Prediction or Condition-based Maintenance
Oil and Gas
Power and Water
Anticipating machines failures before they actually happen through predictive analytics
Predictive maintenance prevents unnecessary repairs, maximizes effective asset lifetime, and significantly reduces major failures and downtime. This results in cost savings with an increased return on investment and customer satisfaction. Providing predictive maintenance at the edge and sending only relevant data to the cloud is a lower cost alternative to cloud-only based solutions. It provides real-time actionable insight with extremely low latency and substantially lowers data transfer and storage costs.
The Electric Submersible Pump (ESP) is a long in-ground piece of equipment at the center of an oil well extracting oil from the bottom of the well and pumping it to the surface. Failure of an ESP will stop the entire operation which can take time and result in costly repairs and loss of revenue. An edge predictive maintenance solution can monitor the operational data gathered from the ESP and apply advanced analytics in real-time to predict failures. If a potential failure is detected, the system can automatically stop the pump to prevent damage as well as alert operations to repair or replace the ESP based on current machine health and maintenance models developed by the operators of the ESP. An edge solution can reduce the cost of data transfer to the cloud by preprocessing real-time data at the edge and sending only relevant data to the cloud. It also provides real-time availability without requiring an uninterrupted network connection that cloud-based solutions depend on.