PhD Position in MSCA Doctoral Network ASSESS at EPFL, Switzerland

PhD Vacancy for a Doctoral Candidate (DC12) in Hybrid digital-twins based on physics-informed graph neural networks for composite space structures at EPFL, Switzerland

Job description
The objective of this project is to develop a novel methodology that develops physics-informed graph neural networks (GNNs) for composite space structure analysis. The methodology will aim to leverage the inherent structure of composite materials, embedding physical laws and multi-physics constraints into the GNN framework. This approach will enhance predictive accuracy, reduce computational costs, and facilitate the design, simulation, and optimization of high-performance composite space structures, particularly for aerospace applications.

The ultimate goal of this project is to develop methodology that is able to be updated in real time with SHM measurements, while exhibiting predictive capabilities, in particular with respect to prognostics.

We also foresee a secondment at the beginning of the project in RUAG Schweiz AG (BEG) to make sure that the PhD student has access to all the required data and models. DC12 will also closely collaborate with DC5 and DC6. In particular, during the secondment at Politecnico di Torino, a collaboration will be established on developing FE and damage models to cross benefit between numerical models and hybrid models. Verification is by deliverable reports and scientific publications. This PhD is part of the ASSESS project, you can read more about this here.

Expected Tasks

• Develop the Physics-Informed GNN Architecture further leveraging the physical properties and multi-physics constraints inherent in composite materials
• Integrate real-time SHM data for dynamic updates with stability and accuracy.
• Implement, validate, and optimize the GNN framework for efficiency and predictive accuracy.
• Develop long-term prediction methods for structural health and failure prognostics.
• Ensure scalability for large, complex composite structures in high-performance environments.
• Document results, publish in journals, and present at key conferences.
• Create user-friendly tools and prototypes for practical applications.

Research Fields:

Sensors, materials, AI

Secondments:

• RUAG Schweiz AG (BEG), Switzerland
• Polytechnic University of Turin (POLITO), Italy
• University of Patras (UPA), Greece

Key Responsibilities

• Conduct research as outlined in the project’s objectives.
• Engage in secondments as part of the project’s training activities.
• Contribute to academic publications, project reports, and other dissemination activities.
• Participate in structured training programs and network-wide events organized by the consortium.
• Collaborate with international partners and other Doctoral Candidates within the network.

Principal Supervisor:

EPFL : Prof. Olga Fink

Requirements
Eligibility Criteria

• Master's degree (or equivalent).
• Not in possession of a doctoral degree at the date of the recruitment.
• Recruited applicants can be of any nationality and must undertake trans-national mobility (i.e., move from one country to another) when taking up the appointment. In particular, at the time of selection, the recruited applicant for this position must not have resided or carried out their main activity (work, studies, etc.) in Switzerland for more than 12 months in the 3 years immediately prior to their recruitment. You will be asked to enclose evidence to prove this. Short stays, such as holidays, are not taken into account.

Required Skills:

• Strong analytical background
• Proficiency in geometric deep learning, signal processing, statistics and learning theory
• The ability to work both independently and as part of a team
• Outstanding MSc degree in Engineering, Control, Computer Science, Physics, Applied Mathematics, or a related field

Optional Skills:

• Knowledge of French

English Requirement: High level of proficiency in English (will be tested at interview).

Academic Host
EPFL (CH) - https://www.epfl.ch/en/

Host laboratory: Intelligent Maintenance and Operations Systems (IMOS)

Website: https://www.epfl.ch/labs/imos/

Conditions of employment
Salary

• The successful candidate will receive a competitive and attractive academic remuneration package following the EPFL regulations for doctoral candidates: Basic starting salary of Doctoral Assistants and Postdocs ‒ Human Resources ‐ EPFL (https://www.epfl.ch/campus/services/human-resources/en/basic-starting-salary-of-doctoral-assistants-and-postdocs/).
• The financial benefit includes an estimated starting basic gross salary of CHF 55’150 per annum which is expected to increase by 1.000 CHF per year until year 4.
• The estimated gross salary includes both employee and employer’s contributions. The net salary is obtained after deducting these contributions (and any other nationally applicable direct taxes (e.g. income tax)) from the gross salary.

Other Benefits:

The doctoral candidate will also benefit from comprehensive training within the ASSESS network, including participation in at least six specialized Training Events organized by the MSCA-ASSESS network. Additional opportunities include secondments to partner laboratories, a range of advanced training courses (including transferable skills development), and active engagement in workshops and international conferences, providing a well-rounded and enriching research experience.

Expected Start Date: Q2 2025

Additional information
If you would like more information about this vacancy or the selection procedure, please contact Prof. Olga Fink, via olga.fink@epfl.ch.

Project Description:

The Horizon Europe MARIE SKŁODOWSKA-CURIE ACTION Doctoral Network (DN) ASSESS (Automated online monitoring of Smart compoSitE StructureS - Grant Agreement Number 101168031) offers 12 exciting PhD positions to develop next-generation smart composites for strategic European industries such as aviation, space, and wind energy.

Composite materials with embedded sensors and nanomaterials hold the potential for real-time structural health monitoring. However, industrial adoption is hindered by challenges in sensor integration, automated data analysis, standardization, and operator training. ASSESS bridges the gap between research and industry by advancing the design, manufacturing, and testing of Fibre Reinforced Polymer (FRP) composites.

As part of a multidisciplinary consortium of 9 beneficiaries and 13 Associated Partners, you will engage in pioneering research on:

• Smart composites and cutting-edge sensing technologies
• Advanced simulations and AI-driven data analysis
• Digital twins and Industry 4.0 innovations

Through collaborative projects, international secondments, and specialized training, you'll contribute to safer, lighter, and more efficient composite components, reducing CO₂ emissions, enhancing wind energy production, and supporting Europe's green transition.

Application procedure
Are you interested in this vacancy? Please apply no later than 28 February 2025 via the application button and upload the following documents:

• Detailed CV, including information on the candidate’s proficiency in English
• Motivational letter (1 page), describing why the position fits the applicant
• Brief research statement (one page) describing your project idea in the field of physics-informed deep learning algorithms, making connection to your experience in this area and the related work from the literature
• Copy of Master’s degree diploma
• Transcripts of all obtained degrees (in English)
• Copies of any other relevant certifications listed within CV
• One publication (e.g. thesis or preferably a conference or journal publication, providing a link to the publication is sufficient)
• Contact information of 2 references

Selection Criteria: Candidates will be evaluated based on their academic background, research experience, and alignment with the project’s objectives. The selection process will be open, transparent, merit-based, impartial, and equitable.•Interview: Shortlisted candidates will be Invited for a structured interview with predefined questions and a scoring system.

Shortlisted candidates will be invited to apply to one of the EPFL doctoral schools (e.g. EDCE or EDRS). This parallel application process is necessary to be eligible for a PhD at EPFL

You can address your application to Prof. Olga Fink.

Please note:

• You can apply online. We will not process applications sent by email and/or post.
• Please do not contact us for unsolicited services.

40 hours per week

Mekelweg 5