Our self-learning algorithms – developed for more than 20 years – find correlations between large numbers of system-level signals that are unobservable to human operators.
After an initial learning cycle, MP INtelligence can quickly identify abnormal patterns in real-time signal data and generate the appropriate warnings for O&M specialists.
MP INtelligence can recognize patterns that are early indications of an upcoming failure – even when the incident happens for the first time. This is thanks to our unique approach of combining large-scale unsupervised learning with continuous user feedback, as well as re-training the machine-learning models whenever needed.
The ability to handle thousands of signals simultaneously, and quickly identify any process anomalies, is what sets MP INtelligence apart from monitoring systems that focus on individual parameters only.