Doctoral thesis: development of physics-informed machine learning model for machining processes

Physics-informed machine learning is a rapidly growing area within Scientific Machine Learning (SciML) that integrates physical laws with machine learning and deep learning techniques. This integration is inherently bi-directional: physical principles, such as conservation laws, governing equations, and domain‑specific knowledge, are embedded into artificial intelligence (AI) models to enhance their accuracy, robustness, and interpretability, while AI methods can, in turn, assist in identifying governing equations and unknown model parameters, thereby deepening our understanding of complex physical systems. As a result, physics‑informed machine learning enables high‑fidelity predictions with reduced data requirements and provides efficient tools for solving challenging differential equations.

The objective of this PhD project is to leverage physics‑informed machine learning models for their application in machining processes, including turning, drilling, and broaching. These models will be developed by combining experimental data with analytical and finite element models, with the aim of accurately predicting cutting forces, cutting tool temperatures, and key industrial outcomes such as tool life and surface integrity.

Tool wear prediction plays a crucial role in ensuring the reliability and efficiency of machining operations due to its wide-ranging industrial applications. Similar considerations apply to surface integrity, which is directly linked to component performance and durability. Although significant research efforts have been devoted to tool wear and surface integrity prediction, achieving high prediction accuracy for these critical industrial outcomes remains a major challenge. Addressing this gap represents a key motivation for the proposed research.

The project will initially focus on orthogonal cutting, providing a well‑controlled framework to develop, test, and validate the proposed methodologies. The knowledge and modelling strategies gained will then be extended to fully three‑dimensional machining processes, enabling broader industrial applicability and a stronger technological impact.

Research team: Mecanizado de Alto Rendimiento

PhD Supervisor: Pedro J. Arrazola

Starting date: March 2026.

• www.mondragon.edu/en/faculty-of-engineering
• www.mondragon.edu/mar

• Personal Researcher in Training contract associated with a pre‑doctoral grant. See grant conditions: https://www.mondragon.edu/es/doctorado-ingenieria-aplicada/precios-becas...
• Duration: 4 years
• Grant amount: the grant amounts for each year are indicated below.
• • DOC1 y DOC2: 22.809,77€ brutos
  • DOC3: 24.439,04€ brutos
  • DOC4: 30.548,80€ brutos

The beneficiaries of this aid will not pay any amount for enrollment in the doctoral programme or for reading theses.

We are looking for proactive people who are agents of change; people who are committed and involved in reality, who seek to transform it, giving the best of themselves.

• Educational requirements:
  • Master's degree in industrial/mechanical engineering
• Specific knowledge:
  • Materials
  • Manufacturing
  • Mechanical design
• Languages:
  • English (minimum B2 or similar, C1 will be valued)
• The following will be valued:
  • Experience in previous research projects related to materials, manufacturing, and mechanical design.
• Other:
  • The thesis will be carried out in line with the Horizon Europe EuroFMX project.
  • At least two articles must be published in Q1 journals and one article in a conference.
  • It would be advisable for the student to attend some machining classes from the master’s degree in industrial engineering to acquire the necessary knowledge for the development of the doctoral thesis.