Machine learning algorithm for predictive maintenance
Budget: €250 – €750 EUR
Im working on a predictive maintenance project. I have sensordata (about 0,6 gb) from liquid gas pumps.
I need to extract knoweddge about the state of the pump. i dont know whats possible based on the data i have. (Maybe the remaining useful lifetime of the pump, or a chance that the pump will fail in a certain week would be helpful for maintenance).
In detail:
Visualize the pressure/temperature/electricity of each pump over the time entire period in fancy graphs. Also aggregation of maybe 1d/1week/1month/1year.
Figure out which model works best for my case, then Build ML-model split in train/test. Then visualize comparison between actual data and forecast/prediction.
Then plot the feature importance. Maybe shapbeeplot?
The data will be provided when the deal is settled.
I need to extract knoweddge about the state of the pump. i dont know whats possible based on the data i have. (Maybe the remaining useful lifetime of the pump, or a chance that the pump will fail in a certain week would be helpful for maintenance).
In detail:
Visualize the pressure/temperature/electricity of each pump over the time entire period in fancy graphs. Also aggregation of maybe 1d/1week/1month/1year.
Figure out which model works best for my case, then Build ML-model split in train/test. Then visualize comparison between actual data and forecast/prediction.
Then plot the feature importance. Maybe shapbeeplot?
The data will be provided when the deal is settled.