One June day in Virginia last year, an airplane was grounded by an unlikely adversary: a large swarm of bees. The peculiar story made for great newspaper headlines and serves as a reminder that even with the best technology and planning, some things are truly unexpected. But fortunately, most aircraft delays are caused by far more predictable issues than an unwelcome swarm of bees nesting in a turbine. Airlines, like most asset-intensive businesses, are getting increasingly better at predicting failures and anticipating maintenance problems. Rather than keeping planes grounded for costly and annoying last-minute maintenance — or, worse, exposing passengers to the risk of flying on a faulty aircraft — airlines are investing in cutting-edge technology that detects potential problems before they arise.
Predicting what was once unpredictable
Successful companies know that it pays to get ahead of problems. This doesn’t mean more scheduled, structured maintenance — it means performing maintenance at just the right time. The problem with scheduled maintenance is the guesswork — parts might be replaced before they need to be under the banner of minimizing downtime. To truly be predictive, companies require a real-time understanding of how assets are performing in the field, how their surroundings impact performance, and how those assets compare against similar assets.
Comparing current conditions with the historical data, and using that analysis to predict subtle, early signs of a future problem, help prevent problems and avoid guesswork. Data-driven insights enable companies to transform their approach with predictive maintenance — enabling organizations to automatically trigger preventative actions when certain conditions occur.
This delivers numerous benefits, like cost savings and greater uptime — but that’s not all. Organizations are also using the underlying information to enhance customer engagement and differentiate their businesses. Let’s now take a look at how predictive maintenance and related proactive services are helping to address business needs across a range of scenarios.
Creating a variety of benefits for your business
Increasing quality, reducing waste and saving energy
Leveraging data from across locations and assets enables manufacturers to manage operations in a completely new way. With assets that continually generate performance data, it becomes possible to closely track variations, create alerts and preprogram workflows to respond to changing conditions. By using data to fine-tune processes and automate responses, businesses can reduce energy consumption and prevent costly waste and rework.
Jabil, one of the world’s leading design and manufacturing solution providers, wanted to minimize downtime by creating digital, intelligent and predictive factories. The company connected its machines to Microsoft Azure Machine Learning. Once connected, it began capturing millions of data points and running them through machine-learning algorithms. These predictive analytics capabilities enabled them to catch issues before they even occur. With the ability to identify small deviations in performance, Jabil can now prevent big problems down the line — leading to higher product quality and lower failure rates. After incorporating Azure IoT Suite into its business, Jabil could predict with 80 percent accuracy when machine processes would slow down or fail, which reduced the cost of scrap and rework by 17 percent. It also experienced an energy savings of 10 percent.
Proactively detecting failures to maintain uptime
Asset performance data is not just a source of process optimization or quality assurance. The information from asset monitoring and maintenance solutions also ensures production uptime. For instance, with various machinery involved at all stages of production, one of these assets going down can result in millions of dollars in lost production. KUKA, a robotics manufacturer, runs a connected factory that produces different Jeep Wrangler bodies on the same production line. To deliver a new car body every 77 seconds, KUKA needed continuous, 24/7 uptime without interrupting production flow. After connecting 60,000 devices and robots to the Azure platform, KUKA gained better insight into equipment status. This was the foundation for predicting when a robot might fail and what component was at risk. As a result of this predictive insight, KUKA increased productivity and reduced downtime.
Enhance customer engagement with contextual and responsive services
Customer expectations have evolved. The on-demand, always-on service experiences available from digital businesses are now shaping the way customers think about all services they receive. Customers expect businesses to be highly responsive. These expectations are forcing an evolution from the break-fix model of service. To meet these changing demands, manufacturers like Rockwell Automation are now providing proactive monitoring and predictive maintenance services.
The petroleum supply chain runs 24 hours a day, and getting to the various equipment in that chain takes time. Rockwell Automation wanted to differentiate itself from competitors. Using Azure IoT Suite to monitor its expensive, remote equipment, Rockwell can now offer proactive, data-driven advice to its customers. With full visibility into equipment performance, it’s now easy to alert customers to potential issues and give them precise insights into how to address them. In an industry where a single pump failure in an offshore rig can cost a company $300,000 a day in lost production, these insights are of significant value to Rockwell’s customers. After partnering with Microsoft, Rockwell significantly reduced customers’ troubleshooting efforts and has made the design manual consult a thing of the past.
Support development of innovative and differentiated business models
It’s also possible to use data from connected devices to create new services, new monetization models and even new product designs. By analyzing product data, organizations can better identify which features are used most, how performance and failures vary from user to user, and whether customers are misusing the product and could benefit from training. Insight into these areas enables businesses to reduce costs, innovate faster and open up new revenue streams — all with less money spent on R&D.
For example, Rolls-Royce wanted to improve aircraft efficiency and reduce maintenance costs for its engines, which power more than 50,000 flights around the world each month. With ever-increasing volumes of data coming from all the different sensors on an aircraft, Rolls-Royce needed a better way to get ahead of customers’ maintenance needs. Rolls-Royce partnered with Microsoft to collect and analyze a wide variety of data in real time, including engine health, air traffic control, fuel usage and more. Now, Rolls-Royce not only has a strong approach to predictive maintenance, it has developed a line of new service offerings based on the visibility and insight it gained. These offerings help its airline customers determine optimal actions for saving money and making other operational improvements.
Get started today
We all know that some problems are truly unexpected — the flight crew grounded by bees can certainly testify to that. But fortunately, surprise mechanical delays are increasingly becoming a thing of the past. Learn how predictive maintenance can transform your business today.