FEATURE ARTIFICIAL INTELLIGENCE Not a
lost treasure The four pillars of the Industrial IoT
Words by Jennifer McClure, digital industries software at Siemens P
lugging into the Industrial IoT (IIoT) brings manufacturers a network of new ways to cut costs, improve performance and increase productivity. Although a digital transformation may seem overwhelming, the transition can be achieved in just four phases, known as pillars: connectivity, control, digitalisation and augmentation. Connectivity involves connecting physical devices and enterprise systems to the IoT, to foster system integration, increase transparency, and improve processes remotely across plants. Control allows a company to use data from connected devices to gain complete transparency and control the performance of assets.
Digitalisation takes control a step further by using data to create a digital copy of a product or system to find efficiencies, troubleshoot problems, test solutions, and improve product development. Real-time data from the field is fed back into the digital twin. In the final phase, augmentation, IoT and artificial intelligence (AI) are combined to create smart machines that can use data to operate independently of human influence. Now, to elaborate on what these pillars mean for the IIoT.
PILLAR 1 – CONNECTIVITY To clarify, the first step in a digital transformation is connecting physical devices and systems to the IoT. Even companies with legacy machines can do this with sensors and hardware investments; plants in multiple locations can be connected and then monitored. Real-time data can then be collected, and alarms can be set to notify the manufacturer when an asset isn’t
12 APRIL 2020 | ELECTRONICS
performing properly, removing the likelihood of an urgent emergency repair.
PILLAR 2 – CONTROL The second step in a digital transformation is using the data being collected to optimise machine maintenance, using predictive and prescriptive maintenance. This involves replacing traditional maintenance methods – reactive and scheduled maintenance – with evolving, data-driven approaches.
Predictive and prescriptive maintenance involves servicing machines at the right time to predict and prevent failure. This eliminates ill-timed maintenance and reduces downtime, allows for the remote monitoring of machines, and enables the manufacturer to identity the root cause of production issues. Predictive maintenance saves 12 per cent in costs compared to scheduled repairs, reduces maintenance costs by 30 per cent, and decreases breakdowns by 70 per cent.
PILLAR 3 – DIGITALISATION The digitalisation process uses data to create a digital copy of a product or system to find efficiencies, troubleshoot problems and test solutions, and improve product development. Real-time data from the field is then relayed back into the digital twin.
There are three types of digital twin: product, production and performance. A digital twin of a product allows a manufacturer to test out variations of a proposed product before creating a physical prototype.
A digital twin of production recreates the entire production process; the manufacturer can find flaws in the process without affecting plant output. A digital twin of performance gathers real-time data from operational products and the production line, to enable manufacturers to identify ways to improve the product or process. That data can also be fed back into the digital twins of product and production for continuous improvement.
“Although a digital transformation may seem overwhelming, the transition can be achieved in just four phases: connectivity, control, digitalisation and augmentation...”
PILLAR 4 – AUGMENTATION The final step in the digital transformation is using data gathered from the IoT to inform machine operation, without human interference - AI makes sense of the data collected from the IoT by using machine learning to predict outcomes and react accordingly. By automating the operation of machines, manufacturers can reduce errors and increase productivity, as AI gives companies a way to disrupt existing business models and unlock new and exciting opportunities. Siemens and an analytics company are working together to embed enhanced analytics in MindSphere, to further enable machine learning and AI capabilities in IoT environments. By streaming these enhanced analytics, it provides customers with real-time AI for connected machines. Using predictive and prescriptive maintenance, companies can increase productivity and reduce operational risk.
Siemens
new.siemens.com / ELECTRONICS
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
