INTEGRATIVE DATA PROCESSING FOR CONNECTED CYBER-PHYSICAL OFF-SITE AND ON-SITE CONSTRUCTION
A significant obstacle concerning the aspired co-design approach is a centralized data management.
Even if this obstacle has been targeted from different perspectives in the past, utilizing data models and formats in the domain building engineering as well as from the domain of mechanical and plant engineering, it has not been solved till today. The models represent the flow alongside the process chain. However, a backflow of information is out of scope for most approaches.
This research project aims at addressing this shortcoming by creating a central data management platform with a standardized syntax and semantic that is shared throughout the whole process chain and supports interoperability between all stages, demonstrated by a codesign approach for an adaptive production system.
To identify the relevant data sources and formats, data acquisition is part of the requirements definition. Based on the results of this acquisition, we define the software architecture of the overall data management framework alongside with its internal data model. We further incorporate semantics into the framework and devise proper interfaces for relevant data formats. The data mapped into the central data management platform serves as a basis for algorithms that analyze the deviation of a monitored state from the anticipated state to then allow for the implementation of a feedback loop to devise an adaptive production system.
Prof. Dr. rer. nat. Melanie Herschel
Institute for Parallel and Distributed Systems (IPVS), University of Stuttgart
Prof. Dr.-Ing. Alexander Verl
Institute for Control Engineering of Machine Tools and Manufacturing Units (ISW), University of Stuttgart
- Gazzarri, L., & Herschel, M. (2021). End-to-end Task Based Parallelization for Entity Resolution on Dynamic Data (to appear). Proceedings of the IEEE International Conference on Data Engineering (ICDE).
- Lässig, N., Oppold, S., & Herschel, M. (2021). Using FALCES against bias in automated decisions by integrating fairness in dynamic model ensembles (to appear). In Proceedings of Database Systems for Business, Technology, and Web (BTW).
- Diestelkämper, R., & Herschel, M. (2019). Capturing and Querying Structural Provenance in Spark with Pebble. In ACM International Conference on Management of Data (SIGMOD), 1893–1896. https://doi.org/10.1145/3299869.3320225