[Answer] How Cognitive Analytics & AI Helps in Ship Maintenance?

The webinar saw Steel Ships chief executive Dr Ranjan Varghese, Columbia Shipmanagement POCR manager Pankaj Sharma, and NYK Bulkship (Asia) operations director Capt K K Mukherjee discuss how to extend intelligent monitoring of onboard machinery.

“Cognitive maintenance is the only way forward. We are having serious problem with getting a complete crew on board,” Dr Varghese said, explaining that deploying a predictive cognitive maintenance decision support system has become even more relevant in light of crew change problems during the coronavirus pandemic.

Dr Varghese said implementing the cognitive maintenance system is already producing results.

“We have the availability and maintainability increased by 15%,” he said. “Failures have reduced by 30% and energy consumption by 6% to 10%. The reduction of spare parts consumption is between 7% to 15%.”

Integrating Shore & Ship Activities

One key step in the process is integrating shore and the shipboard activities into one data centre. Columbia Shipmanagement set up its control room in December 2018, and Mr Sharma described how one of the first projects was modernising a superyacht.

A high end yacht, but more than 30 years old with limited sensorisation, Columbia Shipmanagement’s task was to limit downtime and improve predictive maintenance. Wifi was chosen as the onboard communication tool for the new sensors to provide real-time data, and Mr Sharma described some of the project’s outcomes.

“So how did that go? In less than six months, the machine learning AI application was able to create thresholds and benchmarks for almost all the machinery. And it started identifying trends. And it started predicting trends when something was likely to go outside of normal vibration parameters,” said Mr Sharma.

In one case, the predictive system recognised, through vibration analysis, a misalignment in the air conditioning system, one of the crucial systems in a luxury superyacht.

“When we did a cost benefit analysis after this case, we realised the system had already paid for itself,” said Mr Sharma.

Benefits of Deep Sensorisation & Machine Learning

Capt Mukherjee, speaking from Singapore, noted the hidden benefits of deep level sensorisation and machine learning. “This can lead to data mining during inventory control and various search optimization of the operations in relation to navigation or running machinery,” he said.

In summary, Dr Varghese noted that with the drive to extract greater intelligence from technology, safety remained the most important element of shipping operations.

“Whatever we do, the ultimate objective and the paramount goal has to be operating a safe ship.”

Mr Sharma’s final thoughts were that companies and fleets are sitting on a lot of data.

“We were able to demonstrate that the AI applications and predictive maintenance pays for itself and the learning curve will be exponential,” he said.

Capt Mukherjee noted “You have to have an intention or mission to transform your digital resources and not to waste them”.

Some 60% of webinar attendees felt that engineroom ship operations management would be the fastest and most realistic area for intelligent monitoring and 69% agreed that 5G technology will accelerate the implementation of cognitive analytics on board.

The webinar attendees were fairly evenly split as to when intelligent monitoring of machinery on board would become a reality. Just under half thought it would be by 2025 and a similar proportion felt that 2030 was a realistic timeline.

You can view this webinar and all of the webinars from Riviera’s Vessel Optimisation Webinar Week, in full, in our webinar library.