Project 1: Big data for shipyard 4.0: Data stream mining
The project is aimed to perform mining of data coming from devices like sensors and buoys, which have great relevance with quality assurance in the ship building process. Specific research issues include: (i) analysis of feature- and distance-based methods for effective clustering/classification of multidimensional time series data streams, (ii) identification and detection of anomaly behaviours, and (iii) introduction of techniques robust to the presence of noises/imprecise readings.
Project 2: Streaming big data analytics for Shipyard 4.0
Streaming big data generated by smart sensors can be exploited to trigger alarms for emergency situations and events in the manufacturing process. A challenge in analysing the streaming sensor data is the lack of semantics in the data streams. The project aims to design integrated technologies for decreasing the costs for ship monitoring, assessment and maintenance by collecting data from devices like sensors and buoys, processing and uploading to the data lake.
Project 3: The Digital Twin for the Evaluation of Ship Safety
The main objective of this PhD would be to develop the main points needed to set up a Digital Twin with the main goal of analysing and predicting the ship behaviour after damage. This digital twin should receive data from the real vessel and the environment, and generate information regarding the status of the ship, which would be of great interest for the ship master as a tool to take data-based decisions and which would help in the case of an accident.