AI forLife Cycle Management

Important results from the project

The aim of the project was to evaluate how AI can be used together with Systecon´s existing LCM software Opus Suite. The goal was to understand what an AI component that complements Opus Suite could look like, e.g. by integrating sensor data with the system analyses that Systecon is currently a leader in. The project has fulfilled its aim by developing a prototype for data-driven prediction of failure frequencies, an existing use case developed together with BAE Hägglunds.

Expected long term effects

The project has developed a framework where the data input to Systecon’s LCM-models is partly automated through a prototype where a neural network, which is used for parameter estimation, was integrated with Opus Suite. The project identified two areas in the interface between AI and Systecon´s analytical framework for further investigation: semi-monitored labeling of data as support in the manual part of data-driven estimation of parameters; learned distributions as a replacement for traditional, parametric distributions.

Approach and implementation

The project has form most parts worked well. One challenge that appeared during the project was the lack of a sufficiently comprehensive and detailed amount of real world data, which affected Ap2. This was partly solved by developing a data generator that enabled us to use simulated data instead.