For the Shipping Industry, the Information Age means the availability of ever-growing amount of data coming from multiple sensors present on-board. Sensors are nowadays embedded in most of the main new-built ship components and allow the ship owners and builders to monitor virtually every aspect of the vessel operations and use. Recent trends in the Big Data in Shipping Industry show that in the data growth will continue, and therefore there is a need to provide to ship owners and operators efficient solutions able to support them during the decision processes based on ship data statistics and evidence.
Moreover, ships are extensive scale complex engineering systems, composed of many subsystems and components. The interactions among different subsystems are complex but fundamental to understand the overall system performance. As a result, it is mandatory to integrate the information from different sources and to consider their interactions. This requires an overall system understanding and analytic skills able to model and estimate the system performance. Because of the improvement in information systems and advanced sensing technologies, abundant sensing data provide a feasible way to study this analytics process with a data-driven approach. By taking advantage of these big data, significant performance improvement and cost savings can be achieved.
The goal of this research is to enable the monitoring of the onboard equipment by exploiting heterogeneous sensors, enabling diagnosis and prognosis of the system’s components and their potential future failures. The success of this achievement hinges crucially on the capability of developing effective data-driven predictive and prognostic models. A significant challenge of this research is to model and improve the operational resilience, efficiency and safety of ships by integrating condition monitoring and lifecycle management. The challenge is to consider the interactions among different subsystems, the operation profiles, and the environment. Such interactions are complex, therefore the innovation of this proposal is to synergise the information from different subsystems to get a comprehensive view of the overall system.
Ships data analytics has some inherent challenges that need to be addressed by this research. Ships operate in a very dynamic environment; it is instrumental to consider the dynamic operating conditions. Ships are not readily available for diagnosis and repairs in case of unexpected failures, for this reason, condition based maintenance and risk assessment need to consider the relatively long lead time when performing any onboard tasks to achieve the practical yet optimal strategy.
To address these challenges, this research will investigate innovative analytic methodologies, which can integrate domain knowledge with data from different channels, to provide accurate vessel state estimation, conditional monitoring, and performance optimisation.
The main aim of this proposed PhD topic is to fill these gap by developing an innovative data-driven approach to enabling diagnosis and prognosis of the system’s components and their potential future failures.
The specific objectives can be listed as follows:
1. Develop anomaly detection models able to grasp and detect abnormal variation in the behaviour of the ship;
2. Develop analytical models able to nowcast and forecast the state of the components of the ship to plan effective and cost aware maintenance protocols;
3. Exploit the analytics model to influence the design of the ships or ship components based on their actual use;
4. Exploit the analytics models to optimise the performance of the components or their exploitation;
The R&D department of Damen Schelde Naval Shipbuilding (DSNS) will support the project with real ship operational historical data as well as relevant ship details.
Name of supervisor(s)
Dr Andrea Coraddu - University of Strathclyde, Department of Naval Architecture, Ocean and Marine Engineering.
Dr Luca Oneto – University of Genova, Department of Informatics, Bioengineering, Robotics and System Engineering.
Applicants should have a distinction pass at Master’s level in marine/mechanical engineering or a related subject, or first class BEng/BSc Honours degree, or equivalent, in marine/mechanical engineering or in a related subject.
The project requires a mixture of skills, including numerical and experimental, computer programming and statistics.
The funding covers UK/EU student tuition fees and stipend in line with University rates for 36 months.
How to apply
Applicants should send their application directly to Dr Andrea Coraddu.
E-mail: [email protected]
Applications should include:
- Cover Letter;
- CV with two references;
- Degree transcripts and certificates (in English), and if English is not your native language, a copy of your English language qualifications (IELTS).
If you wish to discuss any details of the project informally, please contact Dr Andrea Coraddu e-mail: [email protected]
Start Date and Duration: Preferably January/March 2019, for 3 years.
Application closing date: Prompt application is advised, as this position is only available until a suitable candidate is found.