Photo-K - Fotolia

Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

How can SAP's predictive maintenance help manufacturers?

From react to prevent to predict -- here's a look at how SAP has taken maintenance to the next level and how S/4HANA can lead to fewer machine breakdowns.

With the cloud version of S/4HANA Asset Management, a line-of-business product within the S/4HANA Enterprise Management umbrella, companies can truly make progress in their quest for less machine downtime through the product's predictive maintenance functionality.

Companies have long realized that one of their core assets is their equipment and machinery and, more to the point, keeping those working at an optimal level. This ensures uninterrupted production, not to mention smoother logistics and supply chain management. Indeed, taking a proactive approach to machine maintenance -- or better yet, a predictive one -- leads to reduced machine downtime, lower maintenance costs and higher return on assets.

S/4HANA Asset Management helps realize that potential by attending to maintenance management and business processes by leveraging the constant stream of data flow from sensors, machine tagging, wireless technology and predictive algorithms. The available predictive analytics in S/4HANA Asset Management enables maintenance planners to gain a holistic view of maintenance planning and, in turn, helps ensure effective planning of required spare parts and technical resources and manpower.

The growing availability and lowering costs of sensor devices and associated technology have enabled companies to install sensor devices on all key equipment and machinery that then stream live data of a machine's condition to an SAP ERP system. Maintenance planners ensure that this data streaming from the actual and physical equipment also matches the company's assets created in the SAP ERP system. With the help of the predictive analytics and algorithm, the Asset Management predictive maintenance system is able to intelligently prepare a maintenance and service schedule to keep the equipment in optimal condition. This in turn also leads to higher OEE (overall equipment effectiveness), which measures the asset availability together with its quality and performance. It also leads to effective benchmarking of equipment to see, for example, why there are fewer machine breakdowns at one customer's site as compared to another and be able to identify and trigger root cause analysis.

While this predictive maintenance may just be the beginning of S/4HANA Asset Management offerings, companies can consider also purchasing SAP Business Network, which brings together three different cloud-based SAP companies, Ariba, Concur and Fieldglass. While Ariba is a cloud-based procurement product that companies can use to procure spare parts, Fieldglass offers temporary and contract workforce management that companies can use to plan maintenance activities, while Concur caters to travel and expense management needs of a company and can be used to ensure timely availability of the key maintenance resources required. 

Next Steps

The scoop on Fieldglass purchase and what it means

Expert sourcing and procurement tips

SAP gets Concur, but who loses?

Dig Deeper on SAP manufacturing

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

In what ways could predictive maintenance be most helpful to you?

predictive maintainnance will helps because of the using maintaince in table maintianance generate. its will fetch only single table data what you are choosed



One asks oneself a question. And then I read marketing fluff. HANA improves Equipment uptime.  :) - at what effort, costs and SP implementations ?

HANA should be used do to some predictive maintenance for SAP Systems first instead of letting customers run into system break-downs followed by a tedious SAP Notes search and try and error application of the same. 
That might be helpful as a real value add. Once that works we can look further.
I can share some reality experience having done many PM implementations. Predictive Maintenance needs specialized sensoric systems,  historians and part specific rules. Who puts that into HANA ? 
Thank you for your insights! 
The cheaper sensor technology, equipment tagging and IoT available now is making all possible to stream data into SAP HANA so that it is able to do predictive maintenance. 
The real-time data has to get into SAP HANA to enable it to begin doing its share of work. 
putting all the historian data into HANA is one thing. But the very specialized analysis system for thousands of different equipment types also have very specific prediction rules, trend analysis, add. measures like temperature, throughput or any other type of additional factors. How does that get into HANA? Especially since they are proprietary in most cases.
I don't believe e.g.SKF will give away its logic to kill it's analysis tool sales. There is a lot of know how involved SAP that does not have at all. Why not leave it like it is. Let the historians trigger maintenance activities via the SCADA interface (like they are doing since 1995). The other question is, can SAP display historian information  graphically the way historians can do it? With all the bells and whistles?
HANA is just a dumb maybe fast DB so far. Is the architecture there to support all the necessary transformations? E.g. Fourier analysis, analog signal compression, etc.
I am curious to see some real examples beyond simple temperature, pressure, simple noise (dB) measures.
Still the manufacturer's Specs have to be manually implemented into SAP. I have not seen any of this in R/3 without significant custom coding.
But I curious to learn more about some real examples incl. their ROI.

sap should maintained based on hana server instead of sql or rdbms.

but their is no big difference while at time of hana migration standard code should be changed. as per the changes the logic while at the time of coding some conditions are applied for performance analytic.