DiagFit, the failure prediction software in blind mode

Specialized in the creation of predictive models, DiagFit is the software designed for industry experts to monitor equipment health throughout their lifecycle.

DiagFit, the failure prediction software in blind mode

Specialized in the creation of predictive models, DiagFit is the software designed for industry experts to monitor equipment health throughout their lifecycle.

Failure prediction in blind mode

Serving the most complex equipment

Blind mode fault prediction is a machine learning method that allows for the prediction of equipment failures without relying on historical fault data or knowledge of the equipment or sensors being monitored.

Failure prediction in blind mode

Serving the most complex equipment

Blind mode fault prediction is a machine learning method that allows for the prediction of equipment failures without relying on historical fault data or knowledge of the equipment or sensors being monitored.

Usability

Interpretable detections
By displaying health indicators for each sensor, in addition to the overall health indicator, decision making becomes more rational.
Designed for domain experts
An iterative approach to managing decisions in the field.
Incremental learning
Models can adapt to the evolving contexts encountered by the equipment, making them robust to operational changes.
No-Code
An interface designed entirely in no code to be used without development or data science skills.

Performance

High accuracy
DiagFit has a unique technological core, developed in conjunction with academic research, and produces accurate models (low FP and FN).
Frugal AI
DiagFit is very frugal in terms of processor power and data volume, which allows it to operate in embedded mode in particular.
Agnostic
DiagFit works with all types of equipment as long as they are equipped with sensors producing data in the form of industrial time series.
Fast model production
DiagFit is fast, thanks to the small amount of historical data required, the automated approach and the accuracy of the algorithmic core.

Clients

Use cases

Failure prediction on an anonymized ship's system

Using anonymized healthy data acquired over a year on the ship, DiagFit was used to create a unique model in hours that captures correlations between sensor data and produces a health status per sensor.

Partners

News & Events

Request a demo