Pro-active service with Artificial Intelligence

Artificial Intelligence, or AI, can detect deviations much more quickly than a human. But in order to offer value to our customers, deviations must be clarified and solutions must be provided. AI makes it possible for our experts to spend more time interpreting deviations and formulating solutions. This is both the focus and the added value of AI. AI learns from our experts' knowledge so that its sensitivity also improves while the number of false alarms decreases. The technology used for this is called machine learning.


The term ‘artificial intelligence’, often referred to as AI, was first used in 1956 by John McCarthy, an information technologist at Stanford University. AI stands for a computer program's capacity to learn. The idea is to build machines that can learn just as people do and which can then ‘think’ and act on the basis of what they have learned.


Machine learning (ML) is an application of AI in today’s industry, in which algorithms are written. They help the computer collect information. That way, they learn directly from the data without using logic that has been formulated beforehand. The performance of ML algorithms improves with the frequency of course correction.


The AI in this article consists of various ML algorithms. Generally, an ML system consists of three main components:

  1. the model that makes the predictions
  2. the parameters that are used throughout the model to weigh the input data and
  3. the ‘expert’ who adapts the parameters based on the model's capacity to predict the proper result.

Analysing equipment on board a ship is generally done on the basis of the rule-based method. An example: the temperature of exhaust gas may not exceed X degrees under a given engine load. The Wärtsilä engineer gets a notification if that limit is exceeded and then discusses the solution with the customer. Configuring limits is generally manual work. The limits depend on environmental factors around the engine such as fuel, humidity in the engine room etc.

AI learns the behaviour of the environmental factors for each engine specifically in each unique setting. This intelligent technology consequently reduces manual work. This doesn't mean that fewer Wärtsilä engineers are needed, but rather provides support in order to get more from the data and serve the customer better. It provides new insights with which Wärtsilä experts can get to work.

“The new possibilities and the even higher level of service open the door to new business models.”

Faster and broader

One of AI's big advantages is that it can be fed with data collected by sensors. A deviating measurement can quickly be marked as a precursor to a problem. Collaboration apps allow us to work together with the customer to get at the core of the problem. We do this by allowing Wärtsilä experts to communicate indirectly with the ship's crew in the context of the irregularity. Such a data-driven approach can be expanded to various types of equipment.

Pilot at Royal Caribbean Cruise Lines

AI is relatively new to the industry. Still, Wärtsilä has already performed a small-scale pilot with Royal Caribbean Cruise Lines (RCCL). RCCL is known as a leader in the area of digitalisation. To date, the algorithms, together with the engine experts, have detected many potential problems that would otherwise have remained unnoticed, allowing the crew to take timely action to prevent problems occurring nonetheless.

Other customers will also soon experience the fact that the new technologies improve the performance of our equipment in the field considerably. The new possibilities and the even higher level of service open the door to new business models.

Want to know more about AI?

John Kop

Digital Product Owner in Digital Development

Asset Management

Frank Velthuis

Director of Digital Development

Asset Management

“If we combine the power of AI intelligently with the knowledge our experts have, we will build the premier service system of the future.”