Digital-Twin-Based Real-time Approaches for Sustainable Scheduling

In the current context of the industry, the convergence of technological advancements and
sustainability imperatives has transformed how companies plan and organize their operations.
These processes rely on static models and heuristic approaches using a comprehensive set of
data. However, in the face of increasing complexity in supply chains, it has become essential to
make decisions while considering uncertainty, with partial information or unreliable
assumptions about the future. Digital twins have emerged as a tool providing real-time insights
through virtual replicas of physical systems, enabling dynamic decision-making to optimize
operations and anticipate disruptions.
This thesis proposes new approaches for real-time sustainable planning, leveraging the
capabilities of digital twins and advanced optimization techniques. It aims to develop algorithms
capable of adapting to changing industrial conditions while incorporating sustainability
constraints, such as energy consumption or CO2 emissions. This thesis particularly focuses on
the feedback loop between digital twins and decision-making processes, allowing for continuous
improvement of scheduling strategies and better alignment with sustainability goals.