The Mtell Previse condition monitoring solution uses machine learning to prevent breakdowns, increase asset lifecycle, reduce maintenance costs, and increase production output for any industrial process.

Previse Data Sheet

Contemporary condition monitoring applications use techniques to “trap” anomalies or changes in operational behavior of a machine that might indicate a problem. Such methods are complex, limited to certain equipment, prone to error, and ALWAYS require further expert investigation and validation; producing high levels of false positive

Mtell Previse uses Autonomous Agents  to learn operational behavioral patterns using actual data from sensors on and around a machine or manufacturing process. Mtell Previse recognizes diverse patterns in the sensor signals that indicate degradation, failure, and root cause.

Architecture Integration

Mtell Previse extracts sensor data from the installed plant historian, and the failure work order history from the existing enterprise asset management (EAM) system. Previse works with any equipment- connected sensors and requires only a few to activate machine learning.

Predictive Scheduling

Mtell Previse immediately sends work orders to the EAM system when Agents detect failure signatures. This provides time to plan and organize maintenance. Messages also contain the full scope of work including: cost, tools, labor, time, and safety concerns.

Self-Learning and Training

When an anomaly is detected, Mtell Previse dispatches alert notifications and requests for inspection. The results of inspection determine if the behavior is a failure signature or a new, normal operating state. The system trains itself to become smarter over time; learning and adapting to new asset operating conditions.

Early Warning & Accurate Time-to-Failure

Agents detect issues (the P in the P-F curve) far earlier than prevailing approaches. When a failure anomaly is detected, Mtell Previse searches further back into the sensor history to refine the signature, in order to improve time-to-failure notifications. A 7-day Anomaly alert can easily become a 30-day machine learning alert.

Transfer Learning

Learned behaviors (normal, degradation and failure) captured on one machine are readily transferred to equipment of the same type with the same sensor configuration. After a very short retraining period, every machine shares the same safetyand breakdown protection.

Root Cause

Detecting symptoms is not the same as detecting the cause of degradation. Mtell Previse provides a sensor-ranking chart that shows which sensors contribute most to machine degradation.

Request Live Demo

Behind the Software

  • How Mtell Previse solves machine problems
  • How to set up and configuration overview
  • How to create machine learning agents that detect issues
  • The solution in action; including alerts, work order requests, and asset health monitoring through heat maps and other visuals

I want to know more about

Mtell Reservoir