, the severity is marked ‘X’ in the look-up table. In such case, the severity cannot be determined (category D). Therefore, the user should try to map a given failure to the credible combination available in the look-up table.

A severity is predefined for each of the identified realistic combinations. A sample of a section of this look- up table is given below: Code Service Affected Services Operational functions Type of Failure Extension Scope Duration T1 Severity AR-AGC/000 Area control services Communication Air/Ground Communication Undetected Corruption of function Unit All > T1 AA AR-AGC/001 Area control services Communication Air/Ground Communication Undetected Corruption of function Unit Some > T1 AA AR-AGC/002 Area control services Communication Air/Ground Communication Undetected Corruption of function Unit One > T1 A AR-AGC/010 Area control services Communication Air/Ground Communication Undetected Corruption of function Multiple Suites All > T1 AA AR-AGC/011 Area control services Communication Air/Ground Communication Undetected Corruption of function Multiple Suites Some > T1 A AR-AGC/012 Area control services Communication Air/Ground Communication Undetected Corruption of function Multiple Suites One > T1 A AR-AGC/020 Area control services Communication Air/Ground Communication Undetected Corruption of function Sector Suite All > T1 X AR-AGC/021 Area control services Communication Air/Ground Communication Undetected Corruption of function Sector Suite Some > T1 X AR-AGC/022 Area control services Communication Air/Ground Communication Undetected Corruption of function Sector Suite One > T1 B AR-AGC/030 Area control services Communication Air/Ground Communication Undetected Corruption of function CWP All > T1 X AR-AGC/031 Area control services Communication Air/Ground Communication Undetected Corruption of function CWP Some > T1 B AR-AGC/032 Area control services Communication Air/Ground Communication Undetected Corruption of function CWP One > T1 B AR-AGC/100 Area control services Communication Air/Ground Communication Total Loss of function Unit All > T1 AA AR-AGC/101 Area control services Communication Air/Ground Communication Total Loss of function Unit Some > T1 AA AR-AGC/102 Area control services Communication Air/Ground Communication Total Loss of function Unit One > T1 A AR-AGC/110 Area control services Communication Air/Ground Communication Total Loss of function Multiple Suites All > T1 AA AR-AGC/111 Area control services Communication Air/Ground Communication Total Loss of function Multiple Suites Some > T1 A AR-AGC/112 Area control services Communication Air/Ground Communication Total Loss of function Multiple Suites One > T1 A AR-AGC/120 Area control services Communication Air/Ground Communication Total Loss of function Sector Suite All > T1 A AR-AGC/121 Area control services Communication Air/Ground Communication Total Loss of function Sector Suite Some > T1 A AR-AGC/122 Area control services Communication Air/Ground Communication Total Loss of function Sector Suite One > T1 A AR-AGC/130 Area control services Communication Air/Ground Communication Total Loss of function CWP All > T1 B AR-AGC/131 Area control services Communication Air/Ground Communication Total Loss of function CWP Some > T1 B AR-AGC/132 Area control services Communication Air/Ground Communication Total Loss of function CWP One > T1 B AR-AGC/200 Area control services Communication Air/Ground Communication Partial Loss of function Unit All > T1 C AR-AGC/201 Area control services Communication Air/Ground Communication Partial Loss of function Unit Some > T1 C AR-AGC/202 Area control services Communication Air/Ground Communication Partial Loss of function Unit One > T1 C Figure 6 — Extract of look-up table in Appendix 1 to GM11 SKPI Page 46 of 64 Annex to ED Decision 2014/035/R C. Examples for ATM-specific occurrences Example 1 All communications with aircraft were lost in the sector S outh of ACC X. The failure lasted 1 min 12 sec.

The service provided was ‘ c ommunication’. As the communication with the aircraft was lost, the operational function affected is ‘ air – ground c ommunication’.

No communication with the aircraft in the sector was possible during that time; therefore, the type of failure is ‘ t otal loss of function’. Service affected is ‘Area Control Centre’. The sector South of the ACC was the only ACC sector affected by the failure. As such, the extension is ‘ sector s uite’. In this case , the communication with all aircraft in the sector was lost and, therefore, the scope is ‘All’.

In ACC X, T1 is predefined for total loss of air – ground communication function as being T1 = 20 seconds.

As the total duration of failure is 1 min 12 sec, the duration is higher than T1 and ,therefore, the RAT look- up table may be used.

For these selected options, the corresponding combination in the look-up table is: Code Service Services Operatio - Type Extension Scope Duration T1 Severity Affected nal of functions failure Air/ Total AR- Area Ground loss of AGC/ control Commu- commu- funct- Sector 120 services nication nication ion suite All > T1 ~20s A Therefore, the severity for the failure in Example 1 is ‘A — Serious inability to provide safe ATM services’.

Example 2 Due to telecom failure, there is loss of redundancy of some frequencies affecting several sectors in APP Z.

There were two such occurrences at APP Z: one on day D which lasted 5 minutes and the other on day D+2 which lasted two hours.

The service provided was ‘ c ommunication’. As the redundancy is for radio communication with the aircraft, the operational function affected is ‘ air – ground c ommunication’.

The type of failure is ‘ r edundancy reduction’ and affects several sectors and several frequencies; therefore, the extension is ‘ multiple s uites’ and the scope ‘ s ome’.

In APP Z, the local procedure requires that in case of loss of back-up frequencies (i.e. redundancies), capacity limitations are put in place after 30 minutes, which is our T1.

Therefore, duration of the failure on da y D is less than T1 and the severity is directly classified as ‘E — No effect on ATM services’ and there is no need to use the look -up table.

For the failure on day D+2, the duration is greater than or equal to T1 and, therefore, the look-up table might be used and the corresponding combination is: Code Service Services Operatio - Type of Extension Scope Duration T1 Severity Affected nal failure Page 47 of 64 Annex to ED Decision 2014/035/R functions Air/ AR - Area Ground Total loss AGC/ control Commu - commu - of Sector 120 services nication nication function suite All > T1 ~20s A Therefore, the s everity for the failure in Example 2 on day D+2 is ‘C — Ability to provide safe but degraded ATM services’.

Example 3 Total failure of the radar data processing system (normal and back-up) in an ACC (duration 2 minutes).

Service affected = Area control services The service is ‘ s urveillance’ and the operational function is ‘ air surveillance in the a rea control services’. It is a total loss of function which extends to the whole unit and affects all targets.

For the combination above, T1 is set to ~ 40s, therefore, Duration is > T1 and, therefore, the look-up table might be used and the corresponding combination is: Code Service Services Operationa Type of Extension Scope Duration T1 Se verity affected l functions failure Approach Air/Ground AP- control Communi- communi- Redundancy Multiple 1800 AGC/311 services cation cation reduction suites Some > T1 s C Therefore, the s everity for the failure in Example 3 is ‘A — Serious inability to provide safe ATM services’.

AMC8 SKPI RAT methodology — Monitoring mechanism The Member States’ points of contact established in accordance with Directive 2003/42/EC and Regulation (EC) No 1330/2007 , should collect verified information regarding the application of severity classification using the Risk Analysis Tool (RAT) methodology for the reported occurrences within the scope of the performance scheme Regulation.

When the Member States report on the monitoring of the performance plans and targets in accordance with the performance scheme Regulation, they should report the percentage of occurrences that been evaluated by the use of the severity classification using the RAT methodology.

For the application of the severity classification on an individual basis for all occurrences within the scope of the regulation, Member States should provide the data by making use of existing safety data reporting mechanisms, that is, either the European Central Repository and/or the Annual Summary Template (AST) mechanism, with enhancements where needed.

Directive 2003/42/EC of the European Parliament and of the Council of 13 June 2003 on occurrence reporting in civil aviation (OJ L 167, 4.7.2003, p. 23).

Commission Regulation (EC) No 1330/2007 of 24 September 2007 laying down implementing rules for the dissemination to interested parties of information on civil aviation occurrences referred to in Article 7(2) of Directive 2003/42/EC of the European Parliament and of the Council (OJ L 295, 14.11.2007, p. 7).

Page 48 of 64

Annex to ED Decision 2014/035/R

IV Just culture

GM12 SKPI Just culture — General The just culture KPI aims at measuring the level of presence and corresponding level of absence of just culture at State and at ANSP level. The metrics have been constructed to respond to the criteria of: clearly defined, auditable, verifiable, repeatable and indicative of the level of just culture being implemented. The just culture KPI consists of metrics in the areas of policy and its implementation, legal/judiciary and occurrence reporting and investigation.

The main aim of the indicator and of the questionnaires is to identify possible obstacles and impediments to the application of the just culture (JC).

Reference is made to the ‘State level’ instead of ‘NSA level’ because, although a large number of questions refer to the existing situation in the national authority, a limited number of others deal with elements which go beyond the field of competence of the authority and may have to be addressed at the level of other State entities.

The questionnaires identify several elements related to an effective just culture, each element in turn with a number of sub-elements. These sub-elements are binary, i.e. the answer can only be ‘yes’ or ‘no’. The States and ANSPs may qualify the ‘no’ answers in their respective completed questionnaire (column ‘Justification and remarks’) by indicating the related obstacles.

A positive reply gives an indication of a just culture context, while a negative reply indicates a potential deficit/obstacle in the just culture implementation. However, it is not expected that all replies should be positive but the identification of negative elements would give indication of possible areas of improvement and could be considered as incentives for improving the just culture in a particular State/organisation. The State/ANSP may be asked to provide evidence for justification of the answers supported by written documents such as arrangements, procedures, correspondence or other documents.

AMC9 SKPI Just culture — Reporting at State level A. Reporting The just culture indicator should be reported by verified responses to a questionnaire at State level. The questionnaire which should be answered by the Member State/competent authority is indicated in Appendix 1 to AMC9 SKPI — Just culture Questionnaire — State level (questions P.1 to P.9, L.1 to L.7, and O.1 to O.2). The questions should be answered with ‘yes’ or ‘no’. For ea ch question, the State should provide information and evidence to justify the answers and may add any applicable explanatory remarks.

B. Verification Questionnaires should be dispatched together with those for the EoSM indicator following the same verification process.

The verification mechanism for JC measurement should be the same as in AMC2 SKPI, section C.

GM13 SKPI Just culture — Reporting and verification at State level Some examples of the possible justification material which support the verification of a completed JC questionnaire at State level are provided in Appendix 1 to GM13 SKPI — Just culture — State level — Page 49 of 64 Annex to ED Decision 2014/035/R possible evidence. This Appendix consists of the State-level JC questions with an additional column providing possible evidence and some explanatory notes where considered necessary.

In addition to the filled-in questionnaire, the State may report on the just culture indicator using the following format, including an indication of possible areas for improvement.

No of questions answered with: Yes No Policy and its implementation Legal/Judiciary Occurrence reporting and investigation Identification of possible areas of improvement Policy and its implementation  …  … Legal/Judiciary  …  … Occurrence reporting and investigation  …  … AMC10 SKPI Just culture — Reporting at ANSP level A. Reporting The just culture indicator should be reported by verified responses to a questionnaire at ANSP level. The questionnaire which should be answered by the ANSPs is indicated in Appendix 1 to AMC10 SKPI — Just culture Questionnaire — ANSP level (questions P.1 to P.13, L.1 to L.3, and O.1 to O.8). The questions should be answered with ‘yes’ or ‘no’.

For each question, the ANSP should provide the NSA with information and evidence to justify its answers and may add any applicable explanatory remarks.

B. Verification Questionnaires should be dispatched together with those for the EoSM indicator following the same verification process.

The verification mechanism for JC measurement at ANSP level should be the same as in AMC3 SKPI, section D.

GM14 SKPI Just culture — Reporting and verification at ANSP level Some examples of the possible justification material which support the verification of a completed JC questionnaire at ANSP level are provided in Appendix 1 to GM14 SKPI — Just culture — ANSP level — possible evidence. This appendix consists of the ANSP level JC questions with an additional column providing possible evidence and some explanatory notes where considered necessary.

In addition to the filled-in questionnaire, the ANSP may report on the just culture indicator using the following presentation format, including a self-assessment of possible areas for improvement.

Page 50 of 64 Annex to ED Decision 2014/035/R No of questions answered with: Yes No Policy and its implementation Legal/Judiciary Occurrence reporting and investigation Identification of possible areas of improvement Policy and its implementation  …  … Legal/Judiciary  …  … Occurrence reporting and investigation  …  … GM15 SKPI Interdependencies — evaluation of the impact on safety of the performance plan Purpose The purpose of this guidance material is to describe a possible process to be applied when describing consideration of the interdependencies between key performance areas in the performance plan, including an evaluation of the impact on safety in the performance plan when complying with the performance scheme Regulation.

Description of a possible process to be applied when identifying interdependencies and impact on safety The ATM performance plan includes identifying interdependencies between cost, environment, capacity and safety. The competent authority should be considered as an integral part of the interdependencies because of the competent authorities’ responsibilities in relation to certification and oversight. Planned actions to achieve the targets in the performance areas of environment, capacity and cost-efficiency most likely will bring changes in the functional systems, as defined in Commission Implementing Regulation (EU) 1035/2011 (common requirements Regulation), of the ANS providers and their competent authorities (NSAs).

The performance scheme Regulation establishes provisions for an evaluation of the impact on safety of the performance plan. This is valid for all entities which contribute to the performance plans, including the competent authorities (NSAs).

All entities contributing to the improvement of the performance at local level should make an analysis of impact on their functional systems by the changes which will be introduced by the improvements in the other performance areas foreseen to be implemented within the reference period. Assessment of the identified changes to the functional systems should be done at the time of performance planning and the relevant possible mitigating actions should be identified. Description of the changes with potential effect Article 2(3) of Commission Implementing Regulation (EU) 1035/2011 — ‘’functional system’ means a combination of systems, procedures and human resources organised to perform a function within the context of ATM. ’ Article 11, 3.(e) and Annex II, 3.3 of Commission Implementing Regulation (EU) No 390/2013.

Page 51 of 64 Annex to ED Decision 2014/035/R on safety and the mitigations identified should be included in the interdependencies analyses of the performance plan.

In instances where changes to functional systems are scheduled for medium to long-term future implementation, safety mitigations for safety assurance should be included in the performance plan as far as practicable. If the assessment of planned changes (e.g. by using Safety scanning) shows no effect on safety, they should be referenced in the interdependencies analyses of the performance plan as having no safety impact. However, the Member States may also include a high-level description of some changes in the other performance areas which will not affect their functional systems. The process for the assessment of changes and their insertion in the performance plan are provided in the diagram (Figure 7).

When describing the consideration of the interdependencies between safety performance area and the rest of the performance areas in the performance plan, Member States should, at a minimum, include in the performance plan: — Performance area and the target which achievement will introduce the change to the functional system; — functional systems affected; and — description of:  affected elements of the functional system and the changes introduced in each of them;  planned mitigations and activities for safety assurance and other relevant information.

Page 52 of 64 Annex to ED Decision 2014/035/R Planned changes in performance areas/indicators other than safety environment capacity cost - efficiency Analysis of impact on the functional systems No Change to Functional System with Referenced in the effect on performance plan safety?

as having no safety impact Yes Provide Make description in assumptions on performance plan mitigation for item 3.3 safety assurance Figure 7: Interdependencies evaluation Page 53 of 64 Annex to ED Decision 2014/035/R Examples of changes that may have an effect on safety and how the relevant interdependencies may be described in the performance plan item 3.3 Examples of changes for ANS providers driven by improvement in performance areas which have effect on safety Performance Functional Potential changes to the elements of functional Remarks area/reason for system system and possible mitigation measures change affected/ Change description Cost - efficiency - ANSP xxx, ACCs Human resources Reduction in operational The change is planned f or the beginning of 2019 and will support driven change yyy, zzz etc. personnel; achieving the cost - efficiency target by reducing the unit rate by (reduce cost for 1.06 %. In order for the ATCO to take over the role of the Removal of ATCO additional training for personnel) assistant then , it is likely that the information used by the assistant new role; assistant will have to be p resented to the ATCO. Moreover, in position (tasks Training for technical order to avoid overload , the information used by the assistant go to ATCO personnel .

and the information used by the ATCO will have to be presented and/or Procedures Change to operational and in a different, more user - friendly, form. It may also be necessary automation) maintenance procedures.

to provide additional automa tion to perform some assistant’s tasks. This certainly implies changes to the equipment at the Systems Change to operator interface ATCO’s working position and very probably implies changes to and likely change to functions the functions providing information to those working positions.

for the manipulation and visibility of surveillance and flight data information and management; Possibly , the addition of new flight lists in CWP of planning and executive controllers .

Architecture Removal of assistant position and likely changes to the way information is managed and distributed within the system ; Redistribution of function/responsibility Page 54 of 64 Annex to ED Decision 2014/035/R between human – automation .

Environment Possible change to sector shape/organisation to limit ATCO workload .

Capacity - driven ANSP A and B Human resources Training for new procedures, The change is a deliberate attempt by the provider of ATS to change — airspace organisation and increase the capacity as indicated in the performance plan from Change the increase in equipment; 2017. Daily and seasonal fluctuations in traffic are not considered organisation of traffic in to be a change.

the upper Possible increase in personnel; airspace airspace and The change is actually a change in the environ ment of operation Working hours/shift patterns introduction of that would require a change to the functional system in order to (fatigue and the associated new technology make the operation acceptably safe.

increased risk of human errors) Changes are required to the surveillance or communications systems already present . T he changes may involve the operational use of new or modified information that is already Procedures New or changed procedures within the current system. Such use could involve an (including contingency architectural change to make the information available to the measures) to handle new changed components.

services and increased traffic; Changes to the ANSP organisation for delivering services Page 55 of 64 Annex to ED Decision 2014/035/R Possibly improved surveillance, communications System/ and/or other systems , e.g.

constituents ATCO decision support tools; Changes to the display of operational data to controllers at the point of service delivery; Changes to communications systems (architecture , etc.)

used for the delivery of an ATS service Architecture Possibly , if the surveillance and communications system changes require changes in the interfaces with equipment already present .

Environment Increase in traffic; Airspace change .

Page 56 of 64

Annex to ED Decision 2014/035/R V Safety Performance Indicators AMC11 SPI The application by the air navigation service providers of automated safety data recording systems where available, which shall include, as a minimum monitoring of separation minima infringements and runway incursions The application by the air navigation service providers of automated safety data recording systems used for monitoring and recording of separation minima infringements and runway incursions should be reported under this safety performance indicator, where the system has, as a minimum, the following basic functional capabilities: — Interface with ATC operational systems for detection of candidate events; — Filter (automatic and manual) to extract only relevant events, based on predetermined technical and operational criteria; and — Recording of retained events in a local database for further analysis and reporting.

Those functions are captured in Figure 8 below and detailed in the associated guidance material.

Surveillance data 1. Interface: Flight data Get data from ATS system Detect events Candidate events 2. Filter: Select genuine occurrences Discard non-relevant events Reportable events 3. Record: Database population Figure 8 — Automated monitoring of separation minima infringements and runway incursions ANSPs should report to their competent authorities the beginning of the application period and subsequently, on an annual basis, the application of automatic safety data recording systems at their individual ATS units. If such systems are in use, the ANSPs should report, as a minimum, the following data: — the unit at which the system is used (which ACC, APP, TWR, etc.); — the type of reportable events recorded and the associated definitions for each type, which should include, but may not be limited to, the minimum required by the performance scheme Regulation; and — the number of reportable events recorded in the local databases by type (SMI, RI).

The competent authorities should: Page 57 of 64 Annex to ED Decision 2014/035/R — collect the reports on the application of automatic safety data recording systems submitted by the ANSPs; — review the data contained in the above-mentioned reports from the ANSPs; and — provide the information to the Agency and the PRB for this safety performance indicator for the preceding year by the end of May each year.

GM16 SPI The application by the air navigation service providers of automated safety data recording systems where available, which shall include, as a minimum monitoring of separation minima infringements and runway incursions. Automated safety data recording systems for monitoring of separation minima infringements (SMIs) General The automated safety data recording systems should be used in a just culture environment to improve the information and analysis used by the organisation’s SMS.

It should be recognised, where appropriate, that for various reasons (e.g. the automated system failed to capture some occurrences which were reported by other reporting mechanisms), the number of occurrences captured and reported against this performance indicator did not necessarily coincide with the number manually reported in the SPI, namely ‘the number of, as a minimum, separatio n minima infringements, runway incursions, airspace infringements, and ATM- specific occurrences at all air traffic services units’.

The systems should operate to detect candidate SMIs in the ANSP’s designated airspace. To ensure the systems focus on SMIs, the system should be configured to filter out events which can be attributed to standard operating practice. The remaining occurrences, after filtering, should be considered as genuine and should be reported under this PI.

Recorded data in one ATS unit is not comparable and should not be used for benchmarking with that of another ATS unit.

General description of automated safety data recording systems In order for such systems to detect reportable occurrences, there are several functionally distinct processes as defined in AMC11 SPI.

1. Interface with the ATS operational systems for detection of candidate events The automated safety data recording system should interface with ATS operational systems (surveillance, flight data processing, etc.). According to the implemented algorithm, this system should detect candidate SMIs in the airspace concerned.

2. Filter for genuine SMIs The filtering process should discard those events that are not considered genuine.

All candidate SMIs should be processed to determine if they fall within a predetermined set of operational rules and procedures permitting identification of true SMIs. This should include both automated and manual filtering and should discard spurious events such as bad plots/tracks or not relevant (i.e. operationally correct).

Automatic filtering should be focussed on risk-bearing events and should automatically limit recorded events.

Parameters for SMI detection of the airspace concerned may be eroded by certain values such as the vertical and/or lateral dimensions of the minima (e.g. the error or the resolution capabilities of the surveillance system implemented).

Page 58 of 64 Annex to ED Decision 2014/035/R Manual filtering should further discard the automatically recorded events that are not considered genuine.

Automatically detected events that are due to normal operating practice should also be filtered out. Normal operating practice may include events such as aircraft encounters in the vicinity of an airport which may not be subject to standard separation, encounters with military aircraft or aircraft employing VFR operations.

These operational practices and procedures should be predefined.

3. Recording SMIs The reportable occurrences for both automatic and manual filtering should be recorded in a database. For the purpose of this performance indicator, the database should be capable of providing, as a minimum, a list of recorded events for a specified period of time and the related data extracted from the system interfaces.

GM17 SPI The application by the air navigation service providers of automated safety data recording systems where available, which shall include, as a minimum monitoring of separation minima infringements and runway incursions. Automated safety data recording systems for monitoring of runway incursions (RIs) GENERAL The automatic detection and monitoring of runway incursions is a complex technical task. The reason for this is that both the presence and contents of an ATC clearance are essential factors in determining whether an event can be classified as a runway incursion and these are typically not available in an electronic format.

Article 2(14) of the performance scheme Regulation transposes the ICAO definition of runway incursion.

Some typical situations for an RI may be that: — the aircraft lands/takes off without clearance; — the controller incorrectly clears an aircraft to land or take-off; — the aircraft, vehicle or person enters the runway at the incorrect holding point; or — the aircraft lines-up out of instructed sequence.

What makes automated detection of an RI even more complex is the fact that different operations and the relevant interpretation of ‘incorrect presence’ may lea d to cases when similar situations may be considered as an RI in one instance and as normal operation in another instance.

As a consequence, this GM is written with a view to the development of future automated systems. Similar to the systems for automated detection of SMIs, a future system that automatically detects RIs should comprise three functionally distinct processes as defined in AMC11 SPI.

The automated occurrence recording system should be used in a just culture environment to improve the inform ation used by the organisation’s SMS for the purpose of improving safety.

Recorded data in one ATS unit is typically not comparable to that of another ATS unit.

In order for systems to detect RIs, there are several functionally distinct processes as defined in AMC11 SPI.

1. Interface with the ATS operational systems for detection of candidate events The automated safety data recording systems for monitoring of RIs should interface with ATS operational systems (surveillance, flight data processing including ATC clearances, etc.). The system should analyse the position of every aircraft on the airfield relative to every other aircraft and/or vehicle in its vicinity, and ATC aircraft specific clearance information to determine the presence or not of a RI occurrence.

Page 59 of 64 Annex to ED Decision 2014/035/R This still may leave unresolved the issue of the presence of a person on the runway, which may not be detectable.

A number of events scenarios will then need to be defined and incorporated into the system to enable the detection of candidate events. It should be noted that developing the events scenarios requires careful consideration. The scenarios need to take into account the airfield layout, the type of operation taking place (ILS CAT I, II or III), the status of each aircraft (cleared to take off, cleared to cross the runway, cleared to line up, conditionally cleared to line up, cleared to land, etc.), each aircraft’s position, the status of all stops bars (when in use), and the sequence with which the clearances have been issued. This will enable the criteria for each runway incursion to be established. This is necessary because there are no consistent criteria that can be used to identify a runway incursion. They can occur with a single aircraft, vehicle or person on the runway and do not necessary occur with a simultaneous presence of aircraft, vehicle or persons on the runway.

2. Filter for genuine RIs During this step, the system should filter out genuine events that are due to normal operating practice.

Spurious and/or false targets also need to be filtered out by the system. This filtering function should be fulfilled by an automatic filtering followed by manual filtering, given the complexity of potential situations at an airport and the differences between airports in Europe. Each event should be reviewed against applicable scenarios suitable for the operations which are in accordance with the airport policy.

3. Recording RIs The reportable events for both automatic and manual filtering should be recorded in a database. For the purpose of this performance indicator, the database should be capable of providing, as a minimum, a list of recorded events for a specified period of time and the related data extracted from the interfaces.

GM18 SPI The reporting by the Member States and air navigation service providers on the level of occurrence reporting, on an annual basis, aiming at measuring the level of reporting and addressing the issue of improvement of reporting culture General The level of occurrence reporting should be defined as the proportion of the occurrences received by the ANSP or State occurrence reporting schemes, compared to all the occurrences that happened. This can hardly be evaluated since neither the ANSP nor its competent authority may be sure that all occurrences that happened are known, hence, indirect methods should be used to estimate the level of occurrence reporting. As a general principle, it should be recognised that the level of occurrence reporting may be related to a number of different variables, such as the implementation of just culture principles, ease of report submission and feedback given to reporters after investigation. Direct comparisons or benchmarking of organisations using the number of occurrence reports are particularly misleading for this reason and, therefore, should not be used.

In order to report on the level of occurrence reporting, ANSPs and Member States should prepare a written assessment of the level of occurrence reporting on an annual basis. The ANSP analysis should be submitted for review to the relevant competent authority, and Member State analysis should be submitted to the Agency for review.

At State level, the preparation of this report should take into account the safety performance indicator ‘the number of, as a minimum, separation minima infringements, runway incursions, airspace infringements and Page 60 of 64 Annex to ED Decision 2014/035/R ATM- specific occurrences’ (GM19 SPI and AMC13 SPI). Therefore, the data definition used for both performance indicators should be the same.

At both ANSP and State level, the analysis of the level of occurrence reporting should be a combination of quantitative assessment of occurrences and a qualitative assessment of the successes and limitations of reporting within the ANSP or State (as applicable). In addition, the State level analysis should include an overview of the combined ANSP analysis of the level of occurrence reporting, which should be dis-identified.

Example: Document outline for the annual assessment of the level of occurrence reporting Introduction (Qualitative Information) A brief introduction should provide basic information as to the nature of the reporting scheme, such as: — a description of the methods of collecting data and the ways in which reporters can submit occurrence reports; — whether voluntary reports are incorporated; and — a brief description of the functionality of the database for collection storage and analysis of safety data, system in use, how long it has been in place and who can submit reports.

Data Analysis (Quantitative Information) The overall rate of ANS occurrences, which should be broken down into categories showing the occurrence type and severity classification: — For severity classification, the results using the RAT methodology should be presented for, as a minimum, separation minima infringements, runway incursions and ATM-specific occurrences.

— Where other severity classification methodologies are in use, the results may also be presented separately or with an indication that the severity was not evaluated by using RAT methodology.

— Appropriate units of measurement should be used, wherever available, to calculate the rate. For example:  the rate of runway incursions should be calculated using the number of all IFR/VFR movements under control of the TWR unit (e.g. number of RI/number of arrivals and departures);  the rate of separation minima infringements should be calculated using the number of IFR flight hours as the flight hours may be calculated as sum of the airborne time of IFRs within the area of responsibility of the ANSP (e.g. number of SMI/number of IFR flight hours);  for the airspace infringements (AI), due to their complex definition, it is difficult to propose a proper rate. However, it is possible to divide the number of reported AI leading to a loss of separation by the total number of AI to identify the rate of infringement resulting in loss of separation. Another measure could be to divide the numbers of reported AI attributable to IFR and VFR aircraft by the total number of AI to identify the rate of infringement by IFR and VFR flights respectively. This would then allow comparison between the two and help in determining which set of aircraft was at a greater infringement risk; and  the rate of ATM-specific occurrences should be calculated using the number of operating hours of the relevant ATS unit (e.g. number of ATM-specific/number of operating hours). In case some functions (e.g. FDPS, RDPS) are serving several ATS units, the rate of ATM specific occurrences related to that function number of IFR flight hours could be considered as more suitable.

Page 61 of 64 Annex to ED Decision 2014/035/R — A comparison of the number of high-severity occurrences and low-severity occurrences should be made, since logically in a system with a high level of reporting there should be many times more low-severity occurrences than high-severity occurrences. Low-severity occurrences are defined as severities C and E, high severity occurrences are defined as severities A and B for SMI, RI and airspace infringement, and AA, A and B for ATM specific occurrences.

— The variation in the reporting rate between the major reporters should be measured. For example, the different ANSPs reporting to a State scheme or the different units or sectors within an ANSP. The information should be dis-identified since it is the variation that is of note, not the rates themselves.

Conclusions: Assessment of the Level of occurrence reporting A brief summary of the main conclusions should be provided, including the limitations of the data and the perceived impact of variables applicable to the ANSP/State on the results presented.

Using the data analysis results and any gaps in reporting that were identified in the qualitative information, an assessment should be provided of the level of occurrence reporting, as well as a list of actions that should be initiated to improve reporting. At State level, these actions should be generally valid for all ANSPs under the CA authority and at ANSP level, the actions should be specific taking into account size of the ANSP, services provided, etc. The list of actions provided should include those recently completed, those that are underway and new actions. Timescales for the initiation and completion of the action should be included.

AMC12 SPI The reporting by Member States and air navigation service providers on the level of occurrence reporting, on an annual basis, aiming at measuring the level of reporting and addressing the issue of improvement of reporting culture States and ANSPs should prepare a quantitative and qualitative assessment of the level of occurrence reporting, on an annual basis. The scope of the assessment should be the same as that used for performance indicator ‘the number of, as a minimum, separation minima infringements, runway incursions, airspace infringements and ATM- specific occurrences’. The assessment should contain, as a minimum: — an estimate of the level of occurrence reporting, including both quantitative and qualitative analysis. At State level, this should include an aggregated, qualitative description of the level of occurrence reporting by their ANSPs; and — details of actions identified to improve reporting culture, including actions that have been completed, those that are underway and newly identified actions.

ANSPs should agree with their State the deadline for submitting their report.

States should combine the preparation of this report with the process of validating the performance indicator ‘the number of, as a minimum, separation minima infringements, runway incursions, airspace infringements, and ATM- specific occurrences at all air traffic services units’, ensuring that the final report is su bmitted by the end of May.

GM19 SPI Process for submitting the number of, as a minimum, separation minima infringements, runway incursions, airspace infringements, and ATM-specific occurrences at all air traffic services units The purpose of this GM is to explain the process by which the number of occurrences will be measured, including as a minimum, separation minima infringements, runway incursions, airspace infringements, and ATM-specific occurrences at all air traffic services units.

Page 62 of 64 Annex to ED Decision 2014/035/R It is anticipated that Member States, either directly or through their ANSPs, will submit occurrence reports of separation minima infringements, runway incursions, airspace infringements and ATM-specific occurrences via existing reporting mechanisms, that is the AST mechanism or the European Central Repository (ECR). Hence, the Agency and the PRB will have the data available in order to be able to evaluate the safety performance indicator ‘ the number of, as a minimum, separation minima infringements, runway incursions, airspace infringements, and ATM- specific occurrences at all air traffic services units.’ States should anticipate that they will receive an analysis report sent by the Agency /PRB of the data submitted by them, by the end of April each year, containing the number of applicable occurrences in their State per the previous year with the following scope: — only occurrences within the territory of a State or its airspace; — only occurrences applicable to the performance scheme Regulation; and — the type of occurrence (as minimum, separation minima infringement, runway incursion, airspace infringement, ATM-specific occurrence).

The number of occurrences for the State will be shown both in total and broken down by type of occurrence.

Observations will also be included regarding the quality of the data that the State submitted.

States should, therefore, be prepared to receive this analysis report, confirm the numbers presented in the report and respond to the observations. To confirm the numbers presented in the report, States may limit this confirmation to a ‘gross error check’ instead of re -calculating the numbers themselves. Where data has been submitted, which is preliminary and subject to change, States should retain a record of the preliminary data in order to perform this gross error check.

AMC13 SPI The number of, as a minimum, separation minima infringements, runway incursions, airspace infringements, and ATM-specific occurrences at all air traffic services units To facilitate the implementation of this safety performance indicator, the CA of each Member State should nominate a national focal point to the Agency and the PRB.

When receiving from the Agency and the PRB an analysis report of the reported occurrences data measuring this performance indicator for the preceding year, the Member State should: — validate the numbers presented in the report and advise of any identified discrepancies; — respond to all the observations in the report; and — send a confirmation of the numbers presented and responses to the observations to the Agency by the end of May each year.

Page 63 of 64

Annex to ED Decision 2014/035/R

Appendices

The below appendices will appear as separate document to this Annex.

— Appendix 1 to AMC2 SKPI — Questionnaire for measurement of the Effectiveness of Safety Management KPI — State level — Appendix 2 to AMC2 SKPI — List of weightings for evaluation of the Effectiveness of Safety Management Questionnaire — State level — Appendix 1 to AMC3 SKPI — Questionnaire for measurement of the Effectiveness of Safety Management KPI — ANSP level — Appendix 2 to AMC3 SKPI — List of weightings for evaluation of the Effectiveness of Safety Management Questionnaire — ANSP level — Appendix 1 to AMC9 SKPI — Just culture questionnaire — State level — Appendix 1 to AMC10 SKPI — Just culture questionnaire — ANSP level — Appendix 1 to GM4 SKPI — Verification of ANSP EoSM by NSA/competent authority — Appendix 1 to GM11 SKPI — Look-up table for severity classification of ATM-specific occurrences — Appendix 1 to GM13 SKPI — Just culture — State level — possible evidence — Appendix 1 to GM14 SKPI — Just culture — ANSP level — possible evidence Page 64 of 64