If you’re a developer in manufacturing, agriculture, oil and gas or any other industry using predictive maintenance, you know that some of the biggest challenges include getting timely and consistent — and the right — data from an amalgam of sensors and connectivity solutions spread across your facilities. When uptime is critical and asset availability is crucial, predictive maintenance can make all the difference.
Consider this scenario: Thousands of sensors in your facilities create a flood of vibration and acoustic data from your key systems 24/7. Your job is to make sure your assets and infrastructure integrate so that you have one “source of truth” for watching operational insights and alerts. As the business grows, it’s time to scale, and you’re tasked with managing increasing complexities around frequencies, certifications and power requirements across geographies. Now it’s time to think about mesh networking.
If you’re not taking advantage of mesh networking technologies for predictive maintenance applications, you’re missing one of the most scalable, robust and efficient architectures available today.
Why Mesh Networking?
Mesh networks offer enormous benefits that make setting up and scaling device networks seamless and simple. Mesh networking optimizes bandwidth allocation, extending range capabilities as data hops from node to node. Mesh networks such as Zigbee and DigiMesh® also “self-heal,” meaning they can create alternate paths when a node fails or connectivity drops. Finally, mesh networks are scalable, power efficient, secure and simple to manage.
Why Is Predictive Maintenance Important?
According to IoT Analytics, 95% of companies using predictive maintenance reported a positive return on investment. That’s because unplanned downtime can cost some $125,000 per hour for a midsized business. Just one correct prediction can pay for the entire investment.
To drive the actionable insights and automation that support predictive maintenance requires intelligence at the edge, and mesh networking is an excellent methodology for OEMs building applications that require accurate data from across the network.
Considerations When Selecting a Mesh Network Technology
Going from raw data to predictive maintenance sounds simple, and mesh technologies such as Zigbee and DigiMesh work well for capturing this data. And extremely low-power devices incorporating these technologies enable OEMs to build scalable networks for a range of industries. However, keep in mind that developers still need to create software and analytics dashboards to decide when vibration data might show an emerging issue.
ISO standards can help you differentiate between a bearing failure and a loose mounting. For example, ISO 13373 (condition monitoring and diagnostics of machines – vibration condition monitoring) is one of several standards relevant to industrial applications. ISO 13373 outlines procedures for setting alarm limits, trending vibration data, and performing diagnostics to identify the root cause of machine issues, such as bearing defects or imbalance. By using these standards, you can implement an effective condition monitoring program to predict failures in equipment such as bearings and mounts before they cause downtime.
When it comes to building a mesh network, the size and scale will depend on the amount of data you send and how often you send it. However, a better way to decide on the size of your network is to understand the bandwidth necessary. Mesh networks can easily scale from a handful to hundreds of nodes, but if you’re transmitting lots of raw data over lots of devices, you can reduce the number of nodes and then separate them with network IDs and a single gateway. But if your sensors don’t send much data, you could scale to hundreds of nodes on a single gateway. In other words, mesh networks provide flexibility to design networks with different gateway and node configurations so you don’t have too many bandwidth-hogging devices clustered together.
Prototyping and Testing
Developers working with mesh networks for predictive maintenance can really benefit from IoT development kits. Because they’re relatively inexpensive and come with all the hardware, they can simplify the design and provisioning processes. In fact, running tests through different protocols available with a development kit can even help you quickly decide which setup works best for your application before making a big investment.
Protocols
Zigbee is a popular open standard mesh networking protocol specifically designed for low-data rate, low-power applications. It offers interoperability between vendors and supplies robust security features. Proprietary third-party solutions built on top of the Zigbee protocol provide embedded solutions to make implementation easier for developers and integrators. DigiMesh offers a similar protocol, but is designed for fully battery-powered operation in the field, with simpler design and deployment. There are 900 MHz, 868 MHz, and 2.4 GHz frequency options that allow DigiMesh to run worldwide.
These products often simplify network set-up, enhanced security features and longer ranges. Best of all, some of these third-party devices integrate with remote management software that makes deploying and troubleshooting seamless.
Conclusion
Finally, developers responsible for keeping these systems operational can tap into mesh networking to enable intelligent, real-time predictive maintenance monitoring through intuitive dashboard interfaces. When you’re ready to set up your mesh network for predictive maintenance, select a wireless partner that is technology-agnostic, so you get an objective perspective on which technology best fits your needs for today and tomorrow.
Quinn Jones is a product manager at Digi International who has helped pioneer RF module adoption for IoT since 2011. During his more than 20 years at Digi, Jones has covered a broad portfolio, including the Digi XBee Ecosystem, system-on-modules and cellular modems. His expertise lies in helping customers navigate through different wireless topologies ranging from simple point-to-point networks to complex mesh networks using a variety of technologies including Zigbee, Wi-Fi, 900 MHz, and cellular.