Wärtsilä Expert Insight, part of Wärtsilä’s portfolio of Lifecycle Agreements, uses cutting-edge artificial intelligence and advanced diagnostics to help vessel owners improve asset efficiency and reduce both operating costs
and emissions. With the help of a Wärtsilä expert, the service spots and addresses smaller problems before they can develop into major issues that result in costly downtime. Expert Insight is already providing crews on over 100 vessels with
invaluable support and peace of mind.
Available as part of a Wärtsilä Lifecycle Agreement, Wärtsilä Expert Insight is a predictive maintenance service that combines artificial intelligence (AI) techniques and advanced diagnostics with Wärtsilä’s OEM expertise to identify anomalous behaviour. It proactively identifies and highlights potential failures and early indications of equipment deterioration that cause vessels to run less efficiently.
With Expert Insight Wärtsilä can be far more proactive, providing advice and recommendations to customers based on forward-looking predictions rather than historical data.
Wärtsilä has now combined the experience and knowhow gained from the original rule-based detection system, which was first launched over two decades ago, with a new AI-based detection element that learns and develops over time. The more data that is fed to the system, the better it gets at spotting potential problems, resulting in a continuously improving service.
Expert Insight – which is now available for 4-stroke and 2-stroke engines and as a pilot product for scrubbers – is an evolution of Wärtsilä’s established condition-based maintenance offering. Today there are more than 100 vessels benefiting from the added layer of protection that the service offers.
Expert Insight analyses engine data according to a set of rules based on Wärtsilä’s vast installed base and uses AI techniques to model engine behaviour. Anomalies are identified by comparing the predicted value provided by the model to the actual value measured by sensors installed in the engine. Rather than measuring a selected parameter and generating an alarm when that parameter exceeds set values, Expert Insight predicts the value itself, meaning it can recognise a deviation long before an alarm would be triggered and flag the issue for attention.
Expert Insight incorporates a set of algorithms that is designed to recognise patterns in numerical data. Built by Wärtsilä data scientists using Google TensorFlow, this neural network powers the calculations behind the programme and enables Expert Insight to crunch a year’s worth of engine operational data in a matter of minutes.
AI alone is not enough, however, because while AI might be fast at processing data, people are still a vital part of the equation as they are best placed to interpret the findings.
So when Expert Insight identifies a potential problem, an expert at one of Wärtsilä’s ten Expertise Centres will employ their OEM engineering knowledge to confirm what is going on and provide clear recommendations to the vessel’s crew on how to deal with the issue.
Our experts liaise with crew members primarily via the Expert Insight Collaboration app, which is accessible online through an internet-connected browser. As well as providing a platform for communication and reviewing historical data, the app allows customers to remotely view operating data such as trends and alarms from their assets almost in real time – on any device with an internet connection. In the same way, crews onboard vessels can also use the app to monitor asset operation without the need to visit the engine room. Wärtsilä is planning to add process views for different engine systems in the near future.
Expert Insight has proven itself more than capable in the field, detecting everything from seemingly minor deviations that indicate a bearing failure in the turbocharger to alarming trends in the cooling water system – proactively identifying and solving issues that, if not detected in time, can potentially result in significant engine downtime and financial losses.
In December 2020 Expert Insight more than proved its worth for Altera Infrastructure when it identified a problem with the turbocharger onboard its dual-fuel e-shuttle tanker the Aurora Spirit. The AI-based anomaly detection system had alerted a Wärtsilä diagnostics expert to an abnormality in the lube oil pressure at the turbocharger inlet. The expert first liaised with the vessel’s chief engineer via the collaboration app to verify the problem, and after consulting with colleagues from Wärtsilä’s Technical Services and Operational Support teams recommended that the engine be taken offline as a precaution.
A subsequent inspection confirmed that a single failed compressor bearing was the source of the problem. Had this problem not been identified early using Expert Insight and the Wärtsilä team’s intimate knowledge of the vessel’s engines, it could have resulted in a major engine failure – and of course significant financial and operational consequences for Altera.
In a nutshell, for vessel operators Expert Insight means less downtime and fewer days in port. With the ability to continuously monitor shipboard performance, Wärtsilä experts can step in and determine if and when a port call is needed, ensure that vessels can make the most of scheduled port calls, and provide improved predictability in terms of maintenance costs. Expert Insight complements Wärtsilä’s other digital services such as Operational Support and Data-driven Maintenance Planning.
Building on Wärtsilä’s more than two decades of experience in delivering condition-based maintenance, Expert Insight brings together the latest in advanced analytics and OEM expertise to provide maritime customers with an unrivalled level of protection for their valuable assets. From the perspective of decarbonisation, it also enables small changes that can have a big impact in terms of optimising engine performance to reduce fuel consumption and therefore emissions.