Know what’s left. Plan what’s next.
Lifespan Predictor estimates each unit’s remaining life, and its chance of breaking down before the next service, from the sensor logs you already have. It automates feature engineering and hyperparameter optimisation for you and it supports multiple types of standard survival models.
It comes with an easy to use SDK and example notebooks which illustrate the product features with a toy dataset (NASA C-MAPSS). SDK supports currently AWS SageMaker based training and inferencing.
Free Tier can be used for up to 100 unit fleets. For larger fleets, contact us.
Resources
- Lifespan Predictor SDK: the Python package your team uses to train and predict, with examples, on GitHub
- SDK API Reference: every class and function in the SDK