PhD ''Temporal Behavioral Models of Digital Twins for Networks' F/HOrange

Orange Innovation brings together the research and innovation activities and expertise of the Group's entities and countries. We work every day to ensure that Orange is recognized as an innovative operator by its customers and we create value for the Group and the Brand in each of our projects. With 720 researchers, thousands of marketers, developers, designers and data analysts, it is the expertise of our 6,000 employees that fuels this ambition every day. Orange Innovation anticipates technological breakthroughs and supports the Group's countries and entities in making the best technological choices to meet the needs of our consumer and business customers.
Within the Innovation department, you will be integrated into a research team at the forefront of innovation and expertise in future networks and their modeling within a Digital Twin platform. This team is highly active within the internal and external Orange innovation ecosystem, notably through collaborative projects such as ANR and European initiatives.

Digital twins (DTs) for networks are part of Orange's strategic technological trends, notably aimed at achieving Level 5 of autonomous networks [Richard, 2019]. For Orange, the continuous evolution of networks introduces increasing management complexity of physical entities (core optical/mobile networks), their physical environment (decentralized, shared), and software components. This complexity results in challenges related to automation and understanding of network behaviors.
To address this evolution, a holistic view of networks and their behaviors is essential. Network digital twins (NDTs) are seen as the central enabler to provide this comprehensive perspective, thereby supporting the capabilities of next-generation autonomous networks: monitoring, simulation, configuration, orchestration of heterogeneous networks, impact analysis across different network layers, anomaly detection, etc.
In addition to reflecting the current state of a network, an NDT can be used for simulations or predictive analyses [Raza et al., 2025]. To achieve this, a behavioral model is employed to represent the system's behavior, including actions of individual entities (state changes) and their interactions [Gill et al., 2025]. Various models exist for creating behavioral representations, such as temporal automata, Petri nets, and state machines [Gill et al., 2025]. These models are often complex to implement; recent studies suggest that behavioral models can be inferred from real-time system data [Cornanguer, 2023; Westermann, 2023].
The objective of this thesis is to develop a behavioral model of a network and leverage this model to address specific challenges. The first step will be to identify an appropriate behavioral model based on the use case. This model will be constructed from data generated by the network using machine learning techniques. One key challenge will be to utilize the network's digital twin and its modeling as a knowledge graph. The knowledge graph, which is historized and modeled through ontologies, represents the network's state. An avenue of investigation will be to assess the relevance of ontologies in modeling the behavioral model [Gill et al., 2025]. The second challenge will be to correlate the behavioral model with the digital twin's knowledge graph-that is, to establish correspondences between the two graphs and to leverage the behavioral graph to correct errors in the knowledge graph (such as network alarms).

Skills (scientific and technical) and personal qualities required for the position
You hold a Master's degree or engineering diploma.
You are perseverant and rigorous.
You have knowledge in Machine Learning and Semantic Web.
You have experience in Python, Java, and Semantic Web technologies.
You are fluent in French and English.
Educational background
You possess an Engineering diploma or Master's degree in Computer Science.
Research areas include Artificial Intelligence and Semantic Web.
Desired experiences
Prior research internship experience and a career ambition in R&D.
Programming skills.

La rémunération brute proposée est comprise entre 37 et 40 kEUR. A cela s'ajoutent un plan d'épargne entreprise et retraite, l'intéressement, la participation, une couverture santé et prévoyance, des réductions sur les offres et produits d'Orange ainsi que les activités sociales et culturelles proposées par le comité d'entreprise.