Title : Interoperability in Decentralized Goal Directed Building Management System
Beginning of thesis : 2020‐01‐01
End of thesis : 2022‐12‐31
Abstract : As the BIM domain undergoes a digital transformation, the deployed services must be able to react and adapt to the continuously changing environment. Given the complexity and time response constraints, the decision‐making is being delegated to building management systems, thus transforming BIM systems from being purely data‐driven to becoming goal‐directed. MASs are promising solutions to support this transformation, as agents will be able to control one or more services and to coordinate and regulate themselves.
These BIM systems are becoming increasingly complex, dynamic, and heterogeneous. However, the domain specialized, individual systems must remain interoperable to have a flexible and decentralized decision system. This interoperability implies coping with data, protocol, and service heterogeneities. In this regard, Semantic Web technologies can help foster uniformity of data formats, as well as modularization and reuse of specifications (ontologies).
While both MAS and Semantic Web technologies have been used independently in solutions within the building domain, there doesn’t exist a fully integrated infrastructure where all these proposals are interoperable. Thus, the main objective of this PhD is to propose a solution capable of leveraging the advantages of both MAS and Semantic Web technologies, by defining models and technologies for a fully decentralized goal‐directed building management system.
Keywords : Semantic Web Technologies, Multi‐Agent Systems, Web of Things, Thing Description, Building Management System
Thesis defense : 2022‐12‐31
Supervisor :
- Maxime Lefrançois, Maître assistant, Mines Saint‐Etienne, LIMOS
- Olivier Boissier : Professeur à Mines Saint-Étienne, LIMOS
Partners or/and funders : ENGIE LAB Laboratoire CSAI