Hi community, hoping you are well, I am writing to ask you about the capabilities of Thingworx Analytics.
In the field of predictive maintenance, one of the key objectives is the detection of failure conditions before they present themselves. Concerning that, we'd like to know how that can be implemented in Thingworx Analytics.
For example, say you want to predict a pump failure. Failure conditions are known (e.g. zero pressure), and past data is labeled with a boolean "ok/fail" for each timestep. The main objective is not to know the status of the pump given the current conditions, but to predict whether the pump will fail in the short term, given the current trends in the data.
We've previously developed a model which can give an estimate of the current condition of a machine. The model was trained using historical time-series data. However, we're unsure of the way forward for developing a predictive model, which can give an estimate of the probability of a failure condition in the future.
Can you give us a hand with this? We only need some guidance about the next steps. We'll be very grateful if you have some documentation or a step-by-step that you can share with us. We could even schedule a meeting if you have the availability.
Thanks in advance for your help with this matter.
Kind Regards,
