WHITE PAPER

A Telekom, Intel, and SAP Partner Solution Specialty machine manufacturing is the designing and manufacturing of customizedmachines and equipment. Applications range from machines for constructionand food processing to packaging systems for the pharmaceutical industry. It isessential that this production equipment has a long service life, minimal downtime,and low maintenance costs. This is where predictive maintenance–methodsthat help determine the condition of in-service equipment to predict when main-tenance is needed–plays a role. Its source of data comes from sensors relatedto performance, temperature, or rotation speed. Businesses and manufacturersanalyze this data so they can more quickly identify parts that are fatigued or wornand need to be replaced, thereby proactively preventing costly production down-time. Furthermore, such analyses reduce the cost of service and maintenance.The most important requirement for efficient data analysis is a big data infra-structure capable of evaluating a large volume of data in real time. Take, forexample, Telekom's SAP solutions, which can also be used by medium-sizedenterprises, and are available in the classic sense on-premise in the data cen-ter, or in the security-enabled Telekom private cloud. The core components ofDynamic Services for SAP HANA* include the SAP HANA in-memory databaseand the latest generation of the powerful Intel® Xeon® processor E7 family.

Industry 4.0 is big data in its purest form.Sensor data from industrial equipment,telematics data from vehicles, and RFIDdata from the logistics chain–with thedigitalization of all commercial and tech-nical processes, the volume of data thatcompanies must manage continues togrow. If it is properly prepared, filtered,and analyzed, this data is an invaluablesource of information for any enterprise.This is also true for manufacturing com-panies, in particular those connected tothe fourth industrial revolution.

The term "Industry 4.0" denotes the useof the Internet of Things (IoT) in the in-dustrial sector. Its aim is to create inter-connected, sensor-supported and nearlyself-controlling production processesusing intelligent machines, components,and systems, and all of these processescan be automated according to customerspecifications. Intelligent componentscommunicate autonomously, for example,

via RFID chips, with the equipment. Theythen find their way to the next machineon their own, and even conduct qualitycontrol at the end, meaning customizationsare possible, for example, in the manu-facturing of sports shoes, automobiles,or clothing (lot size 1).

Since the machines, sensors and productsare interlinked, an abundance of unstruc-tured data indicating the status of a machineor equipment is generated in the digitalfactory, along with process and productdata. By intelligently monitoring and an-alyzing production data, even medium-sized companies can respond to changingparameters in near realtime and optimizeproduction accordingly. Furthermore, it ispossible to identify machine maintenancerequirements early on and prevent costlydowntime (predictive maintenance). Evenproduction-related tolerances can becaptured and used in the quality controlof the end product.

„Predictive analytics predictsthe probability of future

events occuring. To do this, ituses sophisticated algorithmsor statistical methods andmathematical calculationsto identify correlations andespecially trends in data sets.“

Predictive maintenance reduces costsPredictive maintenance, that is, proactiveservicing or maintenance, is a key applica-tion for predictive analytics in the industrialsector, and a major theme in specialty ma-chine manufacturing. Predictive analyticspredicts the probability of future eventsoccuring. To do this, it uses sophisticatedalgorithms or statistical methods and math-ematical calculations to identify correla-tions and especially trends in data sets.

In addition to classical data mining meth-ods such as clustering or regression analy-ses, simulation processes, elements ofgame theory, and machine learning alsoplay important parts. In machine learning,the algorithms learn from examples andexisting data over time, create a data model,and use it to predict and make decisions.Predictive maintenance is much the same.

The data comes from machine and sensordata that indicates the status of equipment,such as its performance, temperature,rotation speed, and capacity. This data isthen (usually) transferred to a cloudplatform. The solution analyzes this in-formation to identify error patterns andqualitative substandard components, andcan then predict errors. The predictivemaintenance systems can thus predictthe need for expensive repairs or seriousfailures and initiate preventative actionsbefore major damage occurs. Service canrespond quickly and replace a specificpart or perform maintenance work aheadof schedule.

The manufacturers of this equipment caneither use this data themselves or makeit available to their customers. With thehelp of predictive maintenance, businessesminimize the probability of expensive ma-chine downtime, extend the operational

life and service life of the equipment, andreduce costs for service and maintenance,thereby avoiding unnecessary maintenance.The quality of the manufactured goodsimproves and machine throughput in-creases. Finally, parts subject to wear, suchas air and water filters, are replaced onlywhen absolutely necessary and not routinelyor at scheduled maintenance intervals.

Analyzing large volumes of data inreal timePredictive maintenance faces two chal-lenges: The data must first be pooled fromthe most varied sources and then analyzedin real time. The company's ERP systemplays an important role here, functioningas the integration hub. It collects marketdata, customer information, and supplierand product data, and links it to themanufacturing and logistics data fromproduction and the supply chain, alongwith sensor data from the machinesthemselves. This data must fulfill themost critical quality requirements, suchas completeness, correctness, uniformity,and consistency–and it must be up-to-date as well.

The volume of data and information willcontinue to grow as Industry 4.0 takes hold.That is why ERP and big data systems needa powerful infrastructure they can use toquickly process large volumes of data,preferably in real time. In-memory data-base technologies such as SAP HANA offerthe solution to these challenges. Sincein-memory databases store informationin their main memory and not on harddrives, queries from multiple servers canbe processed simultaneously and withvery low latency. This is an important re-quirement for analyzing data in real time

White Paper | A Telekom, Intel, and SAP Partner Solution

and optimizing production processes inthe industrial sector.

As a market leader in ERP systems, SAPstands ready to support Industry 4.0 withproducts and solutions, such as the ap-plication suite SAP S/4HANA* or SAPPredictive Maintenance and Service*. TheERP suite SAP S/4HANA is based on theSAP HANA in-memory platform, whichanalyzes information from the most di-verse sources in one single database innear real time. This makes it possible forSAP S/4HANA to process multiple mate-rial postings simultaneously. Since thisprocess always provides current data onmaterial requirements and the existingmaterial that is available for productionplanning and control, manufacturers canrespond immediately to emergency sit-uations and avoid delivery delays.

SAP Predictive Maintenance and Serviceenables service technicians to record andanalyze a machine or system's real-timedata, such as temperature, rotation speed,or vibration, through data connections.Costly travel for technicians is often elimi-nated because this method transcendscountry borders and continents. If thetool identifies certain patterns or if de-fined thresholds are exceeded, an alertis issued signaling that action should betaken. The answer could be, for example,an inspection or machine repair. Theservice technician can then decide if asite visit is necessary or if the work canbe performed by operating personnel.

The solution–Dynamic Services forSAP HANABoth SAP S/4HANA and SAP PredictiveMaintenance and Service are available forlocal installation (on premise) or as a cloud

solution. One of SAP's most importantand experienced partners worldwide isDeutsche Telekom, together with its ITservice provider, T-Systems. Telekomoffers traditional SAP solutions formedium-sized companies as well as acloud hosting platform, or both as acombined hybrid solution. Since 2004,SAP's partner has employed DynamicServices for SAP Solutions to operatecomplex SAP landscapes for global com-panies in all industries from a certifiedand secure private cloud.

Since 2013, Dynamic Services for SAPHANA has also given SAP HANA in-memorytechnology the edge in the quick analysisof data in real time as a cloud offering.Dynamic Services for SAP HANA makesthese technologies available as standard-ized, industrialized and integrated solutionelements. This means businesses in (spe-cialty) machine manufacturing benefit fromthe following advantages:

• Rapidly deployable: State-of-the-artin-memory technology without majorinternal investment and operationexpenditure

• Scalable: Easily customized to fulfillcustomer requirements

• Reduced costs: Billing according toactual use

• Solution for test operation: Free licensefor three months to test developmentenvironment

• High availability: Some service levelagreements with 24/7 availability, 365days a year, if required, customer systems are mirrored in the securetwin-core data centers

• The cloud "Made in Germany": Internalphysical resources in Telekom's highlysecurity-enabled data centers in Germany,regulated by the German Data Protection Act

White Paper | A Telekom, Intel, and SAP Partner Solution

The benefits of predictive mainte-nance at a glance

• High availability of systems throughreduction of unplanned downtimes

• Proactive maintenance preventsequipment failure

• Reduction of maintenance costs

• Longer service life for equipmentand machines

• Comprehensive consulting: Telekom'sASP and cloud experts stand ready tooffer consultation, planning, and imple-mentation support

• Powerful hardware: By employing thelatest generation state-of-the-art IntelXeon processor E7 family, SAP HANAinstallation fulfills all requirements

Intel’s fast processors are at the coreThe foundation for Dynamic Services forSAP HANA is the latest generation of theIntel Xeon processor E7 family, with upto 18 cores and up to 1.5 TB of memorycapacity per socket. With four to eightsockets, adequate servers offer sufficientmemory for SAP HANA. In-memory data-base performance is also enhanced thanksto Intel® Transactional Synchronization Extensions (Intel® TSX). Transactional memory or TSX, ensures that interdepend-ent threads rarely encounter locked memoryareas.

The additional options offered by the Intel®Advanced Vector Extensions 2 (Intel®AVX2) 2560bit vector integer instructionsand Fused-Multiply-Add instructionsimprove performance even further. Toensure that systems operate without in-terruption, Intel has improved RAS (relia-bility, availability, scalability) features forthe Intel Xeon processor E7 platform.For example, only areas with data aremirrored in the memory. In the future,Telekom will equip servers with the lat-est generation of Intel Xeon E7 v4processors. They offer twice the mem-ory capacity of their predecessor withthree TB per socket, making the solutionideal for SAP HANA.

Gateways for the integration ofexisting systemsThe majority of legacy systems in theindustrial sector are not interconnectedthrough the Internet. That is why com-panies face the challenge of connectingnew systems to old devices and the cloudfor secure data exchange. With gatewaysolutions for the IoT, Intel offers integrated,prevalidated hardware and software so-lutions with open standards and interfacesto ensure interoperability. These solutionstranslate between logs and normalizeddata to prevent redundancy. The gatewaysare fixed to sensors or equipment, collectdata, and transfer it securely to the net-work or cloud for analysis.

©2017 Intel Corporation. All rights reserved. Intel, the Intel logo, the Intel Inside logo and Xeon are trademarks of Intel Corporation in the U.S. and/or other countries.

*Other names and brands may be claimed as the property of others. 1116/JNW/RLC/XX/PDF

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration.No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at www.intel.com/content/www/us/en/processors/xeon/xeon-processor-e7-family.html.

Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of information to evaluate performance as you consider your purchase. For more complete information about performance and benchmark results, visit www.intel.com/benchmarks.

White Paper | A Telekom, Intel, and SAP Partner Solution

About Deutsche TelekomDeutsche Telekom is one of the world'sleading integrated telecommunicationscompanies, with some 156 million mobilecustomers, 29 million fixed-network lines,and more than 18 million broadband lines(as of December 31, 2015). The Groupoffers fixed-network/broadband, mobilecommunications, Internet, and IPTV prod-ucts and services for consumers, and in-formation and communication technology(ICT) solutions for business and corporatecustomers. Deutsche Telekom is presentin more than 50 countries. With a staffof some 225,200 employees throughoutthe world, it generated revenue of 69.2billion euros in the 2015 financial year,about 64 percent of it outside Germany.

About SAPAs the market leader in enterprise appli-cation software, SAP SE helps companiesof all sizes and industries run better.SAP applications and services enablecustomers to operate profitably, adaptcontinuously, and grow sustainably. Fromback office to boardroom, warehouse tostorefront, desktop to mobile device, SAPempowers people and organizations towork together more efficiently and usebusiness insight more effectively to stayahead of the competition. Approximately320,000 customers from the private sec-tor and public administration rely onSAP applications and services to achievetheir goals more efficiently and effectively.For more information, visit www.sap.de.

About Intel Intel (NASDAQ: INTC) redefines theboundaries of technology to make themost amazing experiences of the futurepossible. For more information about Intel,please visit http://www.intel.de/newsroomand http://www.intel.de.