Activities in today’s society are supported by complex systems organised into networks such as logistics, transportation, communication, healthcare, and humanitarian aid. As these systems and tools become increasingly complex, their intelligent management is essential, requiring analytical, planning, and operational management methods to ensure they operate efficiently and to their full potential. Our researchers develop analytical and decision-making intelligence to enhance the informational intelligence of complex systems. Our methodological approach intersects predictive analytics (artificial intelligence), operations research (OR), operations management (OM), and information technology (IT).
Our researchers are active in different application areas, in which they contribute their expertise which is structured by our two main research areas. The following table shows the complementarity between our the research interests of our regular members:
Application areas
Our research in applications is coordinated by Prof. Anaya-Arenas. We develop intelligent analysis and decision-making tools in the following areas:
- Intelligent transport systems
- Planning of health services
- Service delivery in humanitarian aid and disaster response
- Management of connected, resilient and sustainable supply chains
- Digitisation in business (connected shops and manufacturing)
- Smart cities
Our methodological research is divided into two main research areas as follows.
Research Area 1: Methodologies
Our research in this area, coordinated by Prof. Rei, concerns the development and application of models and methods to entire classes of problems and their applications. In particular:
Theme 1.1 – Modelling and solving complex planning problems [Anaya-Arenas, Cherkesly, Crainic, Djeumou, Gruson, Jena, Ortmann, Rei, Zbib] : This theme focuses on modelling decision-making systems and problems using mathematical programming, as well as solving them (i.e. identifying optimal planning).
Theme 1.2 – Representation and treatment of uncertainty to increase the resilience of systems [Anaya-Arenas, Crainic, Jena, Ortmann, Quesnel, Rei, Zbib]: This theme aims to characterise and reduce uncertainty and includes the integration of machine learning in decision-making under uncertainty (stochastic and robust optimisation).
Theme 1.3 – Empirical research [Bendavid, Jena, Lachapelle, Ortmann, Paulhiac, Quesnel]: The creation of procedures for the capture (Internet of Things), collection and processing of data (quantitative and qualitative), are the core of this theme. This covers data obtained through interviews, surveys and publicly available data.
Theme 1.4- Extracting information from large quantities of data. [Bendavid, Jena, Maïzi, Ortmann, Rei]. The large volume of data available creates a challenge for its use in management systems. This theme aims to advance knowledge on the integration of learning methods (e.g. data mining methods) into planning models, the collection of large quantities of data (often in real time) from various sources, and the rapid processing of data. The aim is to facilitate the use of information flows to monitor operations and system performance in order to forecast and adjust planning to changes.
Theme 1.5 – “Living lab” research methods [Bendavid, Gruson, Maïzi], This research takes place in a collaborative environment where users participate in innovation by cooperating with other stakeholders (suppliers of technological solutions; service providers, researchers, customers, etc.) in a neutral and controlled research environment to understand synergies, design and develop complex systems.
Research Area 2: Tools for experiments and validation
In this research area, coordinated by Prof. Bendavid, we develop tools for experimentation, validation, partner training, transfer to partners and implementation in real processes. A large part of the work done in this Research Area is done in the IoT lab. This research is structured in two main themes.
Theme 2.1 – Digital simulation [Anaya-Arenas, Bendavid, Jena, Maïzi, Ortmann]. This theme aims to create numerical tools for the representation and simulation of complex systems. This is essential for validating the models created in Research Area 1, for demonstrating the results, for technology transfer and for our partners to learn the methods.
Theme 2.2 – Prototypes and pilots [Bendavid, Maïzi, Paulhiac] are developed by researchers, students and professionals with a view to transferring the results obtained to users and partners. This stage requires an interdisciplinary approach that takes into account the applicative, technological, political and social dimensions.