Master Thesis AI-Driven Transformation of Railway System Requirements into Executable Models

Req ID:499951

At Alstom we understand transport networks and what moves people. From high-speed trains metros monorails and trams to turnkey systems services infrastructure signalling and digital mobility we offer our diverse customers the broadest portfolio in the industry. Every day more than 80 000 colleagues lead the way to greener and smarter mobility worldwide connecting cities as we reduce carbon and replace cars.

Background
Alstom is a global leader in the railway industry driving innovation in sustainable and intelligent mobility solutions. As the demand for advanced railway systems continues to grow Alstom remains committed to developing reliable safe and efficient train control and management solutions. A key challenge in this development process lies in the transformation of high-level requirements into executable software models used in train control systems. This process demands expert knowledge precision and extensive validation to ensure that all safety-critical aspects are preserved throughout the lifecyclefrom design to implementation. Currently this transformation relies heavily on manual interpretation and expert supervision which can be time-consuming and resource-intensive. Automating or intelligently assisting this transformation would not only enhance workflow efficiency but also strengthen Alstoms position as a global innovator in the design of safe and intelligent trains..

Problem description and goals

Thesis Objective

The main objective of this Masters Thesis is to develop an intelligent transformation pipeline capable of converting system requirements authored in semi-structured natural language format into optimized executable models that align with Alstoms development workflow.

Recent advances in artificial intelligence particularly in large language models (LLMs) offer promising opportunities to bridge this gap. LLMs have demonstrated capabilities in understanding structured natural language generating formal representations and supporting software engineering tasks. However their application in safety-critical domains like rail transport remains underexplored especially in the context of requirement-to-model transformation and lifecycle traceability.

This thesis positions itself at the intersection of requirements engineering model-based development and AI-assisted software engineering aiming to investigate how the envisioned AI-based system can augment the development process by streamlining the model creation process reducing manual effort and ensuring traceability accuracy and compliance with system-level design principles.

Key Focus Areas

• Collaborate with requirement engineers to understand train system requirements.
• Analyze and become familiar with Alstoms industrial development workflow.
• Design and implement a transformation pipeline that converts textual or structured requirements into executable system models.
• Ensure the developed models are compatible with Alstoms development tools and standards.
• Investigate and apply state-of-the-art AI and model transformation techniques from current research.
• Gain an in-depth understanding of system-level architecture and safety-critical requirements in train software systems.

Expected Outcomes

By the end of the thesis the student is expected to deliver:

• A functional prototype or framework for automated requirement-to-model transformation.
• A detailed technical report and evaluation demonstrating the efficiency and accuracy of the developed approach.
• Recommendations for integrating the solution into Alstoms industrial development process.

Prerequisites:

• Masters student in Computer Science Software Engineering Electrical/Electronics Engineering Artificial Intelligence or related field.
• Strong Python programming skills (experience with AI/ML frameworks such as PyTorch or TensorFlow is an advantage).
• Understanding of Model-Based Systems Engineering (MBSE) or Model-Driven Development (MDD) concepts.
• Basic knowledge of Machine Learning Natural Language Processing (NLP) or AI-based automation.
• Good analytical and research skills.
• Effective communication skills in English (both written and spoken).
• Interest in AI applications for industrial and safety-critical systems preferably in the railway domain.

Duration: 20 weeks

Number of students: 1

Language of thesis: English

Is Swedish a language requirement No

Possibility to work from our office: Yes

You dont need to be a train enthusiast to thrive with us. We guarantee that when you step onto one of our trains with your friends or family youll be proud. If youre up for the challenge wed love to hear from you!

Important to note

As a global business were an equal-opportunity employer that celebrates diversity across the 63 countries we operate in. Were committed to creating an inclusive workplace for everyone.