Thesis Start: 2023
Thesis End: 2026
Expected Defense Date: January 2026
Abstract
The integration of digital technologies with physical systems has given rise to Cyber-Physical Systems, where computational components communicate with each other and interact with the physical world via sensors and actuators in a feedback loop. In this context, Digital Twins (DTs) have emerged as a solution for integrating the digital modeling of an entity with its physical dimension. By considering real-time synchronization between these two dimensions, DTs enable monitoring and control of physical assets. However, despite their widespread use as decision-making tools in various applications, current DTs have a limitation: they lack autonomous decision-making capabilities and rely on external entities to determine actions to be taken to control their physical counterparts. This dependence limits their potential in systems where the ability to react quickly to unforeseen events and adapt in real-time is crucial for maintaining efficiency – for example, in urban logistics, where a delay in decision-making when faced with an unexpected traffic jam can impact the entire delivery chain.
This thesis project therefore aims to develop a new generation of Digital Twins endowed with their own decision-making process and participating in a decentralization of decision-making among the different entities of the system. It relies on multi-agent systems to strengthen decision-making autonomy and on knowledge graphs to structure data representation and promote interoperability among different stakeholders.
The field of urban logistics, and more specifically last-mile delivery, will serve as a case study to validate this approach. Last-mile delivery faces major challenges in dense cities such as road congestion, access restrictions, the growing volume of deliveries, and the consideration of environmental constraints. The integration of autonomous digital twins would not only allow real-time monitoring of physical assets (e.g., delivery truck) but also decentralized management of operations with a real-time reaction capability to disruptions.
Keywords
Digital Twins, Multi-Agent System, Knowledge Graph
Partners and/or Funders
Orange
Sustainable Development Goals concerned
Publications
- In recent years, the application of Digital Twins (DTs) in Cyber-Physical Systems is becoming increasingly relevant, as they enable real-time synchronization between the digital and physical dimensions. They have been used as decision support tools in a wide range of applications, as they enable real-time monitoring and control of assets in the physical world. However, […]
- […]
- […]