Periodic Reporting for period 1 - SafeOPS (From Prediction to Decision Support - Strengthening Safe and Scalable ATM Services through Automated Risk Analytics based on Operational Data from Aviation Stakeholders)
Période du rapport: 2021-01-01 au 2021-12-31
The next generation of Air Traffic Management (ATM) systems are pushed more and more towards digitalization, driven by two goals that are hard to combine. Firstly, the demand for capacity and cost-efficiency of air transport operations increases. Secondly, already high levels of safety and resilience in the ATM system must be maintained and then continuously improved. As a mid-term solution, we propose integrating a digitalized system with human operational management, introducing quantifiable performance predictions into ATM. This digitalized system will be based on big data technologies, including the fusion of data from different sources. Organizing and making use of the vast amount of available sources in aviation, will pave the way for Artificial Intelligence (AI) solutions, such as predictive risk estimation and ultimately decision support tools. These solutions will enable safety applications that create a proactive, data-driven approach for safety management, capable of predicting potential safety hazards in real-time.
Using today's technologies for big data, large historical datasets of radar data can be annotated with on-board aircraft performance data. Through predictive analytics, an air traffic control officer (ATCO) could be informed about potential missed approaches ahead of time. A controller having this information at hand could more quickly predict landing traffic, estimate runway capacity and plan for likely go-around events. It can potentially reduce their task- and workload compared to current ATM operations.
The question addressed by SafeOPS is, how predictive tools and the inherently probabilistic nature of their outputs will change the way the system is operated. Beyond “information overflow”, the ATM human agents will have to adapt to more, but also mostly probabilistic information provided by big data analytics. Clever HMI refinements will certainly help to mitigate the potential overflow of information. However, also research on the impact of information automation on the ATM system needs to be conducted. It must show that an increase of safety and cost-efficiency can be achieved and also the resilience of the system is maintained or further improved. SafeOPS aims to foster a collaborative paradigm that involves both the ATM and airline operations worlds to identify possibly hidden safety risks.
The main objectives of SafeOPS can be broken down into three main branches of work:
1. Investigate concepts for the integration of AI/ML based decision support tools in ATM, and evaluate the effects on capacity, safety and resilience of the ATM operation.
2. Enhance risk assessment methods, such that they can cope with the introduced AI/ML component.
3. Develop an AI/ML tool that supports ATCOs in their decision making in complex situations through the provision of predictive risk estimations. As an example, for decision making in complex situation in ATM, SafeOPS exemplarily investigates go-around handling.
Using today's technologies for big data, large historical datasets of radar data can be annotated with on-board aircraft performance data. Through predictive analytics, an air traffic control officer (ATCO) could be informed about potential missed approaches ahead of time. A controller having this information at hand could more quickly predict landing traffic, estimate runway capacity and plan for likely go-around events. It can potentially reduce their task- and workload compared to current ATM operations.
The question addressed by SafeOPS is, how predictive tools and the inherently probabilistic nature of their outputs will change the way the system is operated. Beyond “information overflow”, the ATM human agents will have to adapt to more, but also mostly probabilistic information provided by big data analytics. Clever HMI refinements will certainly help to mitigate the potential overflow of information. However, also research on the impact of information automation on the ATM system needs to be conducted. It must show that an increase of safety and cost-efficiency can be achieved and also the resilience of the system is maintained or further improved. SafeOPS aims to foster a collaborative paradigm that involves both the ATM and airline operations worlds to identify possibly hidden safety risks.
The main objectives of SafeOPS can be broken down into three main branches of work:
1. Investigate concepts for the integration of AI/ML based decision support tools in ATM, and evaluate the effects on capacity, safety and resilience of the ATM operation.
2. Enhance risk assessment methods, such that they can cope with the introduced AI/ML component.
3. Develop an AI/ML tool that supports ATCOs in their decision making in complex situations through the provision of predictive risk estimations. As an example, for decision making in complex situation in ATM, SafeOPS exemplarily investigates go-around handling.
How big data and artificial intelligence based decision support systems could impact daily air traffic operations has not been explored yet. Over the course of the initial five months, the SafeOPS team held several workshops together with air traffic controllers from two major European hubs to elaborate this question in the context of go-around handling. Based on these workshops, scenarios have been identified in which go-arounds can lead to complex situations in daily operations. Based on these scenarios, potential use cases for a decision support tool, able to predict go-arounds, have been elaborated.
A requirements engineering approach is used to refine the use cases into user stories and finally requirements to shape an initial design proposal. The result of this work defines requirements, that outline the further development process of the project. Additionally, a technical problem statement is outlined. This problem statement gives an overview on the state of the art technical methods as well as the challenges which can be derived from the requirements at this stage of the project.
One aspect of the incorporation of a predictive technology in the air traffic operating environment is the risk associated with the technology integration, management and use. Therefore SafeOPS is assigned to the investigation of this risk, structured as a ‘Risk Framework’. The Risk Framework presented by SafeOPS will analyse the impact of the technology and the information presented to the ATCOs. A first process step in the Risk Framework will be developing a risk model. The risk model will be used to determine the resulting risks of each specific operational context under assessment. The initial phase of the process in the compilation of the Risk Framework, namely a systematic review of current risk models available for application in an aviation context. The review provides a critical assessment of existing risk models and their suitability for use in the SafeOPS Risk Framework. The review provides a recommendation on the most appropriate model for use in the SafeOPS context. The review selected risk models which are widely used and validated in an aviation environment, inclusive of models applied to air traffic control or aircraft operations.
A futher aspect of SafeOPS is to investigate the big-data related challenges, a aritificial inelligence based decision support system poses. Therefore, SafeOPS set up a big data working infrastructure and collected datasets, that will be used to train AI models for the prediction of go-arounds. Data Cleaning and (pre-)processing tasks are perforemed and the artificial intelligence case studies are prepared.
A requirements engineering approach is used to refine the use cases into user stories and finally requirements to shape an initial design proposal. The result of this work defines requirements, that outline the further development process of the project. Additionally, a technical problem statement is outlined. This problem statement gives an overview on the state of the art technical methods as well as the challenges which can be derived from the requirements at this stage of the project.
One aspect of the incorporation of a predictive technology in the air traffic operating environment is the risk associated with the technology integration, management and use. Therefore SafeOPS is assigned to the investigation of this risk, structured as a ‘Risk Framework’. The Risk Framework presented by SafeOPS will analyse the impact of the technology and the information presented to the ATCOs. A first process step in the Risk Framework will be developing a risk model. The risk model will be used to determine the resulting risks of each specific operational context under assessment. The initial phase of the process in the compilation of the Risk Framework, namely a systematic review of current risk models available for application in an aviation context. The review provides a critical assessment of existing risk models and their suitability for use in the SafeOPS Risk Framework. The review provides a recommendation on the most appropriate model for use in the SafeOPS context. The review selected risk models which are widely used and validated in an aviation environment, inclusive of models applied to air traffic control or aircraft operations.
A futher aspect of SafeOPS is to investigate the big-data related challenges, a aritificial inelligence based decision support system poses. Therefore, SafeOPS set up a big data working infrastructure and collected datasets, that will be used to train AI models for the prediction of go-arounds. Data Cleaning and (pre-)processing tasks are perforemed and the artificial intelligence case studies are prepared.
There has already been, and are still ongoing, various research activities within the SESAR JU on advanced automation and the use of data analysis solutions in the Air Traffic Control environment. Also other initiatives, such as Eurocontrol's System-Wide Information Management (SWIM), aim to reveal the potential, incorporated in collected data and the need to further support the decision taking of air traffic controllers.
SafeOPS will progress the state of the art, complementing existing and past projects, and providing new insights and analysis on how the increase in digitization and automation could impact the safety and resilience in the ATM industry. We find that, although extensive research is dedicated to new tools, algorithms and interfaces, the management of the inherent but mathematically measured uncertainty that AI/ML tools provide, will help to progress on how this new support tools affect the decision making processes of ATCO's.
SafeOPS increases the body of knowledge of safety and resilience in the context of new digitalization tools supporting ATM. Particularly, SafeOPS contributes by developing an operational risk framework that leverages the existing research on machine learning. Outcomes of analytical methods, including predictions, are integrated into the operational environments. Implications, benefits, and disadvantages associated with the outcomes are carefully analyzed. The close interaction between aviation stakeholders, including an ANSP, two airlines, and research institutions, with respect to applied data science is a novelty in the research field of ATM.
SafeOPS will progress the state of the art, complementing existing and past projects, and providing new insights and analysis on how the increase in digitization and automation could impact the safety and resilience in the ATM industry. We find that, although extensive research is dedicated to new tools, algorithms and interfaces, the management of the inherent but mathematically measured uncertainty that AI/ML tools provide, will help to progress on how this new support tools affect the decision making processes of ATCO's.
SafeOPS increases the body of knowledge of safety and resilience in the context of new digitalization tools supporting ATM. Particularly, SafeOPS contributes by developing an operational risk framework that leverages the existing research on machine learning. Outcomes of analytical methods, including predictions, are integrated into the operational environments. Implications, benefits, and disadvantages associated with the outcomes are carefully analyzed. The close interaction between aviation stakeholders, including an ANSP, two airlines, and research institutions, with respect to applied data science is a novelty in the research field of ATM.