Amazon Web Services (AWS) has announced that Amazon Lookout for Equipment and its new Amazon Lookout for Metrics service
are now available for general AWS users. The service uses machine learning to autonomously monitor and detect unexpected anomalies in business and operations data. This means that it can quickly diagnose potential issues from their root source before they impact workflow performance. In addition, AWS allows users to rank anomalies by severity, making it easier to prioritize which issues should be addressed first. The company first debuted Lookout for Equipment in December 2020 during reinvent 2020.
Amazon Lookout for Equipment helps customers perform predictive maintenance in their facilities over multiple locations by analyzing sensor data such as pressure, flow rate, RPMs, temperature, and power. It then uses machine learning to predict early warning signs of machine failure, suboptimal performance as well as equipment abnormalities with speed and precision.
Companies, which currently include Siemens Energy, Cepsa, Embassy of Things, RoviSys, Seeq, and TensorIoT, are responsible for uploading their sensor data to Amazon Simple Storage Service and providing relevant S3 bucket location information, while Amazon's service automatically completes the rest. While many companies use basic simple rules or modeling approaches to identify issues based on past performance, this new approach is said to help industrial companies improve operational efficiency by avoiding maintenance time due to equipment failure and reducing false alarms based on misdiagnosed issues.
Lookout for Metrics reportedly uses the same kind of machine learning technology employed by Amazon to monitor and diagnose metrics and data. It is capable of detecting unusual trends such as abrupt changes in sales revenue, spikes in payment transaction failures, customer acquisition rates, web page views, number of active users, high rates of abandoned shopping cart transaction volume, mobile app installations and increases in new user sign-ups.
“From marketing and sales to telecom and gaming, customers in all industries have KPIs that they need to be able to monitor for potential spikes, dips, and other anomalies outside of normal bounds across their business functions. But catching and diagnosing anomalies in metrics can be challenging, and by the time a root cause has been determined, much more damage has been done than if it had been identified earlier,” shared Swami Sivasubramanian, vice president of Amazon Machine Learning for AWS.
To configure and use the service, developers need to include detectors that are responsible for observing data such as a location field for monitoring the availability of an application worldwide or in various AWS regions. Developers can also set the interval for how often the detector imports data for monitoring and detection.
Users can then instantly respond and provide feedback based on the summary provided by the system through the Lookout for Metrics console or another preferred API. AWS stated that the service also constantly improves its accuracy based on feedback and performance over time.
Currently, Lookout for Metrics supports the following data stores:
- Amazon S3
- Amazon CloudWatch
- Amazon Relational Database Service (Amazon RDS)
- Amazon Redshift
The service will also be available for integration with widely used software as a service (SaaS) applications such as Salesforce, Marketo, and ServiceNow. Metrics is also connected to notification and event services such as Amazon Simple Notification Service (Amazon SNS), Slack, PagerDuty, and AWS Lambada. Users can customize alerts or set specific actions such as filing support tickets or automatically removing incorrect information from a product page on retail websites.
Amazon Lookout for Equipment is currently only available in U.S. East (N. Virginia), EU (Ireland), and Asia Pacific (Seoul). While, Lookout for Metrics is now available in major AWS regions in the U.S., Europe, and Asia, namely:
- U.S. East (N. Virginia)
- U.S. East (Ohio)
- U.S. West (Oregon)
- EU (Ireland)
- EU (Frankfurt)
- EU (Stockholm)
- Asia Pacific (Singapore)
- Asia Pacific (Sydney)
- Asia Pacific (Tokyo)
Amazon says both services will be accessible to more regions in the coming months.