From real-time processing and analytics driving greater efficiency and learning, to predictive maintenance, Edge AI is powering a revolution in manufacturing…
Manufacturers around the globe are under huge pressure to maintain a competitive advantage and early interest in Internet of Things (IoT) and AI technology is maturing into ongoing digital transformation programmes. In fact, leveraging industrial IoT to optimise operations has become so established, it has been dubbed ‘Industry 4.0’.
The Industry 4.0 elevator pitch is that by connecting the ‘top floor’ to the ‘shop floor’, vast efficiencies can be made in almost every area of the manufacturing process. From supply chain management, automated assembly, predictive maintenance through to automated fulfillment, the benefits of a fully-integrated, Edge AI-powered manufacturing process are considerable.
In fact, the McKinsey Global Institute has estimated that AI techniques have the potential to create between $3.5 trillion and $5.8 trillion in value per year, across nine business functions in 19 industries. Edge AI’s power to merge industrial data and transform it into usable products and innovative services is the central plank of Industry 4.0.
Dynamic Predictive Maintenance
Using data analytics, IoT sensors and Edge AI, systems can learn baseline operating behaviours for factory equipment to define a model for ‘normal’ activity. When differences between real-time behaviour and this normal model are detected, alerts can be sent, or equipment can be automatically shut down as a safety precaution. For example, ultrasonic sensors might monitor conveyor belts or other mechanical processes, and when a change in pitch is observed, an alert is sent. This allows plant managers to switch from inflexible and often costly scheduled servicing to more dynamic predictive maintenance that focuses on preventing costly failures, and avoids unnecessary servicing.
This data-focused approach does raise processing challenges, as device intelligence needs to be gathered, analysed and actioned in real-time. This is where Edge AI becomes particularly important to Industry 4.0, as it removes much of the latency from a more traditional cloud system – that latency can make the difference between success and failure where industrial machinery is concerned.
In just one example, a chain of garment factories in China began a pilot project, partnering with Alibaba Cloud’s Internet of Things team and immediately saw their scheduling performance increase 6 per cent and delivery cycles became 10 per cent shorter. One of the pilot factories, Hangzhou Dianshi Garment, had data collection camera arrays installed, debugged and production process synchronised via computer vision algorithms in under two days, for a total cost of less than CNY50,000 (USD7,400), according to the companies.
Fast and Flexible VIA Edge AI Solutions
To meet the diverse manufacturing process automation needs of industrial customers, VIA has developed a growing family of Edge AI systems and platforms that can be rapidly customized to meet specific deployment requirements.
One key element of our strategy is to to ensure full compatibility of our systems with leading cloud and edge intelligence platforms from companies such Alibaba, Microsoft, and Foghorn.
Last year, VIA was recognized as a Best Alibaba Cloud IoT Intelligent Manufacturing Partner at the ICA (IoT Connectivity Alliance) Ecosystem Partner Summit 2018 and our VIA ARTiGO A820 enterprise IoT gateway was officially certified as compatible with the Alibaba Cloud IoT open edge computing platform Link Edge.
In addition, VIA also participates in the Microsoft Azure Certified for IoT program, which helps businesses get started on their IoT projects quickly by connecting them with an ecosystem of partners that have offerings that can easily connect to the Azure IoT Suite. We offer a growing number of boards and systems that are Microsoft Azure Certified for IoT, including the VIA VAB-820, VIA ARTiGO A630, VIA ARTiGO A820, VIA ARTiGO A1250, VIA AMOS-820, VIA AMOS-825, VIA AMOS-3005, and VIA Alegro 100 systems as well as the VIA SOM-6X50 and VIA SOM-9X20 platforms.
For industrial customers looking to embed edge intelligence directly into small-footprint IIoT devices as close to the source of streaming data as possible, VIA has also partnered with Foghorn to integrate support for Foghorn’s Lightning software platform into VIA Edge AI systems. By accelerating decision-making using real-time data, the combined solution will open up a plethora of opportunities for industrial customers to increase the efficiency of their manufacturing processes, reduce costs, and optimize business outcomes.
VIA Edge AI Industrial Solutions at Embedded World
VIA will be showing our latest Edge AI Industrial Solutions at Embedded World 2019 from February 26 – February 28 at the Nuremberg Exhibition Center in Germany. The VIA booth is located at #2-551 in Hall 2. To find out more, please click here.