Zhiyuan Lin

Staff User

AI4RAILS2024

by

Welcome to AI4RAILS 2024

The 5th International Workshop on “Artificial Intelligence for RAILwayS”

co-located with the International Conference on Optimization and Decision Science – ODS2024

September 8th-9th (provisional), 2024, Badesi, Sardinia, Italy

The impact of Artificial Intelligence (AI) on the industry has been so disruptive that it gave rise to a new wave of research and applications. Several industries, mainly in logistics and manufacturing, have benefited significantly from AI adoption and this positive trend is planned to be kept in the future. Modern and future railways represent one of the fields in which AI is expected to have a significant impact in a medium to long term perspective, to get higher levels of automation. Despite this opening for unprecedented scenarios in railway systems, it also raises concerns regarding system dependability and new threats associated with a high level of autonomy.

Therefore, the first step towards the adoption of AI in the railway sector is understanding to what extent AI can be considered reliable, safe and secure. In such a context, building upon the  RAILS project (Roadmaps for A.I. Integration in the Rail Sector, funded by Shift2Rail) the annual AI4RAILS workshop series are organized with the aim to provide annually a forum for researchers, practitioners and business leaders to discuss and share new perspectives, ideas, technologies and solutions for effective and dependable integration of AI techniques in rail-based transport systems.

The program includes:

  • Keynote speaker’s talk
  • Technical presentation sessions: detailed program will be announced later.

Keynote talk

Hybrid OR-AI algorithms for Railway Traffic Management

Speaker: Prof Paola Pellegrini, Université Gustave Eiffel

Abstract: 

Real-time railway traffic management is a challenging task that is mostly carried out manually today. Several models have been proposed in the literature, and OR techniques have been used for optimizing this task. A promising research direction consists in exploiting AI techniques in combination to OR ones for optimizing traffic. This talk will present three different approaches for hybridizing AI and OR for railway traffic management optimization. First, an Ant Colony Optimization-MILP algorithm will show how we can reduce the search space of instances to promising regions, and achieve their effective exploration. Second, a Graph Convolutional Network-MILP algorithm will allow optimizing traffic considering passenger demand prediction. Third, a consensus-MILP algorithm will make self-organizing traffic an actual option for the railway system. Some unexplored research directions on hybrid AI-OR approaches for railway traffic management will be sketched to conclude the talk.

Bio:

Prof Paola Pellegrini is an expert in railway planning and operational management. Her research focuses on the development of optimization approaches to effectively exploit railway infrastructure capacity, aiming to the automation of the involved processes. Paola has participated in a number of national and international research projects, and she regularly collaborates with various SMEs to facilitate the knowledge transfers from academia to practice. She is a member of the Board of the International Association of Railway Operations Research (IAROR).

Themes and Goals

AI4RAILS workshop is addressing topics related to the adoption of artificial intelligence technologies in railway transport. AI4RAILS is set to showcase AI opportunities over the holistic railway system including traffic planning and management, passenger mobility, predictive maintenance, autonomous driving, transport safety and policy, and including mainline and high-speed railways, metro, tram, and hyperloop. We welcome various applications of AI including machine learning, neural networks and reinforcement learning, evolutionary algorithms, computer vision and knowledge-based approaches in railway transport for both freight and passenger transport. Also, both pure AI models as well as hybrid models combining AI and optimization are foreseen.

Topics of interest include, but are not limited to:

  • AI for rail traffic state predictions
  • AI for train timetabling, rolling stock and crew scheduling
  • Self-organising, decentralised and conventional real-time rail traffic and disruption management
  • Railway predictive maintenance and defect detection and prediction
  • Predicting passenger demand and flows
  • Data analytics for railway transport
  • Using AI for improving optimization models for planning and management
  • Using optimization for improving machine learning models
  • Novel problems enabled by AI techniques
  • Autonomous and cooperative driving, including virtual coupling
  • Cognitive Digital Twins for Railways
  • Explainable and trustworthy AI in railway applications
  • Transferability to railways of AI approaches from other sectors (e.g., aviation, road, etc.)
  • Datasets and real-world case studies

Important Dates

February 24, 2024

March 4th, 2024 (extended)

 Deadline for short papers
February 24, 2024

March 8th, 2024 (extended)

   Deadline for abstracts
March 15, 2024 Abstract acceptance notification
March 17, 2024 Short paper acceptance notification
March 24, 2024   Deadline for camera-ready papers
March 24, 2024   Deadline for early registration
March 31, 2024   Deadline for the payment of the deposit at the discounted rates to the resort Le Dune (25% of the total amount due is requested). After this date, available rooms can be booked at higher rates by mail to meeting@delphina.it by the same deposit.
June 30, 2024   Registration deadline for inclusion in the final program and deadline for payment of the balance to the resort “Le Dune”  
September 8-9, 2024 Workshop

Short paper / abstract submission

All researchers, academics, practitioners, and students working on the conference themes are invited to participate to ODS2024/AI4RAILS2024 submitting either a short paper (8-10 pages, including front matter and references) or an abstract (max 2000 characters, space included). Please use the official Word template provided by the conference: http://www.airoconference.it/ods2024/submissions

Please email your abstract directly to our AI4RAILS official email address: rails.project.eu@gmail.com

Workshop Organizers

Workshop Program Committee

  • Valentina Cacchiani, University of Bologna, Italy
  • Lorenzo De Donato, Unversity of Naples Federico II, Italy
  • Francesco Flammini, Mälardalen University & Linnaeus University, Sweden
  • Rob Goverde, TU Delft Transport Institute, Netherlands
  • Bisheng He, Southwest Jiaotong University , China
  • Ronghui Liu, University of Leeds, United Kingdom
  • Zhenliang Ma, Royal Institute of Technology (KTH), Sweden
  • Stefano Marrone, University of Naples Federico II, Italy
  • Roberto Nardone, University of Naples “Parthenope”, Italy
  • Paola Pellegrini, Gustav Eiffel University, France
  • Felipe Rodrigues, Technical University of Denmark
  • Stefania Santini, University of Naples Federico II, Italy
  • Valeria Vittorini, University of Naples Federico II, Italy
  • Pengling Wang, Tongji University, China
  • Jiateng Yin, Beijing Jiaotong University, China
  • Shuguang Zhan, Hefei University of Technology, China
  • Yongqiu Zhu, TU Delft Transport Institute, Netherlands

Contact

For all questions related to the workshop, contact the organizers via email.

 

Presentation slides from the RAILS Final Workshop & Roadmapping Event

by

We are thrilled to announce that our workshop was a huge success! We had a great turnout of participants who engaged in lively discussions and shared their insights and experiences. We are very grateful to everyone who joined us and contributed to this amazing event.

We know that some of you may have missed the workshop or want to revisit some of the topics we covered. That’s why we are happy to share with you all the presentation slides and video recording of the session on our project webpage. You can access them anytime and learn more about our project and its goals.

We hope you enjoy this valuable resource and stay tuned for more updates from us. Thank you again for your support and participation!

Presentation slides:

 

Recording:

 

Photos from the events:

AI4RAILS 2023

by

Welcome to AI4RAILS 2023

The 4th International Workshop on “Artificial Intelligence for RAILwayS”

co-located with the International Conference on Optimization and Decision Science – ODS2023

September 4th, 2023

Ischia, Italy

The impact of Artificial Intelligence (AI) on the industry has been so disruptive that it gave rise to a new wave of research and applications. Several industries, mainly in logistics and manufacturing, have benefited significantly from AI adoption and this positive trend is planned to be kept in the future. Modern and future railways represent one of the fields in which AI is expected to have a significant impact in a medium to long term perspective, to get higher levels of automation. Despite this opening for unprecedented scenarios in railway systems, it also raises concerns regarding system dependability and new threats associated with a high level of autonomy.

Therefore, the first step towards the adoption of AI in the railway sector is understanding to what extent AI can be considered reliable, safe and secure. In such a context, building upon the ongoing RAILS project (Roadmaps for A.I. Integration in the Rail Sector, funded by Shift2Rail) in 2019 the AI4RAILS workshop series has been started with the aim to provide annually a forum for researchers, practitioners and business leaders to discuss and share new perspectives, ideas, technologies and solutions for effective and dependable integration of AI techniques in rail-based transport systems.

The program includes:

  • Keynote speaker: Prof Mathijs de Weerdt, TU Delft.
  • Technical presentation sessions: detailed program will be announced later.

The AI4RAILS workshop is supported by the AIRO stream “OPTSM Optimization in Public Transport and Shared Mobility”.

Keynote talk

Hybrid algorithms for train shunting and servicing at railway hubs

Prof Mathijs de Weerdt, TU Delft

Abstract: To support trains executing the railway timetable, they need to be cleaned, undergo security checks, and be parked in an order that matches their departures according to the timetable. When shunting yards are nearing their capacity, deciding on the order of these activities becomes so computationally challenging that state-of-the-art mixed-integer linear programming solvers fail to produce solutions for average-sized railway hubs. In this talk a number of alternative models and algorithms are explored to address this computational challenge, inspired from successes with for example search algorithms, decision diagrams, multi-agent path planning, and pebble motion. Underlying this is a more general lesson about what to include in a model, what to decide upon later, and how to make such modeling decisions.

Bio: Mathijs de Weerdt is full professor on Algorithms for Planning and Scheduling and section head of the Algorithmics Group at Delft University of Technology. His main aim is to develop new algorithmic techniques that cope with challenges prohibiting successful use of algorithmic and artificial intelligence planning and optimization techniques in practice. He mainly works on such algorithmic challenges related to the energy transition and railway transportation.

 

Themes and Goals

AI4RAILS workshop is addressing topics related to the adoption of artificial intelligence technologies in railway transport. AI4RAILS is set to showcase AI opportunities over the holistic railway system including traffic planning and management, passenger mobility, predictive maintenance, autonomous driving, transport safety and policy, and including mainline and high-speed railways, metro, tram, and hyperloop. We welcome various applications of AI including machine learning, neural networks and reinforcement learning, evolutionary algorithms, computer vision and knowledge-based approaches in railway transport for both freight and passenger transport. Also, both pure AI models as well as hybrid models combining AI and optimization are foreseen.

Topics of interest include, but are not limited to:

  • AI for rail traffic state predictions
  • AI for train timetabling, rolling stock and crew scheduling
  • Self-organising, decentralised and conventional real-time rail traffic and disruption management
  • Railway predictive maintenance and defect detection and prediction
  • Predicting passenger demand and flows
  • Data analytics for railway transport
  • Using AI for improving optimization models for planning and management
  • Using optimization for improving machine learning models
  • Novel problems enabled by AI techniques
  • Autonomous and cooperative driving, including virtual coupling
  • Cognitive Digital Twins for Railways
  • Explainable and trustworthy AI in railway applications
  • Transferability to railways of AI approaches from other sectors (e.g., aviation, road, etc.)
  • Datasets and real-world case studies

 

Important Dates

January 30, 2023 Submission of abstract opens
March 31, 2023 Early registration opens
May 18, 2023 (extended) 23:59 BST Deadline for abstract submission (max 2000 characters, space included)
May 27, 2023 Abstract acceptance notification
May 31, 2023 Deadline for early registration
June 15, 2023 Deadline for camera-ready abstract

Abstract submission

All researchers, academics, practitioners, and students working on the conference themes are invited to participate to ODS2023/AI4RAILS submitting an abstract (max 2000 characters, space included). Please use the official Word template provided by the conference: http://www.airoconference.it/ods2023/call-for-papers

Please email your abstract directly to our AI4RAILS official email address: rails.project.eu@gmail.com

Special Issues

After the conference, regular papers can be submitted for publication in special issues of indexed international journals. Information will be provided in the subsequent announcements and on the website.

Workshop Organizers

Workshop Program Committee

  • Paola Pellegrini, Gustav Eiffel University, France
  • Ronghui Liu, University of Leeds, United Kingdom
  • Rob Goverde, TU Delft Transport Institute, Netherlands
  • Valeria Vittorini, University of Naples Federico II, Italy
  • Francesco Flammini, Mälardalen University & Linnaeus University, Sweden
  • Lorenzo De Donato, Unversity of Naples Federico II, Italy
  • Stefano Marrone, University of Naples Federico II, Italy
  • Roberto Nardone, University of Naples “Parthenope”, Italy
  • Stefania Santini, University of Naples Federico II, Italy
  • Ruifan Tang, University of Leeds, United Kingdom
  • Jin Liu, University of Leeds, United Kingdom
  • Zhenliang Ma, Royal Institute of Technology (KTH), Sweden
  • Felipe Rodrigues, Technical University of Denmark

Contact

For all questions related to the workshop, contact the organizers via email.

 

Trustworthy AI for safe autonomy of smart railways: directions and lessons learnt from other sectors

by

Our PhD student Lorenzo De Donato will present a talk “Trustworthy AI for safe autonomy of smart railways: directions and lessons learnt from other sectors” in the The 13th World Congress on Railway Research is coming to Birmingham, UK 6-10 June 2022 (https://www.wcrr2022.co.uk/website/938/ ). See here for details:

https://www.researchgate.net/publication/361114998_Trustworthy_AI_for_safe_autonomy_of_smart_railways_directions_and_lessons_learnt_from_other_sectors

AI4RAILS 2022

by

Welcome to AI4RAILS 2022

3rd International Workshop on “Artificial Intelligence for RAILwayS”

colocated with EDCC 2022

September 12, 2022

Zaragoza, Spain

*** The program is now available here ***

The impact of Artificial Intelligence (AI) on the industry has been so disruptive that it gave rise to a new wave of research and applications. Several industries, mainly in logistics and manufacturing, have benefited significantly from AI adoption and this positive trend is planned to be kept in the future. Modern and future railways represent one of the fields in which AI is expected to have a significant impact in a medium to long term perspective, to get higher levels of automation. Despite this opening for unprecedented scenarios in railway systems, it also raises concerns regarding system dependability and new threats associated with a high level of autonomy.

Therefore, the first step towards the adoption of AI in the railway sector is understanding to what extent AI can be considered reliable, safe and secure. In such a context, building upon the ongoing Shift2Rail project RAILS (Roadmaps for the AI integration in the raiL Sector), in 2019 the AI4RAILS workshop series has been started with the aim to provide annually a forum for researchers, practitioners and business leaders to discuss and share new perspectives, ideas, technologies and solutions for effective and dependable integration of AI techniques in rail-based transportation systems.

Themes and Goals

AI4RAILS 2022 is the 3rd workshop in the series, designed to gather the best idea, comments and experiences related to the adoption of AI in railways. The series is supported by an ongoing Shift2Rail project, named RAILS (Roadmaps for the AI integration in the raiL Sector), and represents an open forum for researchers working on these themes. In fact, the Advisory Board of the RAILS project will be invited since it includes railway stakeholders and companies operating in both the AI and railway market as well as the entire set of companies and research centres of the Shift2Rail community.

The themes of the workshop are relevant for the EDCC community and consistent with its aims and scope. The railway domain represents a critical domain in which all the aspects of dependability and security should be continuously addressed as a consequence of technical and technological progress. In the context of AI adoption, current literature includes different attempts of adopting AI techniques for increasing the automation level and the overall dependability of railways. AI4RAILS represents a unique forum for researchers, practitioners and business leaders to present and discuss their latest research results on theory, approaches and tools for the design, validation, operation and evaluation of dependable and secure railways by means of AI techniques.

Topics of interest include, but are not limited to:

  • AI for railway safety, availability, reliability and security

  • Transferability to railways of AI approaches from other sectors (e.g., avionics, automotive, etc.)

  • Autonomous and cooperative driving, including virtual coupling

  • Automatic detection of signals, pedestrians and obstacles

  • Predictive maintenance and defect detection

  • AI for traffic planning and management

  • Adversarial perturbations to AI with a focus on railway case-studies

  • Cognitive Digital Twins for Railways

  • Resource allocation with a focus on railway case-studies

  • Run-time model checking and AI with a focus on railway case-studies

  • Explainable and trustworthy AI in railway applications

  • Safe autonomy standards and railway certification of compliance

  • Datasets and real-world case studies

Important Dates

Paper submission deadline: June 20th, 2022 (EXTENDED)

Author notification: July 2nd, 2022 (EXTENDED)

Camera-ready paper due: July 9th, 2022 (EXTENDED)

Paper submission

Submitted papers must be unpublished and not considered elsewhere for publication. Submissions will undergo a rigorous review process handled by the Technical Program Committee. Papers will be selected based on their originality, significance, relevance, and clarity of presentation. Only electronic submissions in PDF format through the submission site (https://easychair.org/my/conference?conf=ai4rails2022) will be considered.

The AI4RAILS proceedings will be published by Springer in the series Communications in Computer and Information Science (CCIS).

Papers must be in English, up to 12 pages in Springer format, including references and appendices. Information for authors, LaTeX and Microsoft Word templates, as well as formatting guidelines, can be found here.

Submitting authors are requested to note that the publication of the paper in the proceedings will be subject to attending the conference and presenting in person (unless the pandemic situation does not allow it) and register their attendance upon submitting the camera ready version.

Workshop Steering Committee

Workshop Chairs & Organizers

Program Committee Members

  • Nikola Bešinović, Delft University of Technology, Netherlands

  • Lorenzo De Donato, Unversity of Naples Federico II, Italy

  • Francesco Flammini, Mälardalen University & Linnaeus University, Sweden

  • Rob Goverde, TU Delft Transport Institute, Netherlands

  • Ronghui Liu, University of Leeds, United Kingdom

  • Fredrik Heintz, Linköping University, Sweden

  • Leonardo Impagliazzo, Hitachi Rail STS, Italy

  • Zhiyuan Lin, University of Leeds, UK

  • Stefano Marrone, University of Naples Federico II, Italy

  • Claudio Mazzariello, Hitachi Rail STS, Italy

  • Roberto Nardone, University of Naples “Parthenope”, Italy

  • Alberto Petrillo, University of Naples Federico II, Italy

  • Danijela Ristic-Durrant, University of Bremen, Germany

  • Carlo Sansone, University of Naples Federico II, Italy

  • Stefania Santini, University of Naples Federico II, Italy

  • Valeria Vittorini, University of Naples Federico II, Italy

Slides from our RAILS Mid-term Workshop

by

Slides from our RAILS Mid-term Workshop (25 February 2022) are now available on our website:

Invited speakers:

“New steps forward in railways smart maintenance”: In2Smart2

“Electrical and electronic applications for railways”: Cenelec_TC_9X_SG34

“Validate AI Based Perception”: Validate_AI_Based_Perception

“The need for standard XAI — Explainable Artificial Intelligence”: ExplainableAI

RAILS:

Bird’s eye view: RAILS_MT

WP1: RAILS_MT_WP1

WP2: RAILS_MT_WP2

WP3: RAILS_MT_WP3

WP4: RAILS_MT_WP4

RAILS Mid-term Project Workshop (online)

by

The RAILS mid-term workshop will be held online (MS Teams) on 25th February 2022.

The main objectives of the workshop are:

  • to present the results of the analysis phase;
  • to introduce the next steps and the pilot case studies identified so far;
  • to discuss the safe integration of AI in autonomous driving, predictive maintenance, and traffic management.

Talks from RAILS researchers and invited speakers are planned for the morning.
A panel of experts will be held early in the afternoon to discuss aspects related to the development of roadmaps for the integration of data-driven and AI techniques in the rail sector.

The draft program of the workshop is now available here!

To attend the workshop, please register here.

 

RAILS will attend several conferences

by

Members of RAILS will attend several conferences presenting output of the project in 2021.

– RAILS at 2021 INFORM Annual Meeting

(https://meetings2.informs.org/wordpress/anaheim2021/)

 

– RAILS at IsoLa 2021, DisCoRail Session

(http://isola-conference.org/isola2021/)

 

– RAILS at INNORAIL 2021

(https://innorail2021.hu/en/)

Lorenzo DE DONATO PhD Student in Computer Engineering, RAILS Ambassador, University of Naples Federico II, Italy: Roadmaps for the Integration of Artificial Intelligence in Railways

 

– RAILS at SmartRail Europe 2021

(https://www.smartrailcongress.com/)

Suicide prevention strategies & effective track detection technologies

Francesco Flammini
RAILS Technical Manager

Wed 24, Nov 3:00 pm – 3:20 pm

During this session we examine the links between fatalities and factors such as homelessness and lifestyles and the strategies and technologies being used to effectively reduce occurrences in what is a difficult but important issue to address. We will also consider new developments in track detection technologies to prevent trespassing incidents.

Challenges and State-of-Practice Survey of Artificial Intelligence in Railways

by

(Original webpage: https://docs.google.com/forms/d/e/1FAIpQLSeUNBU3VUCBXzaM5O3sqM4kpMM8GgwPTs7wpU_zXkmecOTO3g/viewform)

This survey aims at obtaining a deeper understanding of the Challenges and State-of-Practice (SoP) of Artificial Intelligence (AI) in railway transportation from the perspective of diverse stakeholders. Results are collected in the context of the H2020 Shift2Rail project RAILS (Roadmaps for A.I. integration in the raiL Sector), to understand priorities and requirements when defining the roadmaps for AI integration in the rail sector.

The survey is distributed to Shift2Rail members, project partners and other project participants, as well as to other railway organizations and regulatory bodies. Aggregated and anonymous results will be disseminated primarily through the project website https://rails-project.eu/.

You are welcome to add customized answers by using the text box “Other”.

We would like to thank you in advance for the time you will invest in filling this survey!

IMPORTANT NOTE:
Please note that at the end of the questionnaire you can decide whether to share your personal data with the project team and if you are available to be contacted, otherwise you can remain fully anonymous.
Please note that even if you agree to be contacted, we will not share your name and affiliation with anyone except the project team. Only aggregated and anonymous results of the survey will be publicly disclosed.

If you encounter any problems, need clarifications, or want to report your feedback on the questionnaire, please contact:
Prof. Valeria Vittorini (project leader): valeria.vittorini@unina.it
Prof. Francesco Flammini (technical leader): francesco.flammini@lnu.se

THE SURVEY WILL CLOSE ON 31st MAY 2021!

First Advisory Board Meeting

by

We successfully ran the first Advisory Board Meeting on 1st March 2021 online. We would like to thank all AB members who participated in this meeting and for their valuable contributions! We saw many interesting presentations there and would like to keep this momentum in the future.

AB memebers to joined the meeting:

  • Aitex (IT)
  • Comesvil (IT)
  • European Union Agency for Railways – ERA (EU)
  • Hitachi Rail STS (IT)
  • Netherlands Railways (NL)
  • NextTechnologies Ltd. (HU)
  • ProRAIL (NL)
  • SYENMAINT (IT)