‹ Back to the directory

Elie Hachem

Elie Hachem

Professor

Center · CEMEF

Awards & distinctions

  • 2025 ERC Proof of Concept Grants for its project combining virtual reality, artificial intelligence, and healthcare.
  • 2022 ERC Consolidator Grant for the “CURE” project: “Mechanical-fluid coupling and machine learning for the treatment of unruptured aneurysms.”
  • 2020 Fellow 2020 (distinction de l’International Association for Computational Mechanics)
  • 2019 1st Prize, Atos-Joseph Fourier Award, in the category of numerical simulation and high-performance computing,
  • 2015 Intel and HP Award for Best Presentation in “High-Performance Computing for Fluid Mechanics,” École Polytechnique, Palaiseau, February 5, 2015
  • 2014 Prix IBM Faculty Awardde 10 000$
  • 2010 ECCOMAS Best Thesis Award
  • 2010 GAMNI-SMAI Thesis Award

Team

CFL - Calcul Intensif et Mécanique des Fluides

Biography

Hachem Elie is a researcher whose work lies at the intersection of computational fluid dynamics, machine learning, and biomedical applications. His research focuses primarily on developing innovative methods for simulating and analyzing complex flows, with particular expertise in hybrid approaches that combine physical modeling and artificial intelligence. His contributions include the use of graph neural networks (GNNs) to accelerate computational fluid dynamics (CFD) simulations, notably through masked pre-training strategies and multiscale learning curricula. This work aims to improve the efficiency and accuracy of models while reducing computational costs, with targeted applications such as the simulation of intracranial aneurysms or the prediction of multiphase phenomena. At the same time, Hachem Elie explores fluid-structure interactions (FSI) and adaptive meshing methods, integrating multiscale variational frameworks to address problems such as thrombus formation or the topological optimization of thermofluidic systems. His research also extends to the optimization of industrial processes, such as the control of gas furnaces or the protection of solar panels against high winds, by coupling deep reinforcement learning algorithms with high-fidelity CFD simulations.

Publication(s)

Teaching

Fluid Mechanics (PI MECAERO)

Lecturer

Mechanics of Materials and Structures (PI MECAERO)

Course Director

Project (PI MECAERO)

Lecturer

Elective Course Period (October and January)

Course Director

Finite Elements

Lecturer

The course consists of a theoretical component (including lectures and small-group sessions) and a practical component based on a mini-project. The ""lecture"" component includes a presentation of the theoretical framework of the method (10 lecture and tutorial sessions), as well as its implementation (5 lecture and tutorial sessions), and the industrial context (5 sessions). The mini-projects account for one-third of the course. Students will choose their projects from a list of topics focused either on specific applications or on algorithmic or mathematical developments: dynamics, statics, contact, thermal analysis, diffusion, fluid mechanics, mesh adaptation, etc. Content Theoretical formulation and implementation of the method Variational formulation: Sobolev spaces, weak solutions, the Lax–Milgram theorem, connection to virtual work and the calculus of variations. Finite Element Method: Description of the method, examples of finite elements, convergence results. Matrix formulation: Elementary matrices, assembly, boundary conditions, solution, algorithms in mechanics and thermal analysis. Computational environment Connection to CAD: Design workflows. Meshing techniques: Delaunay methods, frontal meshing. Meshing a complex part. Major commercial software packages: Organization of a computational code, some examples. Parallel computing: Machines and associated algorithms, parallel solvers, domain decomposition. Applications Students interested in applications involving real parts will use a major commercial software package or software from the School’s Research Centers. They will perform a calculation that is realistic from an industrial perspective. Those more interested in applied mathematics can use a development platform, such as FreeFem++.

Digital Engineering of Complex Systems (IDSC) track

Course Director

Over the past decade, artificial intelligence (AI) has profoundly transformed the industrial world by revolutionizing the ways in which systems are designed, optimized, and operated. One of the most striking aspects of this evolution is the digital modeling of complex systems. Traditionally based on methods with a strong physical or multiphysical focus, this field now benefits from advances in artificial intelligence, particularly through machine learning and deep learning algorithms. The Digital Engineering of Complex Systems track at Mines Paris aims to provide students with a solid, multidisciplinary foundation in AI, enabling them to gain a deep understanding of its applications and apply them to real-world industrial challenges. The disciplines studied in the specialization are: machine learning and its applications in engineering, computer vision, and high-performance computing. The specialization’s curriculum is structured around three weeks of coursework, supplemented by two weeks of a Data Challenge and a one-week field trip. More information is available on the track’s webpage: https://bruno-figliuzzi.github.io/Option/

Fluids (Course)

Course Director

Fluids (Research Quarter)

Course Director

PhD supervision

  • 2025 Simulation of the pneumatic wringing process in galvanizing, based on a finite element model with anisotropic space-time adaptation. NASSER Ali
  • 2025 Digital twins and deep learning for optimized and decarbonized design of industrial glass furnaces AYOUN Yassine
  • 2025 Machine Learning Coupling with Numerical Mechanics via Graph Neural Networks BAZZI Amir
  • 2025 Advanced simulation of polymer quenching for sustainable aerospace manufacturing CHAKIB Amal
  • 2025 Multiphase flow simulation for the transport of complex fluids in the human pulmonary system. YEPES-PEñARANDA Alejandro José
  • 2025 Acceleration of numerical simulations in mechanical engineering by artificial intelligence: towards an integration of Graph Neural Networks in forging processes. SHEHAZI Manar
  • 2025 Digital twins and deep learning for optimized and decarbonized design of industrial glass furnaces STENTA Marion
  • 2024 Artificial intelligence and digital mechanics coupling for the treatment of intracranial aneurysms ZARATE GUEVARA John Sebastian
  • 2024 Numerical fluid mechanics, 4D flow MRI, and deep learning for risk stratification of intracranial aneurysm rupture LANNELONGUE Vincent
  • 2024 Coupling Deep Reinforcement Learning (DRL) methods with graph convolutional neural networks for flow simulation and optimization NGUYEN Xuan Minh Vuong
  • 2024 Aneurysm Treatment by Machine Learning GARNIER Paul
  • 2024 Deep reinforcement learning for innovative design and development of digital twins for packaging molds ZAAYTER Tony
  • 2024 AI coupling and digital mechanics for the treatment of intracranial aneurysms. FAITINI Chiara
  • 2024 Composite hydrogels for dual drug delivery VINCENT Carla
  • 2024 Spatial qualification of ceramic components for satellites using numerical simulation of thermal shock tests. ABDEL GHANI Hasan
  • 2023 Reduction of drag induced by biopolymers in turbulent flows ELKHOURY Ricardo
  • 2023 Artificial intelligence and digital mechanics coupling for the treatment of intracranial aneurysms PELISSIER Ugo
  • 2023 Artificial Intelligence and Digital Twins in Metallurgy - Modeling Front-Tracking of Evolving Interface Networks. YOUNES Eliane
  • 2022 Towards a Digital Twin of a Photovoltaic Power Plant MICHEL Théodore
  • 2022 Numerical simulation of the aorta using the finite element method and application of optimization methods for improving a ventricular assist device RENAULT Maxime
  • 2022 Anisotropic mesh adaptation for multiphase flow dynamics with wetting and capillary effects in coating applications GAULTIER Clément
  • 2022 A partitioned fluid–structure interaction numerical framework for the simulation of intracranial aneurysms with hyperelastic wall behavior RBAH Abdelilah
  • 2021 Deep Learning Application to Prediction and Modeling of Multiphysics Fluid Flows EL HABER George
  • 2021 ARRAY(0x841b2b6a8) ISUKWEM Kindness
  • 2021 Multi-scale modeling of interfacial instabilities and bubble dynamics: application to filling flows in the lost foam casting process. EL ZAHABI Jennifer
  • 2021 Analysis of the impact of arterial wall motion on blood flow in intracranial aneurysms through fluid-structure interaction simulations GOETZ Aurèle
  • 2021 Advanced Hemodynamic Modeling and Simulation of Flow-Diverter Stents in the Treatment of Intracranial Aneurysms JEKEN RICO Pablo
  • 2020 High-fidelity numerical modeling and simulation of the Lost Foam process HAYEK Cynthia
  • 2020 Topological Optimization of High-Efficiency Heat Exchangers Using the Level-Set Method and Anisotropic Mesh Adaptation ABDEL NOUR Wassim
  • 2020 Correction flux strategies for continuous finite element approximations of the compressible Euler equations on unstructured meshes DEVOS Thibaut
  • 2020 Models and simulation of acoustic-thermomechanical coupling in complex fluids BOUTHIER Louis
  • 2020 Machine learning and compliance with physical laws: application to seismic imaging DE SOUZA Léo
  • 2019 On the coupling of deep reinforcement learning and computational fluid dynamics GHRAIEB Hassan
  • 2019 Modeling of fluid-solid coupling with analysis and optimization of residual and thermoelastoplastic stresses KHALIL Joe
  • 2019 A posteriori error estimation and adaptive control of iterative finite element solvers with mesh adaptation: application to the quenching process. MEDGHOUL Ghaniyya
  • 2019 Numerical and parallel modeling of an anisotropic adaptive mesh for industrial quenching applications EL AOUAD Sacha
  • 2019 Turbulence modeling in fluids assisted by deep learning PATIL Aakash
  • 2018 Numerical and experimental study of boiling phenomena - Application to quenching BRISSOT Charles
  • 2018 An adaptive immersed mesh method (AIMM) for fluid-structure interaction (FSI) NEMER Ramy
  • 2018 Rapid and flexible heat transfer by radiation for sintering applications GERARD Rémi
  • 2018 New modeling framework for interface capture and boiling for industrial cooling BOUBAYA Ali-Malek
  • 2018 Convolutional neural networks for steady flow prediction around 2D obstacles CHEN Junfeng
  • 2017 Optimal resolution of iterative solvers with anisotropic adaptive meshing MANZINALI Gabriel
  • 2017 Finite element methods for simulating radiative heat transfer in industrial furnaces MENSAH ASSIAKOLEY Tevi
  • 2017 Study and development of the performance of the Ranque-Hilsch tube and analysis of industrialization prospects ASSEMIEN Fian
  • 2016 Adaptive variational finite element formulation and massively parallel computing for industrial aerothermal applications. BAZILE Alban
  • 2016 A reliable and adaptive CFD framework for airship design GUIZA Ghalia
  • 2016 Numerical modeling of the solidification of molten steels under levitation in the International Space Station AALILIJA Ayoub
  • 2016 Advanced numerical methods for simulating industrial quenching processes BAHBAH Chahrazade
  • 2016 ANR HECO Contract: from heating to cooling: load monitoring for thermal processing. DALDOUL Wafa
  • 2015 Advanced numerical methods for simulating industrial quenching processes AGARWAL AMAN