Envisioning future resilient AI-enabled Flight Information Service
Woltjer, Rogier LU
; Stefansson, Boel
and Bjursten Carlsson, Christian
(2025)
Joint 11th Biennial Symposium of the Resilience Engineering Association and 14th Annual Resilient Health Care Society Meeting
p.183-189
- Abstract
- This paper investigates resilient performance of Air Traffic Management with future Artificial Intelligence (AI)-based tools. A case study was performed on an early prototype of an AI-enabled Flight Information Service (FIS) tool, using a survey of flight instructors and air traffic controllers, and a focus group with controllers with flying experience. The study focused on what characterizes good Flight Information Service, what works well today, and how this may be transferred to an envisioned AI tool. Data analysis through thematic coding resulted in eight themes: 1) Adapting to context; 2) Strategies of work; 3) Understanding user needs; 4) Uncertainty management; 5) Communication /coordination; 6) System interactions; 7) Cognitive... (More)
- This paper investigates resilient performance of Air Traffic Management with future Artificial Intelligence (AI)-based tools. A case study was performed on an early prototype of an AI-enabled Flight Information Service (FIS) tool, using a survey of flight instructors and air traffic controllers, and a focus group with controllers with flying experience. The study focused on what characterizes good Flight Information Service, what works well today, and how this may be transferred to an envisioned AI tool. Data analysis through thematic coding resulted in eight themes: 1) Adapting to context; 2) Strategies of work; 3) Understanding user needs; 4) Uncertainty management; 5) Communication /coordination; 6) System interactions; 7) Cognitive work; and 8) Information management. Good Flight Information Service: a) is highly dependent on adapting to new circumstances and being flexible; b) entails adapting to sudden changes in, e.g., airspace allocation, weather conditions, and traffic; c) is partly formalized and partly improvised; d) considers the controller’s goals to always provide the highest level of service, as proactively as possible, timing actions in context, balancing these with other demands; e) provides a form of decision
support; and f) involves uncertainty management, with incomplete, uncertain, and potentially incorrect information, adding to complexity. Work-as-done in current FIS sets a high bar for future AI-based FIS. The study indicates that this light method for assessing resilience is suitable for early AI-based operations. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/17530a27-fa76-49f1-9130-e7639e40e664
- author
- Woltjer, Rogier
LU
; Stefansson, Boel
and Bjursten Carlsson, Christian
- organization
- publishing date
- 2025
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Resilience, Safety-II, Flight information service, Artificial Intelligence, Work-as-done
- host publication
- Proceedings of the Joint 11th Biennial Symposium of the Resilience Engineering Association and 14th Annual Resilient Health Care Society Meeting
- editor
- Henriqson, Éder ; Wachs, Priscila and Saurin, Tarcisio Abreu
- pages
- 7 pages
- conference name
- Joint 11th Biennial Symposium of the Resilience Engineering Association and 14th Annual Resilient Health Care Society Meeting
- conference location
- Canela, Brazil
- conference dates
- 2025-10-20 - 2025-10-24
- ISBN
- 978-65-01-74859-7
- language
- English
- LU publication?
- yes
- id
- 17530a27-fa76-49f1-9130-e7639e40e664
- date added to LUP
- 2025-11-19 21:42:38
- date last changed
- 2025-12-09 10:28:50
@inproceedings{17530a27-fa76-49f1-9130-e7639e40e664,
abstract = {{This paper investigates resilient performance of Air Traffic Management with future Artificial Intelligence (AI)-based tools. A case study was performed on an early prototype of an AI-enabled Flight Information Service (FIS) tool, using a survey of flight instructors and air traffic controllers, and a focus group with controllers with flying experience. The study focused on what characterizes good Flight Information Service, what works well today, and how this may be transferred to an envisioned AI tool. Data analysis through thematic coding resulted in eight themes: 1) Adapting to context; 2) Strategies of work; 3) Understanding user needs; 4) Uncertainty management; 5) Communication /coordination; 6) System interactions; 7) Cognitive work; and 8) Information management. Good Flight Information Service: a) is highly dependent on adapting to new circumstances and being flexible; b) entails adapting to sudden changes in, e.g., airspace allocation, weather conditions, and traffic; c) is partly formalized and partly improvised; d) considers the controller’s goals to always provide the highest level of service, as proactively as possible, timing actions in context, balancing these with other demands; e) provides a form of decision <br/>support; and f) involves uncertainty management, with incomplete, uncertain, and potentially incorrect information, adding to complexity. Work-as-done in current FIS sets a high bar for future AI-based FIS. The study indicates that this light method for assessing resilience is suitable for early AI-based operations.}},
author = {{Woltjer, Rogier and Stefansson, Boel and Bjursten Carlsson, Christian}},
booktitle = {{Proceedings of the Joint 11th Biennial Symposium of the Resilience Engineering Association and 14th Annual Resilient Health Care Society Meeting}},
editor = {{Henriqson, Éder and Wachs, Priscila and Saurin, Tarcisio Abreu}},
isbn = {{978-65-01-74859-7}},
keywords = {{Resilience; Safety-II; Flight information service; Artificial Intelligence; Work-as-done}},
language = {{eng}},
pages = {{183--189}},
title = {{Envisioning future resilient AI-enabled Flight Information Service}},
year = {{2025}},
}