Twitter icon
Facebook icon
LinkedIn icon
Google icon
Reddit icon
StumbleUpon icon
Del.icio.us icon

Innovative IoT Solution Helps Manufacturers Predict and Fix Problems Before they Occur

Added to IoTplaybook or last updated on: 04/05/2019

Many leading manufacturers still use paper checklists, manual processes, and legacy closed-loop technologies to monitor and maintain their equipment, in part because the challenge of launching technology that could help them automate their operations and take them online is often daunting. To make the transition easier, custom software developer ActionPoint partnered with Microsoft and Dell Technologies to develop IoT-PREDICT, an industrial IoT solution for predictive maintenance that incorporates machine learning, data analytics, and other advanced capabilities. IoT-PREDICT helps manufacturers quickly reduce downtime, lower costs, and increase the overall efficiency of their equipment and operations.

Factories make money by keeping their production lines running smoothly and manufacturing products without interruption or delay. However, many manufacturers struggle to keep their machinery operating efficiently and to avoid costly downtime. Part of the challenge is in the methods a significant number of manufacturers still use to monitor their machines, and the limitations such procedures place on their ability to detect small warning signs early and intervene before they become serious problems. 

The current system of monitoring key performance indicators such as temperature and vibration is heavily dependent on human observation and manual processes. At many factories, including some of the world’s most advanced, workers still walk around with handheld sensors, taking key readings and recording them on paper. This system inevitably leads to inefficiency and human error. Other systems produce mountains of data, but engineers don’t have time to review the data, nor do they have the tools to unlock the secrets the data holds. 

Many manufacturers recognize the business benefits of the Internet of Things (IoT) and are eager to implement industrial IoT solutions so that they can automate their equipment monitoring, analyze the data, and proactively make changes to increase efficiency on the factory floor. These manufacturers understand how IoT solutions can help them perform predictive maintenance, prevent costly downtime, and improve their overall equipment effectiveness (OEE)—the metric commonly used to measure the productivity of manufacturing machinery and processes. Unfortunately, many IoT solutions to date have been incomplete, dauntingly complex, or unsuited to harsh industrial environments. 

“We developed IoT-PREDICT because we were having a lot of conversations with manufacturing customers, and they were struggling,” says Ivan O’Connor, Head of IoT at ActionPoint, a Microsoft Gold Partner that provides custom software development services and innovative technology solutions to companies worldwide. “Either they didn’t have the right sensors in place to provide all the data they needed, or they had no effective way to analyze their sensor data and visualize the valuable business insights it could provide. We thought there was a relatively easy way to fix the problem.” 

Transforming manufacturing with smarter technology

ActionPoint partnered with Microsoft to develop IoT-PREDICT, an industrial IoT solution for predictive maintenance that incorporates machine learning, data analytics, and other advanced capabilities. IoT-PREDICT can help manufacturers and other companies reduce downtime, lower costs, and increase the efficiency of their operations. The IoT-PREDICT solution uses the Microsoft Windows 10 IoT Enterprise operating system running on Dell Edge Gateway hardware, and combines that with the best of the Microsoft Azure toolset to provide state-of-the-art edge computing. 

For industrial IoT solutions, Windows 10 IoT Enterprise provides enterprise-class power, security, and manageability, allowing organizations to deploy secure IoT devices that their IT teams can quickly provision, easily manage, and seamlessly connect to their cloud services as part of their overall cloud strategy. As well as working extremely well for manufacturers who need to keep their production lines operating at peak performance, O’Connor says that IoT-PREDICT is ideal for original equipment manufacturers (OEMs) who must monitor and manage machinery remotely, and facilities managers who need to monitor and adjust heating and cooling systems to keep their buildings and businesses running efficiently. 

“Because IoT-PREDICT is built on Windows 10 IoT Enterprise, it’s very easy for customers to deploy,” says John Savage, ActionPoint co-founder and Chief Technology Officer. “They’re essentially just installing a Windows 10 IoT device on their network, with the same familiar management and security framework that they already use to manage their other Windows-based devices.” 

With Windows 10 IoT Enterprise, and what Savage calls the “tried and trusted Windows environment,” an organization’s IT team can remotely configure and manage systems and devices on the network from their desktops using many of the tools they already use every day. And Windows 10 IoT Enterprise provides a TLS (transport layer security) encryption connection between the gateway and Microsoft Azure for enhanced data security—something that is increasingly important to many businesses in the digital age.  

Flexibility and easy scalability are other significant benefits of deploying IoT-PREDICT, according to Savage. Because IoT-PREDICT is powered by Windows 10 IoT Enterprise and Microsoft Azure,  manufacturers can start small with IoT and grow at their own pace. 

“The vast majority of manufacturing organizations are struggling to do meaningful industrial IoT,” Savage says. “IoT-PREDICT can give organizations a quick win. Rather than taking on a big, complex project, they can start their Industrial IoT journey very quickly with IoT-PREDICT. This is a very real benefit that our customers tell us they appreciate.” 

According to O’Connor, the combination of Windows 10 IoT Enterprise and Azure helped ActionPoint create a highly effective IoT solution that is easy for customers to deploy and start using within minutes. “Right from the start, we were committed to making the product work 10 minutes out of the box,” O’Connor shares. “We knew if we didn’t achieve that, we would have just one more complicated industrial IoT solution. Everything we did in the design process was focused on that goal.” 

O’Connor further explains that the Microsoft technology that ActionPoint used for IoT-PREDICT played a vital role in his company’s ability to accelerate its time to market while providing the out-of-the-box experience and other key capabilities its customers need for their smart factories. He cites the example of a German company that ActionPoint has been talking to about IoT-PREDICT; the company manufactures industrial systems that are sold worldwide. Currently, the company preconfigures its control systems to connect to a particular network router, so its customers must also install and manage this router to establish the connection to their system, which is a lot of added complexity. 

“Because IoT-PREDICT is built on Windows 10 IoT Enterprise, it’s very easy for customers to deploy,” O’Connor says. “They’re essentially just installing a Windows 10 IOT device on their network, with the same familiar management and security framework that they already use to manage their other Windows-based devices. So, it’s a much easier system to implement and manage.” 

Creating smarter factories by putting data to work

“With IoT-PREDICT, we introduce three sensors out of the box: vibration, temperature, and current,” O’Connor says. “We refer to those as ‘common denominator’ sensors, because they are process sensors that can infer a problem on almost any production line.” 

Since every manufacturing process consumes energy and generates heat,  and every moving part, whether it’s a compressor or a conveyor, has a normal vibration wavelength, O’Connor explains that “if a process is not working as designed, it will consume more energy, generate more heat, or produce more vibration. So, vibration, temperature, and current are leading performance indicators for the machinery in any factory.”  

Because ActionPoint built IoT-PREDICT with experienced production engineers in mind—people who understand industrial technology and have been running factory operations for many years—the company chose to use industrial-grade sensors and the 4–20mA (milliamp) protocol. IoT-PREDICT is designed to support any 4–20mA sensor right out of the box. So, for example, an engineer can take a 4–20mA flow meter they already have in their production line (or one they purchase from a supplier) and connect it to IoT-PREDICT, independent of the three sensors that come with the solution. 

IoT-PREDICT is also designed to support any sensor that uses the IO-Link protocol, which is quickly becoming a global standard for digital sensors. And IoT-PREDICT can integrate easily with customers’ existing systems, such as the SCADA (supervisory control and data acquisition) systems that many companies use to control their industrial, infrastructure, or facility-based processes. 

Another way that the IoT-PREDICT solution helps maximize the impact of manufacturer data: by using the Microsoft Azure IoT Hub to gather data and make it available to several Azure services, including Azure Time Series Insights, Azure Stream Analytics. The solution also uses Azure Blob Storage to store customers’ meta-tagged industrial sensor data in low-cost, long-term blob storage. And Microsoft Power BI can be easily configured to help customers further visualize and share data and insights.

Reducing costs and downtime and increasing OEE

For organizations that use IoT-PREDICT, the business value of the solution—and the predictive maintenance it enables—is in less downtime, reduced costs, and improved OEE. Downtime on a production line can result in millions of dollars in lost revenue each year. By predicting and fixing problems before they occur, manufacturers can potentially minimize downtime, save money, and get more productivity from their factories in return for their existing capital. 

Specifically, sensor and system data that IoT-PREDICT transforms into actionable insights helps manufacturers anticipate problems and resolve them more strategically and proactively. Those insights also give manufacturers critical information to help monitor and manage many other parts of their business, from reducing energy costs and consumption to trimming their maintenance overhead by taking the guesswork out of scheduled maintenance. 

“Many manufacturers have huge data repositories, but no way to use the data effectively,” O’Connor says. “Moving the data to Azure with IoT-PREDICT helps them gain critical insights by applying Azure Machine Learning, Time Series Insights, Stream Analytics, and other Azure services that are baked into IoT-PREDICT.”