Projects and Collaborations

The Mathematical and Industrial Engineering (GMI) department develops research projects in optimization, modeling, and complex systems analysis, in partnership with academic and industrial stakeholders. This work aims to improve industrial, logistical, and energy performance by integrating innovative approaches in artificial intelligence, simulation, and decision-making. This page presents the department’s projects and collaborations.

Current Projects

IADAP

AI for Precision Anomaly Detection

AURA Innovation Partnerships funded by the Region

In this project, we plan to develop a proof of concept based on artificial intelligence algorithms dedicated to image analysis, with the aim of detecting dimensional visual anomalies on the order of a few microns, using AI operating at the pixel level. The project will include a design, construction, and implementation phase for a test bench to validate the system’s performance under conditions representative of its future industrial environment.

The system will be based on the use of a high-resolution camera coupled with a precision lens, a high-performance industrial PC for data processing, and a conveyor allowing image capture under real production conditions. A scientific development phase for the algorithms will be conducted to optimize the model’s precision and robustness. The developed algorithms will then be embedded in the machine assembly (conveyor + camera + PC) to test their performance in real time, with a continuous flow of parts, in a logic of constant learning and adaptation. To ensure the transfer and valorization of results, the project will rely on the industrial platforms of the École des Mines de Saint-Étienne. The IT’M Factory platform will be mobilized for algorithm development, test bench implementation, and system optimization.

The envisioned artificial intelligence will be unsupervised, capable of self-learning, with a configuration adaptable to different types of parts. The objective is to achieve pixel-level detection precision, while ensuring robustness against the natural variability of machined parts, and facilitating the industrial integration of the system.

Dates

  • October 2025 – October 2026

Partners

  • Liaison R

Contact

Keywords or themes

Anomaly detection, Unsupervised AI, Micrometric precision, Industrial vision.

Sustainable Development Goal

ANR ReCircle

Remanufacturing and Circular Economy Logistics

The ReCircle (Remanufacturing and Circular Economy Logistics) project aims to propose optimization tools in the context of a supply chain for remanufacturing, addressing several levels.

  • Dynamic optimization of the product rehabilitation system, which receives batches of uncertain volumes and compositions, and for which operational sequences are discovered progressively during part reprocessing, restricting visibility on upcoming operations and thus increasing cognitive load (stress) on operators.
  • Dynamic optimization of collection and delivery routes for parts to be reprocessed, as well as the connection with the inventory of each reprocessing site. Indeed, when a part cannot be repaired, its components can be stored, either for future use or to be sent to another site with the competence to process the part. This creates a high burden in terms of inventory level management across different product families. It is therefore important that parts be collected as early as possible to avoid overloading a site’s storage area, without multiplying routes, which could have a negative environmental impact.
  • Optimization of the integrated problem, in its proactive form, simultaneously considering the scheduling of the remanufacturing system—taking into account a level of knowledge about the operations to be performed, and vehicle routes aimed at supplying inventories of products to be rehabilitated;
  • Dynamic optimization of the integrated problem, establishing the foundations for an optimization tool that can be integrated into a digital twin.
  • Integration of the designed optimization modules within a digital twin, to provide a proof of concept for the applicability of the methods and models developed within a remanufacturing-oriented supply chain.
A project labeled by Minalogic

Dates

  • 03/16/2026 – 03/15/2030

Partners

  • Mines Saint-Étienne (LIMOS laboratory),
  • Université Clermont Auvergne (LIMOS laboratory),
  • IMT Nord Europe (CERI-SN laboratory)

Contact

  • Participants
    Xavier Delorme
    Frédéric Grimaud
    Arthur Kramer
    Paolo Gianessi

Keywords or themes

Anomaly detection, Unsupervised AI, Micrometric precision, Industrial vision.

Sustainable Development Goal

CIROQUO CONSORTIUM

The CIROQUO Consortium (Industry-Research Consortium for Optimization and QUantification of uncertainty for Expensive data) was created to bring together partners from academic and technological research to solve problems related to the use of numerical simulators. These problems include code transposition (how to move from small-scale to large-scale simulations when only small-scale simulations are feasible), accounting for uncertainties affecting simulation results, and the validation and calibration of calculation codes based on collected experimental data.

This project was born from a simple observation: industries using large-scale calculation codes often face similar difficulties, despite the diversity of their application fields. Although current computing servers are increasingly powerful, the growing complexity of simulations means that computing times are often in the range of several hours, or even a full day. In practice, this limits the number of simulations that can be performed.

Dates

  • 2020-12-01 – 2028-11-30
Website

Partners

  • Centrale Lyon,
  • IRSN
  • BRGM
  • Stellantis

Contacts

  • Participants
    • Didier Rullière
    • Tanguy Appriou
    • Charlie Sire

Keywords or themes:

Industry, applied mathematics, optimization, uncertainty, expensive data

Sustainable Development Goal

Publications

News

ISI FLEET

Development of a Fleet Management System solution, RD Booster AURA

Research and innovation project concerning the development of a tool for managing and optimizing the use of all automated internal transport means (notably a heterogeneous fleet of AGVs and AMRs) on a site.

The project studies the management and optimization of Automated Guided Vehicles (AGV) or Autonomous Mobile Robots (AMR) in Flexible Manufacturing Systems (FMS). Flexible Manufacturing Systems (FMS) equipped with AGVs and/or AMRs have been the subject of intensive research for many years by the international scientific community.

Dates

  • 2023-10-01 – 2026-06-30

Partners

  • ISITEC International
  • MECACONCEPT

Contacts

  • Participants
    • Damien Lamy
    • Arthur Kramer

Keywords or themes

  • Industry, logistics, optimization

Sustainable Development Goal

AURA- PSCP- Easy Smart Factory 2

The objective of the project is to accelerate the digitalization of production workshops, particularly those of small structures, through several major technological innovations (software, electronics, and Telecom innovations). For this project, complementary stakeholders—from IIoT to software—have grouped together to develop a solution with a common goal of making it “simple” and adapted to SMEs. The aim is to develop an innovative solution (hardware, software, service) that is simple to use and will significantly accelerate the digital transformation of medium-sized industrial workshops (manufacturing and packaging workshops) by: (1) Reducing implementation and infrastructure costs (autonomy and simplicity of access to technology), (2) Reducing operating costs (Solution As A Service rather than investment), (3) Reducing electricity consumption (shared infrastructure). A global Solution As A Service (100% cloud MES software and hybrid 5G IIoT hardware representing a technological breakthrough), EASY SMART FACTORY will allow a workshop to be digitalized quickly and independently. Depending on their needs, the manufacturer will configure the different blocks of their solution on an e-shop and then receive the hardware. With the software interfaced with the ERP, they will have all the real-time data from their workshop to share with their teams to improve competitiveness.

Dates

  • 2020-11-03 – 2026-03-26

Partners

  • Astrée Software
  • Editag
  • Eurécom

Contacts

  • Participants
    • Damien Lamy
    • Ehsan Yadegari

Keywords or themes:

Industry, digitalization, optimization

Sustainable Development Goal:

Digital Twins for Industrial Production Systems Sponsorship Chair

The general objective of the Chair is to develop, in close connection between the Parties, international-level research dedicated to the theme of Digital Twins for Industrial Production Systems.

In this context, the main research axes of the Chair are:

  • Development of methods for coupling simulation models and Artificial Intelligence for Digital Twins.
  • Development of an approach for digital twin engineering: interoperability, usage, and maintenance.
  • Development of holistic modeling methods and rapid implementation of Digital Twins for inter- and intra-company production systems.

Dates

  • 2023-06-01 – 2026-05-31

Partners

Contacts

  • Participants
    • Damien Lamy
    • Xavier Delorme

Keywords or themes:

Industry, digitalization, optimization

Sustainable Development Goal:

News

ANR ReConFidurable 

Design and management of reconfigurable and sustainable production systems

Reconfigurable Manufacturing Systems (RMS) are not just systems with customized flexibility; they can be a basis for developing sustainable production systems. The objective of this project is to implement an effective methodology for integrating sustainable development criteria into the design and reconfiguration of RMS. The approach is based on the principle of RMS modularity. It will involve choosing equipment modules to use and assigning production operations to these modules while taking into account demand, types of products to be manufactured, and constraints. The 3 steps to consider are: design, reconfiguration, and real-time management. We will integrate sustainable development criteria into the models for all 3 steps.

Dates

  • 2022-01-01 – 2025-12-31
Website

Partners

  • IMT Atlantique
  • Aix-Marseille University
  • Mines Saint-Etienne
  • ENSAM Metz
  • Automatique Industrie
  • Kedge Business School

Contacts

  • Participant
    • Damien Lamy

Keywords or themes

Industry, reconfiguration, management, optimization, sustainable development

Sustainable Development Goals

Publications

News

IMD-SMART DIAGNOSTICS, CARNOT

Multimodal data interpolation for smart diagnostics and inference – Applied intelligence for decision-making in a dynamic system

This project addresses several major scientific challenges.

  • Heterogeneity of multimodal data: The coherent integration of data from different modalities, each with distinct structures and formats, constitutes a major challenge. This requires innovative solutions to ensure compatibility and convergence of information from various sources.
  • Learning with limited data: Obtaining annotated data, which is expensive but essential for training multimodal models, is a critical difficulty. This context highlights the need to develop effective approaches in environments where resources are limited.
  • Robustness and reliability of models: Model reliability is often compromised by variations and disturbances inherent in industrial environments. This demands innovative strategies to ensure stable and consistent performance, even in the presence of changing conditions.

This project proposes to meet these challenges to improve decision-making in complex dynamic systems.

Dates

  • 2024-02-01 – 2025-09-30

Contacts

Keywords or themes

Data heterogeneity, multimodal models, smart diagnostics, decision-making

Sustainable Development Goal

THUASNE

Optimizing quality controls in the manufacturing of medical textiles by exploiting the various data available in the production chain.
Specifically, it will involve detecting, from this set of available data and information, the elements that allow for tracking potential defects based on the root causes of production defects.
The optimization of these selected quality controls will be based on available data and will rely on the development and implementation of statistical and machine learning techniques with the support of artificial intelligence.

Dates

  • 2024-09-01 – 2025-08-31

Contacts

  • Participants
    • Anis Hoayek

Keywords or themes

Data processing, optimization, quality control

Sustainable Development Goal

AURA Booster-Innovation Industry 4.0- SmartFlow

The objective, through the collaboration of two innovative SMEs from the AURA region (WIPSIM and InfoDream) and LIMOS, is to work on the interoperability and digital continuity of their SmartWip and Qual@xy solutions and to boost the competitive differentiation of the whole by adding Machine Learning methods, with a view to marketing a new product providing operational decision support for production flows in a manufacturing workshop.

Dates

  • 2022-10-01 – 2025-06-30

Partners

  • WIPSIM
  • INFODREAM

Contacts

  • Participants
    • Mireille Batton-Hubert

Keywords or themes

  • Data processing, interoperability, Machine Learning

Sustainable Development Goal

CORENSTOCK Chaire ANR ELM Leblanc

The objective of this CORENSTOCK industrial chair is to provide solutions for optimizing the energy impact of equipment on its global value chain, in a context of energy, economic, and digital transition, by considering each of its life phases: design, industrialization, use, and end-of-life of the equipment.

Dates

  • 2021-01-01 – 2025-10-09
Website

Partner

  • IMT Nord Europe

Contacts

  • Participants
    • Mireille Batton-Hubert
    • Xavier Boucher
    • Damien Lamy

Keywords or themes:

Optimization, multi-line scheduling, energy efficiency

Sustainable Development Goal

Publications

News

Research