Recommending Maintenance Activities for Unplanned Events of Gas Engines
In many industries, like automotive manufacturing, greenhouse or data center,
gas engines are used for distributed power generation. One well-known manufacturer
of such gas engines is INNIO, which produces the Jenbacher gas engines.
In recent years the Asset Performance Management (APM) solution myPlant
was developed by INNIO. The goal of this cloud-based platform is to increase
reliability and safety of the engines while reducing maintenance costs and engine
downtime. This is achieved by using various predictive analytic algorithms for
predicting lifetime of engine parts, such as oil filter and spark plugs, or notifying
customers when any deviation in the sensor data of the engine is occurred.
The aim of this thesis is to create a data-driven system to establish correlations
between the fault symptoms and causes by using machine learning
methods. The results of this system will then be compared with the Service Expert System.
Based on this system, a prototype of an issue resolution assistance application
will be implemented on myPlant to guide the user from the issue to the