Cost reduction through predictive maintenance

Interview Maarten ten Wolde, Technical manager at Wagenborg Offshore

Within the ‘SMArt Maintenance of Ships (SMASH) project’ ship-owners, suppliers, data- and IT-specialists team up to make vessel maintenance condition based. Maritime maintenance is not only important, but also a huge expense. Unplanned accidents come with a hefty repair cost, due to the inability to deploy a vessel and thus causing loss of revenue. Currently, maintenance is done on a preventive or corrective level.  It is the trick to have it take place before a failure happens; this not only increases the deployability of the vessel but also reduces the expenses. 

Wagenborg’s walk-to-work vessel ​Kroonborg is the SMASH-project’s test site for predictive maintenance on the propulsion machinery. Together with its partners, Wagenborg collects data from the propulsion machinery. Analysis should provide insight in the failure itself and its cause: valuable information that could possibly save many tens of thousands of euro’s. 

“The potential cost reduction is indeed significant”

Predictive maintenance

Technical manager Maarten ten Wolde explains what the SMASH-project on the Kroonborg entails. “The Kroonborg supplies offshore platforms in the North Sea with manpower and materials each day.  It is a diesel-electrically driven vessel without a propeller shaft. Its vertical Voith Schneider propellers are driven by an electric motor. The complete propulsion machinery consists of a diesel engine/generator, a switchboard, inverters, an electric motor and the propellers. Maintenance is currently conducted based on the number of operating hours and thus it is mainly preventive. Predictive maintenance could really make a difference, by eliminating unnecessary procedures. A maintenance cycle can be delayed or put forwards, thus reducing your expenses. Furthermore, it increases your reliability.”, according to Maarten. 

Although there are some suppliers who facilitate predictive maintenance, this is mainly suitable for their own equipment. Maarten: “But what we are looking for is a complete picture of the whole propulsion train. This is impossible with individual suppliers, they have too little focus on what we want. But it did trigger us to get started on this subject. It’s currently a trend, also from the supplier’s point of view, to collect data and to do so for different machinery. However, a good analysis requires formulating uniform rules and they can’t deliver. That’s what we are doing with SMASH on the Kroonborg. Eventually, ​this will give us a head start and ultimately a better position.”


The Kroonborg underwent a good deal of preparation before the first results can be expected. “At the end of June, Semiotic Labs and Bachmann have installed a series of sensors.”, Maarten explains. “Semiotic monitors the entire system by measuring high frequency currents in the electric motor. In addition, Bachmann measures temperature, pressure, vibration, humidity, driving current and oil. A very diverse spectrum. Subsequently the collecting of data can begin, from which the first algorithm will take shape. I am very curious about the first results: how are the systems related to one another, how do they influence each other and can we make a reasonable prediction based on the data collected? I have high expectations. Collecting data is one thing: afterwards you will need to analyse it and the best way for this is for a failure to occur, which of course is something you’d like to avoid. We do expect to find an algorithm that allows us to shed some light on failures and their cause. We intend to proceed towards a system that will run on all motors and predicts failure. I expect it to succeed, I’m at least 95 percent certain.”

Towards the future

Should the Kroonborg concept prove to be successful, it would mean a significant potential cost reduction for the entire Wagenborg fleet. If predicting failures and their cause can prevent several days of downtime, the cost recovery will be swift. “The potential cost reduction is indeed significant. Our next step is to equip the rest of the offshore fleet with the system: that would mean 24 vessels. After that, the vessels of the shipping fleet can follow, coming to a total of nearly two hundred vessels”, Maarten concludes.