With the explosion of new sensors and IoT devices on ships, more data is being generated than ever before, but little of that data is actually being put to good use. According to surveys of ship managers by Sea Asia, less than five percent of all shipboard-generated data is analyzed—and only a small portion of that five percent is analyzed in a meaningful way. Shipowners, machinery OEMs, and regulatory entities are increasingly interested in harnessing the value of data being collected onboard ships, but are often less certain of how to do so.
Consider these “what ifs:”
● What if mechanical anomalies could be detected in real time?
● What if that detection was so granular that it could categorize those anomalies into minor, intermediate, or serious?
● What if those anomalies could be shown in 3D, displaying exactly which component inside the machine was causing the anomaly?
● What if automated model building could be used to predict when failure might occur?
● What if your classification society created a notation for vessels with cognitive analytics tools, and based on that notation, your vessel was granted an extension of time for a survey?
Machine learning and cognitive analytics tools provide all of these capabilities, detecting anomalies in machine operations and predicting failure with high degrees of confidence. Machine learning is a game-changer for the maritime sector. What’s more, the ability to apply machine learning tools to shipboard-generated data is now more widely available than ever before.
The progression of analytics from the descriptive (what happened?) to diagnostic (why did it happen?) to predictive and prescriptive analytics (when is it likely to happen again—and what can I do to prevent it?) is changing the way the industry can harness the value of machine learning tools in data analysis.
Machine learning is the latest evolution of data analytics: from logging “OK” into a logbook in the days of yore, to sensing data and setting alarms at a central console, to transmitting data ashore for management’s use, to the present day—using tools enhanced with artificial intelligence (AI) to detect anomalies and predict failure.
Additionally, machine learning expands the toolbox for fleet managers to reduce unplanned out-of-service time, protect against malicious threats, and provide cognitive query of relevant vessel-operating information from a variety of sources. This further allows for savings in maintenance and capital cost replacements, extending the life of critical shipboard assets, and extending the marine engineering knowledge base to new employees.
Operators in the maritime space are increasingly interested in this new technology for three reasons:
1) Cognitive analytics provide the capability to ingest the terabytes of data that are already being generated, and find the insights contained within to save money and reduce off-hire
2) Cognitive analytics allow more intelligent planning of major maintenance periods such as special surveys and drydock periods, spare parts and consumables inventories, and support to seagoing staff in assessing whether maintenance needs to be performed in-voyage, in a port turnaround, or over a longer period of time
3) Shipowners and operators find value in developing a deeper understanding of shipboard machines rather than leaving it to yard periods or warranty and insurance claims
Dependency on a new technology to understand the behavior of machines is a paradigm shift away from thinking about ships. Lloyd’s Register, in its “Global Marine Technology Trends 2030,” estimated a 4,300% increase in the annual data generated by ships by 2020, and says that “by 2030, that figure will have increased even further as this is an accelerating trend.”
In his address to the SNAME Maritime Convention in October 2017, the CEO of the American Bureau of Shipping remarked that understanding and managing data has become more important than ever. Data itself can now be considered a new class of “asset.” And if shipowners look after the asset that is their data, their data will help look after everything else.