Today’s asset-intensive companies collect a lot of data along their value chain. Suppliers are providing data on their products and shipments. While new assets are getting smarter and can communicate status and performance data using internet protocols, even older machinery equipped with new sensors are constantly churning out data from the shop floor. Moving from a product into a service world, customers are also providing data on machine performance and maintenance to service providers or even to the original equipment manufacturers (OEM), in business models in which they only pay for the performance of the asset. And service technicians out in the field are using and generating data on their site visits. The amount of information generated in a typical industrial ecosystem today is truly remarkable.
For example, GE’s next-generation turbofan engines found on Boeing 787 and 747-8 aircraft contain 5,000 data points that are analyzed every second. To put that into perspective, a single cross-country US flight generates 240 terabytes of data, which may result in hundreds of petabytes of data for one airline operator. While not all of this is transferred to the ground and kept for analysis, the challenge remains for organizations: How do they navigate through these oceans of data?
Data that is collected and simply sits in a system, database, or even the cloud is not going to improve your business. This data needs to be translated into actionable information and improve knowledge, so it can be used for better decision making. Because at the end of the day, humans — not data warehouses — are driving the decisions to power your business.
In this article, we’ll review some of the most recent ways in which big data is being made accessible to business users in asset-intensive industries and also look at how those users are leveraging the data to improve both planning and execution through predictive maintenance.
How Mobile Changed … Everything
Manufacturing and asset-intensive industries were well out in front of the mobile device curve. Long before there were smartphones and tablets in every business user’s hands, field service technicians were using mobile devices to measure asset performance and quality degradation in the field. But the process was extremely slow. A technician could review data from the machine and use a device to record that data, but would have to bring the device back to a shop to load it into the system and compare that machine’s performance to others or its own historical performance. And then, when business decisions were made, the technician would have to travel back out to the machine to make any repairs or updates. If the technician needed certain parts once onsite, that required another trip back to the office or parts warehouse. In certain industries, this extended cycle produced longer downtimes and lowered the return on investment (ROI) on those very expensive assets.
Smart devices and mobile technology have changed many of those processes. Today, for example, using applications based on SAP Mobile Platform, a service technician can travel to an asset in the field with a mobile device and have all the data needed to maintain that asset, including work orders and repair instructions visualized in 3D, so the technician can accurately fix any issues. If a new part is required, the technician can research the options and initiate the ordering of the part at that moment. The technician can take photos of the asset with the same device to document repairs that are needed or that have been made. If there is network coverage, all of that data can be directly synchronized with the back-end system, but even when lacking connectivity, activities can be completed offline.
While the use of mobile devices is obvious in the field, even within a plant mobile devices and offline capabilities are often needed to monitor assets and performance and escalate any production issues identified through the proper channels. Applications and content can be tailored to the technician’s needs to ensure the latest information and best practices are employed in keeping production up and running. Mobile access to analytical data provides technicians knowledge of what is going on at the shop-floor level so they can assess it against historical or benchmark data and optimize asset performance.
With mobile devices across all industries providing data on demand, the question then becomes, what’s next? What could possibly improve upon accessing data on a device whenever you request it?
The next wave is having the right analytics presented to the right user instantly, rather than requiring a query. This is the thinking behind the concept of “augmented reality.” Imagine a production manager walking through a manufacturing plant wearing a pair of smart glasses that can scan items as they are viewed and instantly display important data about those items. When the manager turns to look at a certain asset, an alert shows the asset is in need of service soon. When a production line is viewed, data is displayed to tell the manager that this line is ahead of its daily quota.
As SAP demonstrated at SAP TechEd in Las Vegas last year, a service technician can scan an item and instantly know all of its technical data, including the equipment number, the status of the asset, and more (see Figure 1). And rather than having to figure out where to go and what to fix next, by combining data sources, the technician can be simply guided there by a system that uses both facility and GPS data.
Of course, the key to the effectiveness of augmented reality is the presentation of data. With so much data available, understanding what analytics are most important and how they should be displayed is critical to the success of mobile and augmented reality initiatives. Smart glasses are only one way this data can be easily displayed to users in manufacturing and other asset-intensive industries.
Business analytics have come a long way in a relatively short time. It was not long ago that business users would have to submit a formal request into their IT organization to run a historical report to get the most updated information available. And often by the time that report was completed, the data was out of date. Today, that data can be made available directly to the business users with real-time analytics.
But access — even instant access — to the right data may not be fast enough for some asset-intensive organizations. If a production manager has to wait until an asset fails or its performance degrades before repairing or replacing it, then valuable time is lost. That production manager wants to know that an asset is in need of service before it fails, so the maintenance can be scheduled at the right time and the downtime can be minimized or avoided altogether. That type of data has become available, but processing it was the challenge — running smart algorithms to process massive amounts of data often took hours.
By combining the wealth of data available with analytics applications and in-memory computing, asset-intensive industries can employ predictive maintenance strategies that increase uptime and speed ROI on assets and equipment. For example, not only can sensors track a machine’s performance degradation and provide an alert when the machine’s performance dips below a predetermined threshold, but they can also forecast the machine’s condition based on predictive algorithms and statistical models. The asset maintenance can then be scheduled during a convenient time so that production is not being impacted, which reduces downtime and improves safety. In a very large organization with hundreds of machines to track, the ability to prioritize maintenance based on the actual and forecasted asset condition — balanced with the workforce’s capacity — significantly increases operational performance while lowering risk and cost.
With powerful new analytics applications based on SAP HANA’s in-memory computing power on the back end and more user-friendly interfaces on the front end, data analysis based on advanced models and algorithms can be conducted quickly to improve on overall equipment effectiveness. A predictive report comparing two data sets and projecting them into the future — an exercise that might have taken hours to process in a traditional data environment — can be done in seconds in memory, facilitating instant access to actionable knowledge, rather than simply access to huge sets of data that might have been aggregated for a different purpose. Figure 2 shows how SAP Predictive Analysis can help users visualize data and gain better insights on their information.
To close the loop, the analysis on the machine’s uptime and performance quality can be tracked and then integrated into the product design to improve the next generation of the machine. If the predictive analytics show a pattern of an asset failing more often than it should or under specific conditions in the field, that data can be used by the manufacturer to improve the product’s design or installation processes, lowering the number of both warranty cases and spare parts in stock. By collecting and sharing this data, the overall value chain from supplier to owner and operator becomes more integrated.
Today in most industries and organizations, data is swirling all around. Your organization has the choice of simply drowning in the data or leveraging it to make informed business decisions. By combining the latest mobile apps, augmented reality capabilities, and predictive analytics powered by SAP HANA, businesses can finally get the right information to the right users in the right format and at the right time. This is the strategy that will help your employees make better business decisions and achieve operational excellence.
To find out more about these topics, visit www.sap.com/idea-performance.