Increased and strategic automation is one of the promising outcomes of Industry 4.0. As a result, human operators will shed repetitive manual tasks and focus on maintenance and up-skilling. To realize this goal, manufacturers will continue to collect the troves of machine data that the Industrial Internet of Things (IIoT) delivers. They will use artificial intelligence (AI) to analyze these data sets and act on the recommendations that AI delivers.
To illustrate, manufacturers might already be getting data from a variable frequency drive (VFD) that controls an AC motor in a conveyor system. If that system doesn't have a native sensor that indicates when and if excessive heat and moisture are present, the operator might not be able to avoid a VFD failure which could lead to other issues. The IIoT in manufacturing effectively enables you to capture environmental, external, metadata, etc. to provide you real-time information upon which you could take preemptive actions to avoid mechanical downtime.
AI and mobile edge computing help unlock the full potential of the Industrial Internet of Things
While data-gathering is the first step, the sheer amount of data generated by machines can quickly become overwhelming. Given that the Internet of Things (IoT) will be generating a whopping 80 zettabytes of data by 2025, you need a way of sifting through all of it and separating the signal from the noise. This is an example of what artificial intelligence (AI) can do. Machine learning (ML) algorithms, a subset of AI, can be developed to look for problems based on past history and flag them. AI and ML help make sense of the data that IIoT delivers in an automated way to help organizations make better business decisions.
Traditionally, data garnered from machines had to be analyzed somewhere. AI and ML applications, depending on the use case and desired outcome, could be deployed on-site, on an edge or in the cloud. Manufacturers that leverage AI and ML as part of their digitization and automation strategy have to take in consideration the requirements (like application performance and latency) that need to be in place for data-driven automation to be successful. After all, what's the point of gathering information about machines in real-time if you can't act on it quickly?
Edge computing processes the data close to the source, helping to decrease response times for applications. Mobile edge computing (MEC) drives efficiencies even further by using 5G. 5G can enable massive sensorization, an order of magnitude higher than what's possible with today's cellular technologies, paving the way for large-scale IIoT deployments. The 5G architecture inherently supports lower latencies and faster response times, natively amplifying the benefits of an edge computing solution.
MEC is not meant to replace cloud computing but introduce a hybrid option where the cloud works on heavy-duty computation for long-term insights, leaving on-premises edge computing for real-time analysis. Mobile edge computing also enables devices that don't just harvest data but also act on it at the edge.
How technology enables smart manufacturing
IoT, IIoT, AI, ML, 5G and MEC can be foundational technologies for creating a solution that supports Industry 4.0 use cases. The promise of Industry 4.0 lies not just in automation but real-time automation at scale. Below are a few ways in which these technologies can help on the plant floor:
Enable predictive maintenance
Product data management (PdM) software uses equipment data to evaluate the performance of assets in real-time and minimizes costly downtime. When that evaluation happens at the edge (MEC for example), not only do manufacturers realize the benefits of real-time analytics, but costs are cut by eliminating round trips to the cloud. PdM helps manufacturers that want to be more efficient with their maintenance program and want to avoid downtime surprises.
Bring in remote expertise
MEC and 5G networks can enable augmented reality (AR) such that a worker on the plant floor can use a smart mobile device to call for (and receive) help from a more experienced technician. AR enables both the remote technician and the on-site worker to "see" the situation on the ground so they can troubleshoot efficiently, which decreases the need for dispatching expensive help to the site. AR can also layer diagrams of what machine parts should look like, and display guided tutorials that workers can use to diagnose problems on their own.
Better inventory management
Another promising Industry 4.0 use case is real-time inventory management. By tagging raw materials, unfinished parts or finished products with purpose-built sensors, manufacturers gain real-time location visibility of their inventory. This location data can be integrated into warehouse management and supply chain management systems to enable visibility of selected inventory to upstream or downstream suppliers, which in turn helps strengthen supply chains.
The future of smart manufacturing
Advanced networking will underpin Industry 4.0 and IIoT and MEC will play a huge role too. While many large enterprises have embraced the promise of Industry 4.0, the democratization of these technologies is making room for smaller original equipment manufacturers to dive in and reap promising results as well.
Learn more about edge computing and IIoT in manufacturing with Verizon.
The author of this content is a paid contributor for Verizon.