AWS Aims 5 AI Tools to Modernize Manufacturing and Industrial Plants
Amazon Web Services is moving to expand industrial AI use with five new industry-focused AI tools that can watch over manufacturing plants 24/7 to detect problems on assembly lines and other systems, while also predicting needed maintenance tasks.
The new AI tools are machine learning services that can help industrial and manufacturing customers bring machine intelligence into their production processes for improvements in operational efficiency, quality control, security, and workplace safety, according to AWS. Using machine learning, sensor analysis, and computer vision capabilities, the tools aim to make it easier for manufacturing and industrial operations to address common technical challenges by using cloud-to-edge industrial machine learning services.
Providing the new AI services are Amazon Monitron, an industrial monitoring system comprised of sensors, a gateway and ML services that can detect abnormal equipment conditions; and Amazon Lookout for Equipment, which allows customers with existing equipment sensors to adapt them to use AWS machine learning models to detect equipment problems and provide predictive maintenance services. In addition, the new AWS Panorama Appliance enables manufacturing customers that already have video camera systems in their facilities to use computer vision to improve quality control and workplace safety, according to AWS.
Also now available are the AWS Panorama Software Development Kit (SDK), which allows industrial video camera makers to embed computer vision capabilities in their latest cameras; and Amazon Lookout for Vision, which uses AWS-trained computer vision models on images and video streams to find anomalies and flaws in products or processes.
AWS recently unveiled the new tools at its AWS re:Invent conference, which is being held virtually in December due to the ongoing COVID-19 pandemic.
James Kobielus, principal analyst with Franconia Research, told EnterpriseAI that the significance of the AWS announcements is that they provide an onramp for small- and midsized manufacturers to adopt these technologies in stages.
“Computer vision is a huge piece of the commercial AI arena, and industrial internet of things is a primary proving ground for this tech,” said Kobielus. “Many manufacturing firms have already deployed smart sensors for digital process monitoring, anomaly detection, and quality control. Computer vision is a huge and growing segment of the IoT smart sensor revolution, and there's still a large potential market for it that is untapped.”
Designed for industrial users who do not have existing sensor networks, Amazon Monitron includes sensors, a gateway, and machine learning services to watch over industrial and manufacturing facilities. Monitron can detect anomalies and predict when industrial equipment will require maintenance so they can focus on other core manufacturing, supply chain, and operations functions. Monitron uses the sensors and other system tools to detects when machines are not operating correctly based on abnormal fluctuations in vibration or temperature, according to AWS. Monitron is now generally available.
The system notifies customers to examine the questionable machinery to determine if preventative maintenance is needed. Using Monitron, maintenance technicians can start tracking machine health in a matter of hours, without any development work or specialized training. It can be used on a wide range of rotating equipment, including bearings, motors, pumps, and conveyer belts in industrial and manufacturing settings. It can be used to monitor critical machines such as cooling fans or water pumps in data centers, or in large-scale installations in manufacturing facilities with production and conveyance systems.
Plant technicians can watch over Monitron in real time using a mobile app and receive alerts about equipment problems, while also having the ability to check up on machine health and schedule needed maintenance. Technicians can enter feedback on the accuracy of the alerts in the mobile app, which then learns from the feedback to improve the app over time.
Amazon Lookout for Equipment
Designed for customers that have existing sensors but don’t want to build machine learning models, Lookout for Equipment allows users to send their sensor data to AWS to build models for them and return predictions to detect abnormal equipment behavior.
Customers upload their sensor data to Amazon Simple Storage Service (S3) and provide the S3 location to Amazon Lookout for Equipment to get started, or data can be pulled from AWS IoT SiteWise, according to AWS. It also works with other popular machine operations systems like OSIsoft. Amazon Lookout for Equipment analyzes the data, assesses normal or healthy patterns, and then uses the learnings from all of the data on which it is trained to build a model that’s customized for the customer’s environment.
It then uses the machine learning model to analyze incoming sensor data and identify early warning signs for machine failure for predictive maintenance evaluations. Amazon Lookout for Equipment gives customers more value from their existing sensors, and helps customers make timely decisions to improve their industrial processes.
Amazon Panorama Appliance
Built to allow organizations to add computer vision to existing on-premises cameras, the new Panorama Appliance connects to their network and automatically identifies camera streams and starts interacting with existing industrial cameras. Integrated with AWS machine learning services and IoT services that can be used to build custom machine learning models or ingest video for more refined analysis, the Panorama Appliance extends AWS machine learning to the edge to help customers make predictions locally in sites without connectivity. Each Panorama Appliance can run computer vision models on multiple camera streams in parallel, making possible use cases like quality control, part identification, and workplace safety. It works with AWS and third-party, pre-trained computer vision models for retail, manufacturing, construction, and other industries. Customer-developed computer vision models developed in Amazon SageMaker can also be deployed on the Panorama Appliance.
Panorama Software Development Kit (SDK)
This SDK enables hardware vendors to build new cameras for uses such as detecting damaged parts on a fast-moving conveyor belt or spotting when machinery is outside of a designated work zone, according to AWS. The cameras can use chips designed for computer vision from Nvidia and Ambarella. Video cameras developed with the Panorama SDK can use computer vision models that can process higher quality video with better resolution for spotting issues, as well as used to build more sophisticated models on low-cost devices that can be powered over Ethernet and placed around a site. Customers can train their own models in Amazon SageMaker and deploy them on cameras built with the AWS Panorama SDK with a single click.
Lookout for Vision
Amazon Lookout for Vision is a high accuracy, low-cost anomaly detection product that uses machine learning to process thousands of images an hour to spot defects and anomalies. Camera images can be sent to Amazon Lookout for Vision in batch or in real-time to identify anomalies, such as a crack in a machine part, a dent in a panel, an irregular shape, or an incorrect color on a product. Lookout for Vision then reports the images that differ from the baseline so that appropriate action can be taken. It is sophisticated enough to handle variances in camera angle, pose, and lighting arising from changes in work environments. Customers can accurately and consistently assess machine parts or manufactured products by providing as few as 30 images of the baseline “good” state, according to AWS. It also runs on Amazon Panorama appliances. Lookout for Vision is available now and in 2021 customers will be able to run Amazon Lookout for Vision on AWS Panorama Appliances and other AWS Panorama devices.
AWS industrial machine learning services are being used by AWS customers and partners including Axis, ADLINK Technology, BP, Fender, GE Healthcare, and Siemens Mobility.
Moving Industrial IoT Forward
“What's particularly intriguing is how extensively AWS has invested in industrial IoT and what a well-thought-out package of offerings they're providing for on-ramping their customers to sophisticated computer vision for industrial automation,” said Kobielus of Franconia Research. “In a year when industrial supply chains (especially Amazon's) have been stressed to the max and people rely more on these levels of industrial automation in order to protect factory workers, this set of announcements should appeal to many manufacturing firms all over the world. This level of automation is necessary to protect the health of industrial workers as we head into the darkest, most deadly months of the pandemic.”
Also interesting is that the latest AWS AI announcements nicely encompass both units in Amazon’s larger business model, from the increasingly AI-automated supply chain behind its retail powerhouse and a cloud-to-edge AI pipeline in which computer vision and other smart sensors are becoming core development platforms, said Kobielus. “These new products are the fruit of the synergies that [CEO Jeff] Bezos and team continue to build between the two hemispheres of Amazon.”