Amazon Web Services (Aws) announced the general availability of Amazon Lookout for Equipment, which allows companies to use machine learning to fully exploit their investments in equipment sensors and perform predictive maintenance on a large scale at all their industrial sites.
The new Amazon Lookout for Equipment service, based on Aws’ machine learning models, works by inserting sensor data from the industrial equipment of a customer company (e.g. pressure, flow rate, RPM, temperature and power) and then training a customer’s own equipment
With Amazon Lookout for Equipment, Aws pointed out, companies can detect equipment anomalies with speed and precision, quickly diagnose problems, reduce false alarms and avoid costly downtime by acting before machine failures occur.
There are no initial commitments or minimum rates with Amazon Lookout for Equipment, AWS explained: customers pay for the amount of data, the calculation hours used to train a custom model and the number of hours of inference.
With Amazon Lookout for Equipment, industrial and manufacturing companies can quickly and easily build a predictive maintenance solution for an entire plant or multiple locations.
To get started, customers upload sensor data to Amazon Simple Storage Service (S3) and provide the location of the respective S3 bucket to Amazon Lookout for Equipment. The service will automatically analyse data, evaluate normal or healthy models and build a machine learning model tailored to the customer’s environment.
Amazon Lookout for Equipment will then use this model to analyze incoming sensor data and identify the first warning signals of faults or malfunctions of the machines. For each alarm, the service will specify which sensors are indicating a problem and will measure the extent of its impact on the detected event.
This allows customers to identify and diagnose l’alert, prioritize the necessary actions and perform precision maintenance before problems occur, saving money and improving productivity by preventing dead times, making timely decisions.