Use of artificial intelligence (AI) in digital twins can help to analyze large amounts of data. These activities can be used to model system performance, asset interoperability, equipment wear and maintenance and visualization. Even energy consumption can be monitored and predicted with AI's algorithms.
The list goes on and on.
Long short-term memory (LSTM), for example, can be used to forecast asset behavior, such as a motor's overheating, without requiring a thermal model.
In this article from partner publication Automation World, DataArt's Max Ivannikov describes the new frontier of AI applications in industrial settings.