Predictive maintenance has gained traction in recent years as one of the many applications enabled by the Internet of Things (IoT) that is revolutionizing manufacturing. At Microsoft manufacturing, we’ve worked with a variety of partners to deliver predictive maintenance solutions based on Azure IoT Suite and Azure Machine Learning.
In this two-part series, we will hear from our partners at BlueMetal, Dynamics Software, Harman, and ICONICS, whose responses are listed in alphabetical order below. In this first post, we’ll discuss what got them started with predictive maintenance, and what benefits they’ve realized for their customers and themselves. Edited excerpts follow.
What was your impetus for getting started with predictive maintenance?
BF, BlueMetal: Businesses need to find ways to keep up with born-in-the-cloud competitors and meet the increasing demands of customers, partners and employees. One way they’re evolving and transforming their businesses is by taking advantage of real-time data. There’s real potential for data to drive operational efficiencies as well as create new revenue opportunities.
WP, Dynamics: Many companies have made the shift from corrective to preventive maintenance in the past years. Especially in industries that require high capital investments in manufacturing equipment, the need to avoid breakdowns or unscheduled maintenance during production hours was imperative. Predictive maintenance is the next step for these industries.
PK, Harman: Our customers wanted real-time, 24/7 visibility into equipment performance so they could better track key metrics like mean time to repair (MTTR) and mean time between failures (MTBF) to assess the reliability of their systems. For us, predictive maintenance represented the ultimate commercial use case for the Internet of Things. Nothing offers the same value and service to customers in our industrial vertical. Our expertise in devices, mobility and analytics converged around industrial IoT solutions that could help our customers decrease MTTR and increase MTBF.
MT, ICONICS: In addition to staying on top of industry trends and the latest technologies, we closely monitor the challenges commonly faced by our customers. So, when they started expressing a desire to be able to address maintenance issues ahead of the curve rather than constantly putting out fires, we knew it was the time. We started looking at the historical data customers had already been collecting using tools like ICONICS Hyper Historian, and how that data could be put to better use than traditional trending and analysis. We worked with those companies to develop predictive algorithms, or fault rules, based on the most popular types of industrial equipment. This new concept, called Fault Detection and Diagnostics, provided the basis for our predictive maintenance solution.
What drove you to choose a predictive maintenance solution with Microsoft? What key differentiators about Microsoft’s approach made it a good fit for your business needs?
BF, BlueMetal: We believe that bringing predictive maintenance solutions successfully to market requires a platform approach. A platform approach implies you invest in people, process and tools to create a foundation of business and technology capabilities that serves as a building block to modern SaaS solutions. This approach makes it possible to support frequent release cycles, maximize device compatibility, and provide a multitenant authentication scheme that provides secure access to the underlying information. BlueMetal has been designing, developing, deploying and operationalizing IoT solutions based on the Microsoft Azure IoT platform for several years now. We standardized on the Azure cloud because you don’t have to manage virtual machines and perform operating system patches — the platform provides complete support for device management, registration, authentication, telemetry ingestion, and command and control of your connected products.
EV, Dynamics: Our longstanding relationship with Microsoft and the easy end-to-end Azure IoT Suite platform made it an easy choice. Everything Azure delivers is already integrated well with our existing systems. With the cloud-based technology Microsoft provides, we have all the tools from one supplier — from Azure to Dynamics 365.
PK, Harman: For us, we found that many control system vendors and manufacturers we’ve worked with were already adopting Microsoft technologies. In addition, the Azure subscription can be configured so that it keeps costs low for our customers to support their business needs.
MT, ICONICS: Once the Azure IoT Suite was announced, we recognized it would enable us to offer more advanced and intelligent algorithms, visualization, analytics and reports to our customers. It also gave us the ability to seamlessly and securely scale the solution while also taking advantage of the existing security features that we are familiar with as longtime Microsoft partners. Microsoft continues to innovate, evolve its capabilities and strengthen its services in the cloud. By partnering with Microsoft, we are right on the cutting edge — maximizing the value we can deliver to our customers.
What are the key manufacturing verticals you are working with for predictive maintenance solutions?
BF, BlueMetal: We worked with one of our customers, Weka Solutions, to develop an IoT-enabled device that keeps track of vaccines doses while keeping them fresh and secure. The Weka Health Solutions Smart-Fridge provides refrigeration and security for vaccines ranging in value from $100 to over $10,000 per vial. Collecting real-time inventory and expiration data, Weka could provide data visualization of inventory levels for customer care personnel and automate replenishment orders. Using predictive analytics, clinics are alerted to upcoming vaccine shortages at specific locations or in certain areas so they can plan accordingly.
EV, Dynamics: We serve customers ranging from process and discrete manufacturers to industrial equipment manufacturers to field service organizations. We worked with a company that rents out generators to predict the best time for maintenance. Generators were either used for just an hour as a backup or they ran 24/7 for several weeks. By developing a solution based on Azure IoT Suite, Dynamics could help them shift to predictive maintenance based on actual usage rather than scheduled maintenance that was too early or too late.
PK, Harman: Harman’s solutions serve a broad range of customers ranging from semiconductor manufacturing, oil and gas, power, and control systems makers serving multiple end verticals. Our deep knowledge of the industries we serve, along with our over-the-air capabilities and analytics expertise, is helping our customers capitalize on the growing IoT market.
MT, ICONICS: One of the earliest applications of our predictive maintenance solution was with an electric utility company. It used the technology to forecast demand and optimize its operations during peak and off-peak hours. The more common use case, predicting equipment failures before they occur, spans many different verticals because it is a challenge shared by virtually all industrial automation customers.
What benefits from predictive maintenance have you seen customers experience that they didn’t expect upfront?
BF, BlueMetal: Our customers can create new monetization models by delivering new services. Shifting from being solely “product-based” to “service-based,” customers are adding new services that complement their existing product offerings. An example of adjacent new services could include provisioning consulting to customers based on real-time insights on how products and equipment are used. These insights were not possible before the Internet of Things.
EV, Dynamics: Adding on to that — we’ve seen customers implement measurable changes to their business models. With IoT, they can virtually eliminate return visits and no-shows by improving asset and information management and customer engagement.
PK, Harman: With a single view of customer operations and real-time visualizations, we’re able to provide predictive insights into detailed equipment performance. Having insight down to the individual component level enables our customers’ decision-makers to make more informed decisions around capital management.
MT, ICONICS: In contrast to a traditional preventive maintenance system, predictive maintenance solutions enable customers to strategically plan their maintenance tasks and group them in a way that allows them to perform the required maintenance more efficiently. By carrying out those work orders only as frequently as they are needed, rather than on a strict schedule, it results in a lower number of maintenance tasks over time. This leads to even more dramatic savings in terms of labor and maintenance costs, as well as an accelerated ROI. Another benefit that customers do not necessarily expect upfront is the potential savings resulting from integrating the ICONICS dashboard with their existing CMMS or work order management system. Being able to initiate work orders directly from the predictive maintenance system, or in some cases automatically based on certain rules being met or exceeded, saves the operator from having to switch between systems — further reducing labor and maintenance costs.
As you can see, our partners are creating innovative predictive maintenance solutions to help customers across a variety of verticals improve their business in numerous ways. From driving operational efficiencies to creating new revenue opportunities, leading manufacturers are adopting a predictive maintenance strategy on their journey to digital transformation.
Not convinced of the value of a predictive maintenance solution? Be on the lookout for part 2 in our Partner Perspectives series. In it we’ll discuss the concerns and barriers to implementing predictive maintenance solutions partners are seeing across their customer accounts.
For more information about how IoT can transform your business, visit www.InternetofYourThings.com.