rfid white paper technology, systems, and applications

1 of 50

RFID White Paper Technology, Systems, and Applications An overview for companies seeking to use RFID technology to connect their IT systems directly to the “real” world.

2 of 50

Publisher BITKOM German Association for Information Technology, Telecommunications and New Media e.V. Albrechtstraße 10 10117 Berlin-Mitte Germany Tel.: +49 (0)30/27 576 – 0 Fax: +49 (0)30/27 576 – 400 [email protected] www.bitkom.org Compiled by RFID Project Group German Version published August 2005 English Version published December 2005 Contact person: Dr. Birgit Heinz, BITKOM e.V. Tel: +49 (0)30/27 576 – 243 E-mail: [email protected]

3 of 50

Table of Contents 1 Foreword..............................................................................................................................5 2 Introduction ..........................................................................................................................7 3 Management Architecture for RFID Solutions .....................................................................9 4 RFID System Architecture and Standards .........................................................................11

4.1 Market Trends and Usage ......................................................................................... 11 4.2 Architecture of RFID Systems ................................................................................... 12 4.3 The Internet of Things Envisaged by EPCglobal ....................................................... 14 4.4 RFID Standards......................................................................................................... 19 4.5 Outlook ...................................................................................................................... 19

5 RFID System Components ................................................................................................20 5.1 RFID Tags: Chip and Antenna .................................................................................. 20

5.1.1 Structure of an RFID Tag/Transponder.............................................................. 20 5.1.2 Tag Antenna....................................................................................................... 22

5.2 Air Interface and Bulk Reading.................................................................................. 22 5.3 Reader and Antenna ................................................................................................. 24

5.3.1 Reader Structure ................................................................................................ 24 5.3.2 Reader Antenna ................................................................................................. 27

6 IT Infrastructure..................................................................................................................29 6.1 Middleware and Service-Oriented Architecture (SOA) .............................................. 29 6.2 The Layer Model for the Real-Time Enterprise.......................................................... 29 6.3 Event Architecture ..................................................................................................... 31 6.4 EPCglobal’s Savant Middleware Concept ................................................................. 34 6.5 Databases ................................................................................................................. 34

7 RFID Security Aspects.......................................................................................................36 8 Sample Applications ..........................................................................................................39

8.1 Case Study: Logistics Processes at Hewlett-Packard GmbH.................................... 39 8.2 Case Study: Flexible Automotive Production Processes at BMW............................. 39 8.3 Case Study: Mobile Maintenance Solution at Fraport AG ......................................... 41 8.4 Case Study: Large-Scale RFID Roll Out for Metro AG Logistics............................... 43

9 Glossary.............................................................................................................................45 10 Other Sources of Information.............................................................................................47

4 of 50

Table of Figures Figure 1: The Layers and Components of an RFID Management Architecture .........................................9 Figure 2: System Architecture for RFID Applications ...............................................................................12 Figure 3: The Key RFID Frequency Ranges and Their Applications .......................................................13 Figure 4: Example and Structure of an EPC Number ..............................................................................15 Figure 5: Structure of the EPCglobal Network Compared with the Web..................................................15 Figure 6: Questions Solvable by the EPCglobal Network in the Future...................................................16 Figure 7: Gillette’s EPCIS Service............................................................................................................17 Figure 8: EPCglobal Network Standards..................................................................................................18 Figure 9: ISO and EPCglobal Standards..................................................................................................19 Figure 10: Inlays for RFID Tags with 13.56 MHz (HF) .............................................................................20 Figure 11: Inlays for RFID Tags with 868 MHz (UHF)..............................................................................21 Figure 12: Reflections in the Reader Antenna Range..............................................................................23 Figure 13: The RFID Reader Anti-Collision Protocol................................................................................24 Figure 14: Typical Connection Scheme for a Reader with RFID Antennas .............................................25 Figure 15: Bulk Reading and Multi-Tag Handling.....................................................................................26 Figure 16: Overlapping RFID-Reader Fields............................................................................................26 Figure 17: Reader Antenna with Antenna Array and Field Regions ........................................................27 Figure 18: Layers in a Software Architecture for the Real-Time Enterprise.............................................30 Figure 19: Implementation of Global Data Synchronization (GDS)..........................................................31 Figure 20: Auto-ID Event Architecture......................................................................................................32 Figure 21: Event Information in an Automotive Process Chain................................................................33 Figure 22: EPCglobal-Oriented RFID Application and Communication Structure ...................................35 Figure 23: Goods Issue with RFID Antennas at HP’s Chester Plant .......................................................39 Figure 24: Basic Architecture for RFID in Industrial Manufacturing .........................................................40 Figure 25: RFID Transponder with Data Store on the Body Panel at BMW ............................................40 Figure 26: Paperless Maintenance Scenario at Frankfurt Airport, Optimized by RFID............................42 Figure 27: Layer Model for a Comprehensive Auto-ID Application Architecture......................................43 Figure 28: Configuration of RFID Antennas on Doors in the Metro Warehouse......................................44

5 of 50

1 Foreword Dear readers, Allow me to take you back in time, back to the Greek city of Abdera. It was here, around 2,400 years ago, that the philosopher Democritus was setting out his cognitive theories. It was his belief that the objects of this world emit images, or “Eidola” – idols we would say today, or more simply: IDs. He proposed that it was only by perceiving these idols that men was able to assign characteristics and interdependencies to the objects. And this, he believed, was what gave rise to thought and cognition. In other words: Without IDs – without the images emitted from the objects – there is no thought. I think Democritus is pretty close to describing the objectives of RFID: Objects and articles disclose information about their origin, their destination, their status – in short, about their very identity, their ID. And people, machines, and

means of transport can use this information to identify what needs to be done with the objects and articles. Today, virtual bits and bytes are making their way to the other side of the World Wide Web along data lines on the Internet. Tomorrow, real-life containers, pallets, and packages on cargo ships and trucks will make their way to the other side of the world on the “Internet of things,” as RFID has also been dubbed. Even today, it is not unusual for tourists to pay visitors’ tax by RFID chip or for chips to be used in ski passes, parking tickets, and tickets for the football world cup. Even the balls will know when they’ve scored a goal. And yet despite the ubiquity of this fascinating technology, the state, the economy, and society still have a long way to go as regards education and understanding. Few people know what radio frequency identification actually means or what this technology can and can’t (yet) do. Consequently, I am most grateful to the project group within our organization. The group has approached this topic from a range of angles, making it easier even for small companies to use RFID to gain a strong foothold in the retail or logistics value chain, for instance. The concrete case studies illustrate how users can benefit from RFID in a tangible way. The Düsseldorf retail company Metro Group claims the chips will help cut costs by up to 20 percent in the future. However, and this is the important aspect for me as a representative of small and midsize companies, RFID also opens up intelligent logistics to all market players. It reduces economic barriers and allows small and midsize companies to enjoy all the benefits of efficient logistics. Sophisticated retail tracking systems will soon cease to be the exclusive luxury of large organizations. Shorter product life cycles alone can hugely reduce a manufacturer’s capital lockup – an added bonus for small and midsize companies in particular. It’s hardly surprising that RFID is causing a stir in the world of transport and stockholding. RFID technology is a great opportunity for Germany. When it comes to usage, development, and research, we are leading the way, maybe even world class. BITKOM has long called upon politicians, the economy, and society to support RFID. What we do not want is for this technology to be stifled by a characteristically German tangle of regulations, leaving us once again (recall the PAL television and mp3) to discover new technologies only to watch them come to fruition abroad.

6 of 50

I am convinced, despite concerns voiced by some regarding data protection, that RFID will improve our working world, our environment and will not become another form of surveillance. Up to now, there has been a gap between the real world of physical objects and people and the virtual world of information. RFID bridges this gap. If the philosopher Democritus claimed 2,400 years ago that without IDs there could be no thought, then I say that without RFID, there can be no successful trade. Sincerely, BITKOM Vice-President Heinz Paul Bonn

7 of 50

2 Introduction The abbreviation RFID (radio frequency identification) has come to signify system solutions for tracking and tracing objects both globally and locally using RFID tags. RFID is one of several technologies collectively known as Auto-ID procedures – procedures for identifying objects automatically. It bridges the gaps to IT systems that were previously bridged by manual data entry. Companies are increasingly deploying Auto-ID procedures in their supply chains. RFID solutions usually go beyond enterprise boundaries and have to be integrated in extensive IT systems, which is why almost every IT provider, be they product or service provider, is taking a position on the issue. RFID is already used in several areas of everyday life, such as in the central locking systems for vehicles or in ski passes. RFID functions are now integrated in German passports and will even find their way into health cards. Nonetheless, it will be years (if at all) before yogurt pots and other small consumer articles are fitted with RFID labels in supermarkets. What has become reality is the use of RFID labels by carmakers for managing containers in production logistics and for identifying pallets and cartons in retail logistics. Yet these industrial fields of application, which have already contributed significantly to the improvement of such processes, are rarely the subject of public debate, which tends to focus primarily on the tagging of consumer articles. Even though the labeling of such articles is not currently a reality (see above), the potential threats posed by such technology nonetheless stir public emotion. It goes without saying that companies using RFID solutions have a duty to respect the public’s right to privacy and as such, its right to determine how private information is used. This white paper provides an overview of the numerous application-related and technical aspects of RFID systems. It shows that even in the highly competitive IT market, it is possible to define basic structures upon which different providers can agree. This benefits companies seeking to deploy RFID procedures and therefore needing to combine components from different providers to establish comprehensive, stable, and inter-enterprise IT infrastructures. We could compare it to the task of an architect, who draughts the plans while the individual parts of the building are completed by different specialists. More than ten authors have contributed to this paper. While they represent quite different companies from the ITC sector, where they face each other both as partners and competitors, they are at one with the content and objective of this white paper. We have deliberately kept the technical details to a level that will enable company directors – especially from small and midsize businesses – or managers in user departments to understand the interdependencies of RFID systems without being IT experts or specialists in electrodynamic antenna fields. The aim is to help them bridge the gap between the sector-specific requirements laid down by their business processes and the IT systems they need to purchase, develop, or integrate to support RFID solutions. Finally, the “Sample Applications” chapter describes RFID solutions with a proven track record in industry. The interested reader will also find references to additional literature and Internet addresses to pursue his or her own research. Auto-ID and RFID technologies are developing at an astonishing rate around the world with new information appearing daily, particularly on the Internet. It is well worth keeping track of new developments. Despite widespread skepticism about the spread of Auto-ID procedures, the overriding tenor is that these procedures look set to meet with similar acclaim to that received by the bar code. And this is why it is particularly important for companies involved in inter-enterprise supply chains to address the issues surrounding this technology.

8 of 50

This white paper is the joint work of the BITKOM RFID Project Group and was compiled in the first half of 2005. Berlin, July 30, 2005

9 of 50

3 Management Architecture for RFID Solutions Running business applications smoothly is a matter of survival for enterprises today. The enterprise itself is not the only one to suffer if its IT systems fail – relationships with customers and business partners can also be put under severe pressure as a result. It is therefore vital for inter-enterprise business processes to be structured reliably. This calls for a comprehensive management infrastructure that encompasses everything from an enterprise’s strategic goals to the individual components of the IT systems. Auto-ID and RFID solutions, in particular, need to be embedded suitably in such an environment. Responsibilities must be assigned clearly and interfaces defined accurately. This increases transparency and reduces costs.

Figure 1: The Layers and Components of an RFID Management Architecture Source: Hewlett-Packard Components of an RFID Architecture Figure 1 shows the management and function components in a comprehensive IT architecture. The left column shows the different layers. The components in the middle and right columns have to be suitably interlinked in order to ensure smooth IT operations. The right block (red) refers to the management disciplines, the lower middle block (blue) to the application, hardware, and software infrastructure, and the blocks above it (green) primarily to application systems (such as ERP) and services. Appropriate security structures must, of course, underpin the overall construct. The RFID tags and readers are located on the blue layer second from the bottom. More details about the management disciplines are given below:

IT Infrastructure Management – optimization of mission-critical components This area is responsible for managing and controlling the IT infrastructure, that is, components such as servers, data stores, networks, PCs, and printers. It improves the availability of the entire system, increases system performance, and supports cost-effective operations.

Application Management – availability of applications Application Management monitors and controls complex, mission-critical application environments. It improves the performance, availability, and quality of standard applications and company-specific applications. Problems are identified, located, and eliminated quickly as a result.

10 of 50

IT Service Management – IT as a service IT Service Management optimizes IT as a service within the enterprise. The impact of a breakdown on business processes is recorded in the form of key figures such as lost sales or increased workforce requirements, and service level agreements (SLAs) are defined. Best practices and the IT Infrastructure Library (ITIL) can enable services to be improved considerably. Such standards also enable benchmarking with comparable enterprise areas or competitors.

Business Management – the link to business processes Business Management links the IT resources to the business objectives and priorities. All business processes, as well as their interdependencies with IT applications or IT infrastructures are monitored using key performance indicators (KPIs), which represent an enterprise’s strategic goals. IT resources can be allocated more appropriately depending on their relevance for the business processes, troubleshooting can be prioritized, and the performance of business processes optimized.

Lifecycle Management – planning ahead for change Enterprises use Lifecycle Management to plan their business processes strategically in advance. This is where future business process requirements are linked to the enterprise’s medium and long-term objectives. Enhancing systems “just in time” using capacity management lets companies prevent excess capacity and, as such, contributes significantly to cost efficiency.

Many commercial software products are available to support the management disciplines. When selecting suitable products, companies must consider the following criteria to ensure that the products are economically viable and offer an attainable return on investment (ROI):

o Superior application availability

o Reduction in unplanned downtimes, minimization of planned downtimes for the installation of additional components, upgrading of hardware and software (updates and upgrades), and so on

o Impact on customer relationships (SLAs)

o Potential increase in employees’ productivity

o Lower costs as a result of adapting business processes more quickly

The implementation of Auto-ID and RFID procedures affects many areas, such as supply chain control, product lifecycle management (PLM), and the production process. At present, there are diverse data islands, data sources, and integration gaps within the processes. Documents are printed and entered again manually in different systems. This results in errors and delays, inefficient usage of asset capacity (return of net asset), and inflexible processes. RFID enables machine-to-machine communication – automatic, event-driven communication. Data is provided and processed in real time and across different media.

11 of 50

4 RFID System Architecture and Standards

4.1 Market Trends and Usage Radio frequency identification (RFID) denotes the technology that enables objects to be identified without contact and without their RFID tags being visible. The list of areas that have already embraced RFID technology is substantial, including ski passes, remote controls for vehicle central locking systems, anti-theft protection in department stores, logistics processes for retail objects such as pallets, containers, clothing, and identification for domestic and slaughter animals. In many market sectors, RFID technology is set to replace bar codes. In supermarkets, however, where consumer articles are concerned, the bar code will continue to dominate for many years. Much has been written in recent years about the way RFID technology will affect the retail industry, based primarily on the experiences of the department store chains Metro Group (Germany), Tesco (United Kingdom), and Wal-Mart (United States). The impression given was often that the mass merchants of these companies were about to equip all their products with RFID tags, fully automate the shelf-stocking process, and have customers check out their goods themselves. At the time of writing (mid 2005), only a marginal number of articles amounting to single-figure percentages are fitted with RFID tags, while widespread RFID-supported self-service checkouts are still very much a thing of the future. In this area, bar codes continue to be the dominant and globally accepted means of product identification. Nonetheless, Metro and Wal-Mart are pioneers when it comes to deploying RFID tags in the supply chain. Their suppliers are increasingly attaching RFID tags to cartons and pallets, mostly with conventional bar code labels on the front so that both procedures can be used complementarily. This level is expected to become ever more widespread in logistics in the coming years. Most RFID applications are currently found in “closed” processes, where the tags are reused time and again and within one organization or company. One example would be in libraries for the loan of books, or in the areas of asset and document management. Such cases are referred to as “closed loops.” The retail industry usually operates “open” systems with labels on the packaging that are discarded at the end together with the RFID tags. Such cases are referred to as “open loops.” With the lower price limit for RFID tags currently between 10 and 30 cents, tagging consumer articles such as yogurt pots – which themselves have a similar price – is yet to become a viable economic option. Pallets and containers with permanent RFID tags are also in use. Their data is updated continually as the pallets are reused. As such, more expensive tags can be used and still be economically sound. This is also a “closed loop” situation, since the pallets flow in a circle. RFID tags can also be used for spare parts. Once installed in equipment, aircraft, or vehicles, the tags can remain on the part and enable it to be identified at any time. Additional data could be stored in such tags and be queried at any time, for instance data about installation, life cycle, or maintenance cycles. Airbus and Boeing are currently planning to use this technology for aircraft parts. Track and trace describes the process of tracking and locating products automatically within supply chains. In this process, RFID technology requires RFID antennas and readers, for instance at warehouse entrances and logistics hubs. The antennas could, of course, be connected to mobile devices. Nokia is the first company to provide a cell phone with an RFID reader, which could prove attractive to consumers and industry alike.

12 of 50

When the movement of containers or other mobile objects is being tracked, a different Auto-ID technology comes into play – the global positioning system (GPS), or the European Galileo system currently in development. This technology has become familiar through its use in vehicle navigation systems. It is used to locate objects around the globe, and is not discussed further here. GSM, GPRS, UMTS, or the INMARSAT satellite communications system is used to transfer localization or RFID data when mobile devices are being used. Permanently installed RFID antennas have either a cable connection or a wireless WLAN connection to “back-end” systems, where the data received is processed further. (Other technologies such as GPS or location-based services can, of course, also be used for such purposes.)

4.2 Architecture of RFID Systems An extensive IT architecture is essential to provide stability for the RFID systems and to ensure they run reliably. In architectures of this type, system and application functions are usually arranged in different layers, between which data is exchanged.

Figure 2: System Architecture for RFID Applications Source: Infineon In Figure 2, the RFID readers that read the RFID tags are located in the lower hardware layer. The data read (“raw data”) is transferred to the “edgeware” layer, which could be implemented on a server in a warehouse, for instance. In this layer, data is filtered and only the data (“events” or “alerts”) that is relevant for the higher layers is transferred to the middleware. Middleware is typically installed in the data center. This kind of architecture simplifies the process of developing, installing, operating, and maintaining RFID systems. Middleware acts as a bridge to the business applications in the back end. An infrastructure responsible for monitoring and updating the components and for communication between the components supports this model. Various components are described below with reference to Figure 2. They are described in more detail in the “RFID System Components” chapter.

ERPEnterpriseResourcePlanning

SCMSupplyChain

Management

MESManufacturing

ExecutionSystem

EnterpriseResourcePlanning

SupplyChain

Management

ManufacturingExecutionSystem

BPM - Business Process Management

Hardware RFID devices: Transponders, readers, printers, sensors …

Middleware

Backend

Processes

EAI - Enterprise Application Integration

Edge ware

RawData

Device management layer

Data management layer

Tag Business layer

Edgeware Edge ware

RawDataRawdata

Configdata

ConfigConfig.

Events &

alerts

Events &

Events &

CommandCommandCommand

RFID softwareDevice management layer


Tag business layer

ERPEnterpriseResourcePlanning

SCMSupplyChain

Management

MESManufacturing

ExecutionSystem

EnterpriseResourcePlanning

SupplyChain

Management

ManufacturingExecutionSystem

BPM - Business Process ManagementBPM - Business Process Management

Hardware RFID devices: Transponders, readers, printers, sensors …Hardware RFID devices: Transponders, readers, printers, sensors …

Middleware

Backend

Processes

EAI - Enterprise Application Integration

Edge ware

RawData



Tag Business layer

Edgeware Edge ware

RawDataRawdata

Configdata

ConfigConfig.

Events &

alerts

Events &

Events &

CommandCommandCommand



Tag business layer

Edge ware

RawData



Tag Business layer

Edgeware Edge ware

RawDataRawdata

Configdata

ConfigConfig.Configdata

ConfigConfig.

Events &

alerts

Events &

Events &

Events &

alerts

Events &

Events &

CommandCommandCommandCommandCommandCommand



Tag business layer

13 of 50

RFID tag or RFID transponder In the simplest case, this is a label composed of a chip and an antenna. Tags made into adhesive labels can be attached to parts. RFID tags in the form of glass cylinders can be implanted under the skin of animals and humans, an application that has already been approved in the United States. RFID tags can also be cast in synthetic resin to enable them to withstand difficult conditions. Simple tags are passive – they cannot emit information unless electromagnetic waves from reading devices supply them with power. Active tags have their own power source and are capable of sending information themselves. In their simplest form, RFID tags have just one tag ID, a globally unique serial number, which they emit if required. Tags can also have writable memory areas, enabling an electronic delivery note, for instance, to be stored on a pallet tag. This is important if shipments need to be checked in locations without access to the Internet or the EPCglobal Network. The way tags are fitted is crucial for their electronic identification. Tags on metal surfaces need special plastic underlays, while tags concealed beneath liquids (containers, bottles) are effective only in conjunction with certain technologies (125 KHz, 13.56 MHz), if at all. RFID Frequency Comments

125 KHz (LF) A globally standardized and approved frequency, primarily for inexpensive, passive RFID tags for identifying animals.

13.56 MHz (HF) A globally standardized and approved frequency, primarily for inexpensive, passive RFID tags for identifying individual objects.

400 MHz Used, for instance, for the remote control of vehicle central locking systems.

868 MHz (UHF) A frequency standardized in Europe for active and passive RFID tags for logistics.

915 MHz (UHF) An analogous frequency used in the United States. The tags usually support the entire frequency channel from 850 to 950 MHz and can thus be used in global logistics processes.

2.45 GHz An industrial, scientific, and medical (ISM) band approved globally which does not require a license or registration. Used for active transponders, for example, with temperature sensors or GPS localization.

Figure 3: The Key RFID Frequency Ranges and Their Applications

RFID readers/writers and antennas Antennas are connected to electronic control devices: the readers. They generate electromagnetic fields, via which data is received from or transmitted to RFID tags. Data is transferred without a line of sight to the tag. Note, however, that unfavorable conditions can cause transmission problems with certain technologies, such as metallic environments or liquids. Tags and readers/writers must have compatible frequencies. Figure 3 illustrates the most important frequencies and their ranges of application. Other important factors besides the frequency are the antenna design and range (the maximum distance between the RFID tag and the antenna) and the transmission power. The necessary antenna geometry always has to be determined and verified on site, that

14 of 50

is, at the point where the data is to be read. For more details, see the “RFID System Components” chapter.

Middleware and Edgeware One characteristic of the tracking and tracing with Auto-ID procedure is that data accumulates on the basis of events, which means that the data has to be accepted at the time it is identified and transmitted by an antenna. In addition, much more data accumulates than is required by the downstream business systems in the back end. As a result, such IT systems require one or more special functional layers. In Figure 2, these layers are the middleware and the edgeware. The edgeware cleanses the data of any read errors and multiple readings not already cleansed by the readers and buffers it in a database. Next, the data is filtered in accordance with the business process requirements and transferred to the ERP systems. Since the ERP systems may have to trigger other actions in response, the data has to be transmitted in real time. Buffering the data means that it is available for additional analysis or review purposes. In the case of permanent antennas/readers, the event data is transferred from the antennas/readers to the edgeware or middleware via Ethernet or WLAN technology, and in the case of mobile devices, via GSM, GPRS, or INMARSAT. In addition to its RFID-related tasks, the middleware is responsible for integrating distributed application systems across the company. That’s why it is also referred to as enterprise application integration (EAI). Business process management can take place within the ERP modules or alternatively in the middleware, if it has to function across different modules.

ERP and business IT systems (back end) This is usually the highest layer in a system architecture. It comprises IT systems from standard software makers such as SAP, Oracle, Retek, and Microsoft (Navision) or systems that have been programmed individually. They use the data they receive to support the business processes. The umbrella term for such systems is enterprise resource planning (ERP). Application-specific aspects can be covered by complementary systems such as a manufacturing execution system (MES) or supply chain management (SCM).

4.3 The Internet of Things Envisaged by EPCglobal The Electronic Product Code (EPC) on RFID tags will increasingly take on the role of the familiar EAN code on bar code labels. EPC is a numbering scheme that can identify all objects uniquely around the world. It is issued by EPCglobal Inc., New Jersey (United States) and Brussels (Belgium). EPCglobal Inc. is an international organization that resulted from the merger of EAN International and the Uniform Code Council, Inc. (UCC): www.EPCglobalinc.org. EPC numbers have a fixed structure, as illustrated by the table in Figure 4. EPCglobal runs the EPCglobal™ Network. It provides standards for deploying RFID technology in the global value chain, especially in the retail industry. The EPCglobal scenario works on the assumption that every individual object will have its own home page in the future, which will be accessible via the EPCglobal Network. It will be located somewhere on the Internet and typically be run by the object’s producer. The information provided on the home page for a given product would include:

o Data about design, production, shipping, sales, maintenance dates, and expiry dates

o Certificate of authenticity, instructions for use

o Delivery of the object on particular pallets and its position within the supply chain

15 of 50

Gillette Venus Cartridge: 00.47400.114008.9780. (Structure in accordance with EPCTM Tag Data Standards Version 1.1 Rev.1.24)

Field Name Length (Bytes) Comments

Data header 2 For control purposes within the network, version names, formatting specifications

Manufacturer (EPC manager)

7-9 In Germany, the number is allocated by GS1.

Part number (object class)

6 The producer and part numbers correspond to the EAN from the EAN.UCC system; each EPC can accommodate one million articles.

Serial number 9 This number is used to identify individual parts uniquely; capacity per manufacturer: 100 billion articles

Figure 4: Example and Structure of an EPC Number In design engineering, this information would also support product lifecycle management (PLM). Together, the home pages constitute the much-cited “Internet of things,” a concept which is now found in literature as well, the “things” being articles, cartons, pallets, spare parts, and other objects. The scientific approach for the EPCglobal Network was developed by the Auto-ID Center, a project at the Massachusetts Institute of Technology (MIT) in Boston (United States). It gave rise to the network of Auto-ID Labs, a federation of university institutes around the globe, including St. Gallen (Switzerland) and Cambridge (United Kingdom). After conducting basic research into RFID, MIT handed over responsibility for the EPCglobal project and its promotion to EPCglobal Inc. The EPC representative in Germany is GS1 Germany GmbH (previously CCG) in Cologne: www.gs1-germany.de The EPCglobal Network is founded on scientific principles with a global character and, as such, is dedicated extensive coverage in this white paper. Needless to say, RFID tags can also be used completely independently of EPCglobal, particularly in closed loops.

World Wide Web EPCglobal Network

DNS A central system that controls the querying of Web

sites and e-mails.

ONS A central directory of manufacturers

registered with EPC, controls queries for product information.

Web sites The location (resource) at which information about

a particular subject is stored.

EPC Information Services The location (resource) at which information

about a product is stored.

Search engines Tools for finding Web sites.

EPC Discovery Services A tool for finding EPC information.

SSL The security standard for Web sites.

EPC Security Services A tool ensuring secure access on the basis of

authorization. Figure 5: Structure of the EPCglobal Network Compared with the Web

16 of 50

The EPC data standard is a cross-industry standard. User groups propose industry-specific requirements. Expert groups have been established already to cover “Fast Moving Consumer Goods,” “Healthcare/Pharmaceuticals,” and “Transport/Logistics,” with other groups planned as required. The interests of small and midsize enterprises and of countries/regions not currently represented will be covered in this standardization process by the national GS1 member organizations. They ensure a general balance of interests and the feasibility of the results as open, global standards. Application Area Question to the EPCglobal Network

Production We need to recall 1,000 parts due to production errors. Where are these parts?

Marketing Our product promotion is going extremely well in the North East region. In which regions do we have products in stock that we can move to the North East?

Retail We need a subsequent delivery of product xy, fast. Where can we get this?

Distribution The truck has arrived at customer xy but one carton is missing. Where is the carton?

Maintenance When was this part installed? When does it have to be replaced?

Figure 6: Questions Solvable by the EPCglobal Network in the Future The EPCglobal Network connects local servers that contain the object home pages. It is accessed via the Internet. Various service components in the network control the servers as well as authorization and access to the information. The Object Name Service (ONS) will be a key element of the EPCglobal Network. It works on the same principle as the familiar Domain Name Service (DNS), which converts URL addresses (www) to machine-readable IP addresses for locating objects on the Internet.

The ONS server This server will contain all EPC numbers and their associated IP addresses, at which the home pages for individual objects can be accessed on the Internet. Participants authorized by EPCglobal can use these services. Figure 6 lists business questions that the EPCglobal Network will be able to answer in the future.

EPCIS: EPC Information Service

Individual companies use the EPCIS software component to communicate with the EPCglobal Network and the ONS server. EPCIS is interface software, or a gateway, defined by EPCglobal. Companies can use it to send EPC numbers to the ONS server and thus determine the IP addresses of the object home pages. EPCIS also supports data exchange between companies. In this case, the EPC number is used to find the relevant communication partner. Figure 7 shows how data would typically be queried from an object home page using EPCIS. In this example, Gillette informs Wal-Mart that a product in the supply chain was separated from a complete delivery and suggests appropriate action.

17 of 50

Figure 7: Gillette’s EPCIS Service Source: GS1 Germany/EPCglobal

RFID standardization activities at EPCglobal A standardized infrastructure (such as that illustrated in Figure 8) is essential for RFID to be used in open systems across company boundaries. The EPC concept describes a standardized procedure for extracting/recording, storing, and exchanging EPC information in open systems. At present, the following standards for an EPC-capable RFID system are being developed and defined under the umbrella of EPCglobal:

o Tag data standards (the data standard) This data standard regulates how the different identification standards (such as SGTIN, NVE) are coded as EPC tag information.

o Air interface protocol (GEN 2 AIP) The air interface protocol, also known as “Generation 2,” regulates communication between the reader and the tag.

o Reader protocol Describes data exchange and the command structure between EPC-capable middleware and the reader.

18 of 50

o Reader management specification Standard functions for the individual configuration of EPC-capable readers and the control of a multiple reader environment.

o Tag data translation The conversion of EPC information from the tag to a format that is compatible with the Internet.

o Filter and collection ALE (F&C ALE) The way in which EPCs in a reader environment are read and counted in accordance with various criteria.

Figure 8: EPCglobal Network Standards Source: GS1 Germany/EPCglobal

o ONS application layer interface Provides information about the location in the EPCglobal Network of information about a particular EPC number.

o EPCIS protocols Describe how the EPC information can be stored and accessed via the EPCglobal Network.

o Security specification The requirements regarding secure information exchange between the participants in the EPCglobal Network.

The EPCglobal standards are technical specifications and functional descriptions. All IT service providers can integrate the standards in their solutions (hardware and software) and make them available to their users. Most of the specifications are being compiled at present. The air interface protocol (GEN 2 AIP) and the data standard are already complete. Initial drafts of the other specifications are expected at the end of 2005.

ONS

Internet

ALE F&CALE F&C

Application ProgramInterface (API)

Security Specifications

AuthentificationAuthorization

Tag Data Standards

EnterpriseEPC IS

Internal System(ERP, WMS, etc.)

EPC IS

Private network

Enterprise

Private network

EPC ISEPC IS

MiddlewareEPC enabled

ReaderEPC enabled

Tag Data TranslationReader ManagementReader Protocols

GEN 2 Air Interface Protocol

MiddlewareEPC enabled

EPCIS protocols Reader

EPC enabled

ONS

Internet

ALE F&CALE F&CALE F&CALE F&C

Application ProgramInterface (API)

Security Specifications

AuthentificationAuthorization

Tag Data Standards

EnterpriseEPC IS

Internal System(ERP, WMS, etc.)

EPC IS

Private network

Enterprise

Private network

EPC ISEPC IS

MiddlewareEPC enabledMiddlewareEPC enabled

ReaderEPC enabled



GEN 2 Air Interface Protocol

MiddlewareEPC enabledMiddlewareEPC enabled

EPCIS protocols Reader

EPC enabled

19 of 50

4.4 RFID Standards Technology standards describe the technical basis of RFID systems. They define frequencies, transmission speeds, codes, (anti-collision) protocols, and other factors. The International Organization for Standardization (ISO) issues authoritative standards, as do EPCglobal and the Auto-ID Labs. Of course, the ideal would be for the EPC standards to be ratified by the ISO as well, which is why EPCglobal has submitted the new Generation 2 (Gen 2) standard to the ISO. Standardization is expected by the beginning of 2006. A selection of key standards is listed below: Standard Subject of the Standard Frequencies

Auto-ID Class 0 Parameters for air interface communication 860 - 930 MHz

Auto-ID Class 1 Parameters for air interface communication 860 - 930 MHz

EPCglobal Gen 2 Parameters for air interface communication, intended as replacement for Class 0 and Class 1, submitted to the ISO at the beginning of 2005

860 - 930 MHz

ISO 14443 Regulation for contacless / proximity ID cards, reading distance 7 – 15 cm.

13.56 MHz

ISO 15693 Regulation of contactless / vicinity cards, reading distance up to 1 m.

13.56 MHz

ISO 18000 Family of RFID air interface standards, examples:

ISO 18000-1 Generic parameters for air interface of globally accepted frequencies.

ISO 18000-2 125, 134.2 KHz

ISO 18000-3 Reading distance max. 1.5 m, successor of ISO 15693

13.56 MHz

ISO 18000-4 2.45 GHz

ISO 18000-5 Has been withdrawn. 5,8 GHz

ISO 18000-6 EPCglobal Generation 2 Tags (under development) 860 – 960 MHz

Figure 9: ISO and EPCglobal Standards The identifier in ISO 15693 is not the same as the EPC, which identifies the object bearing the RFID transponder. The identifiers are allocated to the different manufacturers in a way that prevents overlapping and permits unique identification. The associated product data (such as EPC) can be stored on the transponders or in external databases. The 18000 standards describe measurement procedures for checking RFID structures. This concerns field strengths, modulation grades, ranges, and processing times.

4.5 Outlook With EPCglobal and the envisaged “Internet of things,” RFID technology it set to permeate many areas of industry, logistics, and the consumer environment. The challenge now is to find commercially viable solutions capable of being integrated in a global network for tracking and tracing, while at the same time protecting consumers as well as patents and intellectual property.

20 of 50

5 RFID System Components

5.1 RFID Tags: Chip and Antenna

5.1.1 Structure of an RFID Tag/Transponder RFID tags, also known as RFID transponders, comprise a memory chip and an antenna. They can differ in many respects. One fundamental criterion is the frequency, which currently ranges between 125 kHz in the LF range and 5.8 GHz in the UHF range, and which is a determining factor in the antenna design. The following analysis relates primarily to transponders with frequencies from 13.56 MHz (HF) or between 860 and 960 MHz (UHF). Such transponders are mainly deployed in retail and logistics and can be manufactured as high-volume one-way transponders. A tag’s performance parameters are its read range, transmission speed (data rate), bulk-read capability, and the impact caused by surrounding objects. The frequency, the orientation to the reader field, and the design and size of the antenna determine an RFID tag’s read range, its resilience to environmental factors, and its bulk capability. The frequency and the associated transmission protocol (anti-collision algorithm) determine the basic rate of data transmission or the bulk read-speed. A distinction is made between active, semi-active, and passive transponders. Active transponders can use a battery to generate radio signals. Semi-active transponders can be stimulated by the read field to amplify the field influence or to buffer sensor data (for tracking cold chains, for instance). Passive transponders only modulate the field emitted from the reader to transfer data. Figures 10 and 11 give examples of RFID inlays for passive RFID tags. The coiled antennas indicate a frequency range of 13.56 MHz (HF). Figure 11 shows inlays that operate between 860 and 960 MHz (UHF). In this case, the antennas have a dipolar form. The tripole illustrated is a variant of a dipole. The bizarre shapes of the dipole antennas help to establish optimal reception conditions in particular working environments. The inlays are made into labels or ID cards, for instance.

Figure 10: Inlays for RFID Tags with 13.56 MHz (HF) Source: Infineon Another distinguishing characteristic is the write/read capability of the transponders. Three types exist:

o Read-only tags These tags are produced with an n-bit unique serial number for identification.

21 of 50

Central databases are used to link the information to the objects to which they are attached.

o WORM tags (write once – read many) These tags permit a one-time, unchangeable coding of the information on the tag, which can be read virtually any number of times (> 100,000 times).

o Read/write tags These tags have individual, writeable memory areas. The information can be written and changed. Reference data, handling instructions, or process data can be stored and encoded. Writeable tags can transport interface data between applications and help to prevent integration gaps.

A tag’s core is the chip that stores the information. It essentially comprises three components:

o A high-frequency part for processing signals and for extracting the necessary energy from the electrical field.

o A control unit for processing the commands it receives.

o A memory unit.

Figure 11: Inlays for RFID Tags with 868 MHz (UHF) Source: Avery Dennison and Rafsec The most important parameter for the use of chips is the size of the available memory, also referred to as the data width. A distinction is made between the user memory (such as 96 bit for EPC-compatible chips) and other memory cells for status control (such as locks on individual memory cells, kill command for permanent deactivation) and data control (storage of check totals). The largest chip currently available in the UHF range offers 2,048 bits, of which 1,728 bits can be used freely. The simplest tags have a data width of one bit and do not contain a chip. Such one-bit tags are used in the retail industry to protect against theft (electronic article surveillance – EAS). They only contain yes/no information that can be changed (paid for/not paid for). If they are still attached to an article when a customer leaves the department store and passes the RFID antennas located at the exit, a warning alarm sounds, signaling that the article has not been

22 of 50

paid for or that the tag was not removed at the cash desk. These tags do not contain information about the actual article.

5.1.2 Tag Antenna The RFID tag antenna is the interface between the electromagnetic wave generated by the RFID reader and the chip. The smallest antennas possible are used for tags. In the low-frequency (HF) range with short read distances, the tag is in the near field of the reader antenna. The power and signals are transferred by means of a magnetic coupling using inductors. In the HF range, the tag antenna therefore comprises a coil (inductive loops), to which the chip is attached. In the UHF range, in cases where the read distances are larger, the tag is located in the far field of the reader antenna. The reader and tag are coupled by the electromagnetic wave in free space, to which the reader and tag are tuned by means of appropriate antenna structures. The tag antenna should be as small as possible and easy to produce. The associated requirements are described in more detail below. Printed antennas are really very easy to produce. The antenna is attached as a flat structure to a substrate. The next stage in the production process often involves attaching the chip to the substrate and connecting it to the antenna. This assembly is called an inlay. An inlay becomes a tag or transponder when it is fixed to an adhesive label or a smart card. It must be noted, however, that the electromagnetic properties of the materials surrounding the inlay affect the tag’s ability to communicate. In extreme cases, tags cannot be read if unsuitable reader antennas are selected. Another type of usage involves integration into the object that is to be identified. Parts of the object can be shaped to form an antenna and the antenna can be adjusted optimally to suit the object. This significantly increases readability whilst simultaneously protecting against counterfeiting. Dipolar antennas are preferred in the UHF range. However, very small, straight dipolar antennas have unfavorable connection impedances and cannot be adapted to the chip without considerable losses. Consequently, dipoles are often folded to form meanders or fractal structures, which offer good compromises in terms of electrical properties. Similar considerations apply to simple loop antennas. Dipole and loop antennas are ideal for production in printed form, enabling flat tags to be produced economically. However, the disadvantage is that the antenna parameters are changed radically by the material in the immediate vicinity, especially by the object that is to be identified. As such, a dipole on a metal surface is ineffective, while its properties are different on glass and on paper. The structure and thickness of the material behind the dipole affect the properties of the tag antenna. A large metal surface half a wavelength behind a dipole can reflect the reader signal such that the wave is canceled in the tag’s location and the reader can no longer read the tag. Insulating materials can also cause similar effects. It follows, therefore, that the design and selection of tag antennas are critical for the success of the desired Auto-ID functions.

5.2 Air Interface and Bulk Reading The air interface describes the connection between the reader and the tag. The connection comprises various layers. Physically, the reader and tag are linked by an electromagnetic

23 of 50

wave. To enable them to communicate and information to be transferred, the wave has to be modulated. The type of modulation and the form of the modulated signal are just as much a part of the air interface specification as the structure of the commands that are transferred, although the commands themselves are not part of the air interface.

Figure 12: Reflections in the Reader Antenna Range Source: Kathrein If tags are to be readable in open RFID systems worldwide, global standards are required not only for the air interface but also for the frequency ranges in which the RFID systems operate. The regulatory authorities have to harmonize such standards and approve them for their intended use. In the HF range, the frequency 13.56 MHz is available around the world. In the UHF range, Europe uses a frequency of 868 MHz while the United States use 915 MHz. The respective mobile telecommunications frequencies stand in the way of harmonization. Readers are transmitters and, as such, must operate only in the allocated frequency bands. This means that UHF tags for worldwide use must be capable of working in the entire frequency range from 860 to 960 MHz. This, in turn, puts exacting demands on tag antennas and increases the price of the tags. In the HF range, the near-field procedure is typically applied for communication between the reader and tag when the read distance is short. The electromagnetic connection therefore comprises a relatively simple magnetic coupling. Objects in the close vicinity have little impact on the coupling. The force of the coupling is determined primarily by the distance between the reader antenna and the tag antenna and by the location and orientation of the two antennas or coils to each other. The large read ranges called for in the UHF range result in far more complex conditions. Foreign objects behind the tag object, or stronger still, in the measuring field, affect the entire reader-tag system. The effects range from simple attenuation to reflection and refraction through to scattering of the electromagnetic wave. The multitude of potential negative effects means that objects must not come between the reader and the tag during read operations. Exceptions include thin, insulating layers and materials with low attenuating and inductive capacities, such as paper, loose materials, and expanded polystyrene. Tags can also be read if they are situated beneath a dry cover. If the cover is wet, however, it is highly probable that the tags will no longer be readable.

24 of 50

Another critical factor is the orientation of the reader and tag antennas to one another during the read operation. The optimum configuration and alignment of the reader antenna, at a warehouse door, for instance, has to be determined for each individual situation.

Left: The reader identi fies a collision by reading several tags at once.

Right: The reader isolates one tag ID and interrogates only that tag. The other tags are read sequentially.

ID = 01001

ID = 10111

ID = 11001

READ (10)

ID = 01001

ID = 10111

ID = 11001Collision!

READ (1)

Left: The reader identi fies a collision by reading several tags at once.

Right: The reader isolates one tag ID and interrogates only that tag. The other tags are read sequentially.

ID = 01001

ID = 10111

ID = 11001

READ (10)

ID = 01001

ID = 10111

ID = 11001

READ (10)

ID = 01001

ID = 10111


READ (1)

ID = 01001

ID = 10111


READ (1)

Figure 13: The RFID Reader Anti-Collision Protocol Source: Siemens The reader and tag communicate using the antennas and a defined command set. The communication protocol is defined such that the reader first sends a command and the chip responds. The command set includes commands to read and write data, to control the anti-collision protocol, to lock individual memory cells, and to deactivate the chip (kill command). If many tags are located in the measuring field at the same time and are read almost simultaneously, the process is known as “bulk reading.” This is where the anti-collision protocol comes into play. It ensures that during the read operation, the reader specifies part of the tag ID in the read command. If more than one tag responds, the reader keeps adding characters to the ID until only one tag responds. That tag is then switched off. The protocol enables up to 100 tags to be processed per second. The readability of the tags is heavily dependent on the materials, the structure, and the places where the tags are located. For instance, it is impossible to read a UHF tag if it is surrounded by metallic objects, liquids, or other highly conductive or reflective materials. A tag must always, therefore, be affixed to the outside of a batch, on a pallet for example. It may be covered at most by paper, expanded polystyrene, or similar materials.

5.3 Reader and Antenna

5.3.1 Reader Structure Tag information is written and read by an RFID reader-antenna unit, comprising the antennas and the actual write-read device. A device can often be connected to several antennas to identify tags better. Readers differ functionally and technically according to their purpose and ambient conditions:

Gate readers Installed at loading gates or thoroughfares for trucks or forklifts. Critical here is reliable identification at long range or when tags are positioned differently. Gate readers generally operate with multiple antennas.

Compact readers

Combine antenna and write/read device in one compact housing. This type of reader is a cheaper alternative to gate readers if the tags are not so far away and if conditions are unproblematic.

25 of 50

Vehicle-mounted readers

Installed in vehicles such as forklifts.

Mobile readers Handheld devices for mobile use. Such devices either transfer the tag data immediately via radio communication (such as WLAN) or collect the data in a built-in memory. Data stored in the memory is transferred to a PC, for instance, when the device is placed in a docking station. Other requirements, such as the degree of protection necessary for a device, are dictated by the device’s intended purpose. For example, readers and antennas at loading gates must be highly tolerant as regards temperature and must be protected against dust and damp. For integration in the higher (software) layers of the RFID architecture, for configuration, and for diagnostics, the RFID readers must support suitable interfaces, for instance for:

o WLAN and Internet (via Ethernet and TCP/IP)

o Point-to-point connections: RS422 for connection to programmable logic controllers (PLCs) or RS232

o Mobile communication: GSM, GPRS, or UMTS

Antenne(n)Middleware

Speicherpro-grammierbareSteuerung (SPS)

PC-Anbindung(für Konfiguration und Diagnose)

Pow er Supply

Ethernet(TCP/IP)

RS 422

RS 232

ProzesssteuerungDigitale Ein-/Ausgänge (24 V DC)

Schreib-Lesegerät

Figure 14: Typical Connection Scheme for a Reader with RFID Antennas Source: Siemens

Certain readers also provide connection options to enable simple process control mechanisms to be implemented, such as digital inputs and outputs with 24 V, which can be used to control traffic lights or gates that are released once the tag data has been checked at the goods issue/receipt point. Simple PLC couplings can also be realized using this technology. The higher protocol layers have not been standardized yet, resulting in additional time and effort when it comes to integrating readers across different manufacturers. In this regard, a joint initiative undertaken by RFID providers would be most welcome.

26 of 50

Figure 15: Bulk Reading and Multi-Tag Handling Source: Siemens

Functional requirements First, a reader must provide sufficiently high read rates and reliability. This is dependent on parameters such as the number and orientation of antennas, the type of tags used, the volume of data stored, and the positioning of the tags. In good conditions, up to 100 tags can be identified per second. If multiple tags enter the field simultaneously, the bulk-read facility is brought into play. It puts the data stream into sequence to enable the tag IDs to be identified without colliding. When large volumes of data have to be read, and particularly for write access, the multi-tag handling concept is deployed to enable individual tags to be addressed directly.

Figure 16: Overlapping RFID-Reader Fields Source: Siemens

The use of multiple readers can result in overlapping ranges and reflections, which in turn can cause mutual disturbances. Figure 16 illustrates a simple example in which the fields of two adjacent readers overlap. In practice, however, reflections on parts of

R e ad e r A R e ad e r B R e ad e r C

S e n de fe ld Ü b e rlag e ru ngAntenna field Overlapping

Reader A Reader B Reader C

Edgeware Server

R e ad e r A R e ad e r B R e ad e r C

S e n de fe ld Ü b e rlag e ru ngAntenna field Overlapping

Reader A Reader B Reader C

Edgeware Server

All tags are read in sequence(bulk reading)

Data is written to (and read from) a single tag (multi-tag handling)

All tags are read in sequence(bulk reading)

Data is written to (and read from) a single tag (multi-tag handling)

27 of 50

buildings can give rise to far more complex scenarios. Such disturbances also result in identification errors. Readers thus need to be synchronized, a task which falls to the Device Management discipline in the higher-level software layer.

With certain systems, data can be preprocessed in the readers. Tag data can be filtered in line with prescribed templates, perhaps to read only the tags attached to pallets and not those on individual products, for instance. Filtering over a particular time frame would be another option, ensuring that tag information is transmitted only once within a given period of time. “Holes” in the field that would otherwise result in multiple identification can thus be compensated for. Both functions serve to reduce the volume of data in the network. Finally, certain readers permit various parameters, such as the transmission force (to limit the range), to be configured with the support of software. Such parameters are stored in the reader.

5.3.2 Reader Antenna The reader antenna establishes a connection between the reader electronics and the electromagnetic wave in the space. In the HF range, the reader antenna is a coil (like the tag antenna). It is designed to produce as strong a coupling as possible with the tag antenna. In the UHF range, reader antennas (like tag antennas) come in a variety of designs. Highly directional, high-gain antennas are used for large read distances. Regulatory authorities usually limit the maximum power emitted in a given direction (the transmission power plus the antenna gain). As a result, the transmission power emitted from the reader to the antenna must also be regulated accordingly. One advantage of highly directional antennas is that the reader power often has to be emitted only to the spaces in which the tags that are to read are located.

Figure 17: Reader Antenna with Antenna Array and Field Regions Source: Kathrein Generally speaking, physical interdependencies mean that the antenna gain is linked to the antenna size. The higher the gain (or the smaller the solid angle into which the antenna emits), the larger the mechanical design of the antenna will be. It follows, therefore, that highly

8 array antennas

Near fieldarea

Far field region

Field regions

Antenna

8 array antennas

Near fieldarea

Far field region

Field regions

Antenna

28 of 50

directional antennas are not used for handheld readers. Antennas typically used for handheld readers include patch antennas, half-wave dipoles, and helix antennas. Larger antenna structures can be used for stationary readers. In the UHF range, they usually take the form of arrays (see Figure 17). In this case, multiple, small emitting elements are interconnected so that the wave fractions are totaled in the correct direction. The size determines the far field of such antennas. The far field does not take effect until the distance to the antenna is so large that the differences in the routes from the individual emitters are negligible in comparison to the wavelength. Close to the antenna, the field is no longer homogeneous and the situation could arise in which a tag is not read because it is in a range of very low field strength. Highly directional antennas are not appropriate for all scenarios, particularly not in the case of warehouse doors equipped with reader antennas. A better option would be to use specially adapted antennas that generate as homogeneous a measuring field as possible.

29 of 50

6 IT Infrastructure

6.1 Middleware and Service-Oriented Architecture (SOA) Service-oriented architecture (SOA) has become ever more significant in recent years. Its objective is to provide users with IT systems and services that are flexible and suited to the task in hand while being economical to maintain. Such systems and services will be integrated to support business processes optimally and enterprise-wide, and enable the highest possible level of interoperability for the IT systems involved. An SOA also has to process the events (real-time information) transferred to the system as a result of RFID read operations and use the Internet as a global communication network. A familiar tool for realizing an SOA is the Internet standard extensible Markup Language (XML), used for structuring information universally regardless of the media in which it is presented. In an SOA environment, the operational application systems communicate via service interfaces rather than bidirectional interfaces. This means that changes can be made flexibly to the individual software components (for instance, software can be updated or local services shifted to a central data center) without affecting other systems and without having to undergo extensive integration tests each time. While conventional middleware does work without an SOA, doing so would make the above-mentioned flexibility unattainable, resulting in less flexible systems with higher maintenance requirements. RFID middleware connects the RFID reader layer to the business applications. Its task is to process RFID events and present them in such a way to the business applications that they can be processed further by those applications. If we consider an application system for warehouse management that is not currently RFID capable, this would mean, for instance, that individual items would have to be aggregated (many articles in one carton) to form groups (a number of uniform cartons) and vice versa before the events are forwarded to the application. The middleware also monitors the RFID hardware (readers, antennas, and so on). This involves sending confirmations to the feedback providers and monitoring the sensors (light barriers, timers, motion detectors), since this cannot be combined initially with the causal process logic. The middleware synchronizes these events with the RFID read results and ensures that duplicate readings are either tolerated or eliminated in accordance with the relevant specifications. The number of messages generated in the LAN or WAN must be kept to a minimum. Not all RFID read results have to be transferred to the supply chain management system (SCM). Software can thus be configured and adapted to different applications more simply as a result of an SOA. Software development tools can be used to organize the individual services in line with process requirements. Web services, XML, message queuing, and workflows are elements of an SOA and of RFID middleware that can be used to link different services.

6.2 The Layer Model for the Real-Time Enterprise The previous section commented on SOA. A comprehensive software architecture to include RFID solutions could comprise five layers (Layers 0 to 4), as illustrated in Figure 18. Depending on each system architect’s experience, a range of different terms will be used in this context, which together form a common overall picture. Take, for instance, the term “real-time enterprise,” which is located on the highest layer (Layer 4: Application Layer) in the figure mentioned above. The number of layers in the system architectures also differs. Nonetheless, a comparison of the different function components in the layers reveals a high degree of

30 of 50

congruence. Such layer models help us to organize the architecture of information systems that are composed of components from different vendors. The following function components have to be assigned to the layers:

Figure 18: Layers in a Software Architecture for the Real-Time Enterprise Source: Microsoft

o Layer 0: Devices Hardware: terminals such as RFID readers, printers, handheld terminals, and so on

o Layer 1: Data Collection and Management The entire hardware infrastructure, operating systems, networks, and other components support RFID. Filter mechanisms that identify RFID tags that have been read more than once and filter superfluous data.

o Layer 2: Event Management Together with the one above it, this layer enables the business processes and solutions to use the RFID information that was generated in real time. This layer is where other services, business partners, and the EPCglobal Network are integrated.

o Layer 3: Services This layer can take the form of an application programming interface (API) or of an XML Web service based on open standards, such as the look-up service for extracting product information from product catalog databases situated in Layer 2. Other services residing in this layer include business intelligence, analytics and reports, and event notifications (alerts).

o Layer 4: Applications Specific business applications and ERP systems with the strategic goal of supporting the real-time enterprise

XML Web services constitute an important middleware standard for the interoperability of different systems. They are assigned to Layer 2 and are based on the XML standard. The Web Services Description Language (WSDL) for describing such services and the Simple Object Access Protocol (SOAP) for communication are available on almost all system platforms. The XML Web service standard is still being defined. Additional functional areas need to be standardized to ensure cross-manufacturer compatibility: Quality-of-service attributes for RFID integration, scalability, reliable message delivery, load balancing, error tolerance, security, and so on. However, this is an issue that is always accompanied by new, emerging areas of technology. It should not deter users from implementing innovative applications for RFID procedures now.

31 of 50

Figure 19: Implementation of Global Data Synchronization (GDS) Source: Microsoft Integration with the EPCglobal Network and the associated process of global data synchronization (GDS) take place in Layer 2. The communication required for this takes place via XML and Web services.

6.3 Event Architecture Auto-ID/RFID technology enables machine-to-machine communication – automatic, event-driven communication, where data is provided and processed in real time. Auto-ID/RFID has a large impact on the way processes are designed. Processes have to be optimized in order to be sufficiently transparent, flexible, productive, agile, and efficient. An event architecture enables data to be recorded on the fly and summarized where appropriate, and information to be presented and forwarded on the basis of events to the relevant applications for processing. This procedure needs to be fast, easy to integrate, and transparent in terms of costs. When machine-to-machine communication is used and item information is recorded automatically, events and status messages created during the process are forwarded automatically or trigger other actions. This calls for a suitable infrastructure and architecture. The architecture described below applies to an outsourcing model, where a service data center operates the IT systems. Of course, a company could operate the same components itself on its own premises. There are always two options for event control:

o The middleware controls the events and thus reduces the time and effort required for integration and implementation (as described below). This is the model used in our example.

o Existing ERP applications (legacy applications) control the events. Considerable time and effort is required to change processes within existing applications and to adapt them to the optimized processes. Functions from Layer 2, the managed Auto-ID services, have to be integrated in systems from Layer 3. We do not recommend this procedure or describe it further here.

Any decision should always be based on a cost-benefit analysis.

32 of 50

SCM ERP CRM PLM

Enterprise Legacy Applications

Managed Auto-ID/RFID Real-time Event Architecture

Business RulesTrack & Tracing

Real-time DataSynchronizing, Provision,Communication & Network

Controlling &Management

Security & Privacy

WMS

Enterprise Application Integration (EAI)

1

3

2

Auto-ID / RFID-Technology & Hardware

SensorsRFID Readers

Other Auto-ID-Technologies

POSBarcode Scanners0

Web Services & EDI

DMS Interface,XML based

EPCInformation

ServiceTag- / EPC-

AdministrationReaderAdmin.

Event-Management

SCM Hosting

SCM ERP CRM PLM

Enterprise Legacy Applications

Managed Auto-ID/RFID Real-time Event Architecture

Business RulesTrack & Tracing

Real-time DataSynchronizing, Provision,Communication & Network

Controlling &Management

Security & Privacy

WMS

Enterprise Application Integration (EAI)

1

3

2


SensorsRFID Readers




SensorsRFID Readers


POSBarcode Scanners

SensorsRFID Readers



Web Services & EDI


EPCInformation

ServiceTag- / EPC-


Event-Management

SCM Hosting


EPCInformation

ServiceTag- / EPC-


Event-Management

SCM Hosting

Figure 20: Auto-ID Event Architecture Source: T-Systems The layers and elements in the event architecture illustrated are described below:

Layer 0 Data-entry devices for Auto-ID information.

Layers 1 and 2

Managed Auto-ID/RFID real-time event architecture

o DMS Interface Information often exists in the form of documents rather than as pure data. An XML interface can be used to connect a document management system (DMS) or archiving system, enabling such documents to be stored in accordance with set criteria.

o TAG/EPC Administration Administration of the Auto-ID components and EPC directory services deployed; where appropriate, connection to local EPC directories, services, or databases (for instance, for food safety/tracking goods).

o EPC Information Service (EPCIS) Storage of product data (such as date of manufacture or expiry date) in a database.

o Reader Administration Administration of the Auto-ID readers deployed (RFID, bar code, biometrics, and so on) and their interfaces.

o Event Management Scripting engine for creating, managing, and executing automatic events (input/output).

33 of 50

o SCM Hosting Hosting and outsourcing for supply chain management environments for local sites or small and midsize companies.

o Business Rules/Track & Trace Mapping of the business rules (definition of the actions for an event) and reporting functions for tracking and tracing objects with RFID tags.

o Controlling and Management Provision of cost accounting functions for analyzing the mapped processes (key performance indicators (KPIs)) and comprehensive management functions for the architecture and platform.

o Real-Time Data Synchronization, Provision, Communication & Network Provision and synchronization of information on the basis of the business rules relating to the events; communication interface to applications and networks, such as: fixed network, mobile telecommunications network, Internet/IP.

o Access Security & Privacy Definition of access rights and dependencies for users and applications (within M2M communication); delimitation of individual clients and their data and information.

Layer 3: Enterprise Legacy Applications

Business applications (ERP): These applications communicate directly with the event architecture. Since the event information and processes are stored and processed in the middleware, outlay for system integration is kept to a minimum. Nonetheless, business processes do still need to be optimized in this layer.

Figure 21 illustrates a complete production process as found in the automotive sector. It spans the entire process from the production of parts to the assembly of vehicles through to sales. This process involves several enterprise areas and companies (illustrated by the different colors). Every link in this process chain must function optimally if the process is to meet its goals. To enable the individual process steps to be monitored in real time (event driven) in the future, the relevant data will be collected automatically by RFID or other Auto-ID procedures and fed to the system to check the processes (keywords illustrated with dashed lines). This scenario puts an end to the delays associated with manual procedures regarding the availability and completeness of data – it is a huge step on the road to creating real-time enterprises.

Production

of partsShipment of parts

Shipment of vehicles

Quality control

Assembly of vehicles

Inbounddelivery of parts

Inbound delivery of vehicles

Vehicle sales

Supplier Manufacturer Manufacturer Sales Customer

Guarantee & service data

Logistical Data

Lading list, date, origin

Identification data

Quality data

Order data

Logistical Data

Order data

Production data


Production control data

Mounting instructions

Production of parts

Shipment of parts

Shipment of vehicles

Quality control

Assembly of vehicles

Inbounddelivery of parts

Inbound delivery of vehicles

Vehicle sales

Supplier Manufacturer Manufacturer Sales Customer

Guarantee & service data

Logistical Data


Identification data

Quality data

Order data

Logistical Data

Order data

Production data


Production control data

Mounting instructions

Figure 21: Event Information in an Automotive Process Chain

34 of 50

6.4 EPCglobal’s Savant Middleware Concept Savant software was the original middleware component in the EPCglobal Network. It was designed primarily to process the data streams from the readers and other sensors and forward them to the relevant applications. Savant was also intended to filter the data generated by the readers, aggregate it, and supply it to the relevant enterprise software systems without overloading them with large volumes of data. It was to be open-source software, but the plans were not developed further. In fact, EPCglobal has decided to standardize the interface between the RFID middleware and the application software rather than to provide a reference implementation and software specifications. The decision as to how RFID middleware will implement the functions, say for filtering EPC data, thus falls to the respective software providers. EPCglobal refers to the interface between RFID middleware and an application as Application Level Events (ALE) specification. ALE contains functions that support the simple filtering and bundling of aggregated data. For instance, it enables filters to be specified that forward EPCs with particular bit patterns only. Other filters could be designed to bundle into one input event the multiple identification of one RFID transponder in a short space of time. Linking such filters would enable the data supplied from the different connected readers to be formatted and converted to the relevant events for the application. For example, many transponder identifications on one read device could be grouped to one goods receipt event for a delivery. Even if EPCglobal does not provide the Savant software, there are sufficient commercial software components available to handle these Auto-ID procedures effectively.

6.5 Databases RFID empowers companies to track and monitor their assets more closely, to attain greater transparency in their enterprises, and to base their decisions on real-time information. It cannot do this unless the data extracted during RFID procedures is stored appropriately and available reliably. The sample EPC-oriented architecture in Figure 22 illustrates just how complex the flow of data can be through IT systems operating with RFID procedures. In this case, RFID data is fed via RFID middleware servers to a company’s central data center. The data center runs the conventional business applications, whereby we need to distinguish between central and local data management. With central data management, only an object’s identification number is transferred (the information on the tag). Any other data associated with the object is stored in central databases. With local data management, the relevant data is stored directly on the tag in order to optimize the process runtime. A database connection would be needed in order to verify this data. However, this is not necessary for typical, local (autonomous) processes. We cannot say definitively that either one of these two data management models is more appropriate; their suitability depends on the actual conditions of use. In the consumer goods industry, the trend is moving towards a central data management model using EPC.

35 of 50

Database at data center

Standard applications

Carton/pallet with RFID tag-

Custom applications

RFID- antenna

Savant Server

Objects with RFID tag

Mobile RFID write-read device

ONS & PML services

Stationary RFID write-read device

Application server at data center

Savant Server

ONS (cache information)

EPC registry

Objects with RFID tag Carton/pallet

with RFID tag

Figure 22: EPCglobal-Oriented RFID Application and Communication Structure Source: Oracle/GS1 Germany/EPCglobal The underlying database system for data management must be regarded as part of the EPC network, that is, as a fixed component (object directory domain). Since RFID data cannot be processed by all business applications, the raw data must be kept in the database system. The database system must meet the following requirements:

o Quick access to individual pieces of information (index)

o Custom data filtering

o Long-term storage for process data, where appropriate with transfer to a data warehouse

o Secure access and high availability Databases also help tackle various security and confidentiality requirements, as do internal procedures, coding techniques, auditing, and access restrictions that can also be applied to RFID-relevant data. Additional security aspects are discussed in the next chapter. With the advent of RFID procedures and with data being centralized and accessed increasingly in real time, the issue of high availability has really come to the fore. Depending on the confidentiality of the data concerned, a range of technical concepts for high availability can be applied, such as cluster systems, standby databases, and replication databases. The system conditions and the processes being supported dictate which technique is most suitable.

36 of 50

7 RFID Security Aspects RFID data must be used in compliance with clear regulations concerning IT security as well as consumer and data protection. Consumer concerns about the general transparency of their actions cannot be dismissed. Conflicts may certainly arise and may result in restrictions on the systems. For instance, future developments will enable RFID tags on clothing and other consumer articles to be destroyed (kill command), removed, or rendered unreadable at the cash desk. The extension of mobile telecommunications networks also gave rise to heated debates about consumer protection. Yet the majority of consumers today carry their cell phones with them continually, fully aware that doing so permits their every movement to be tracked and traced. Nonetheless, those operating RFID systems must understand the gravity of their task and strive to provide transparency for consumers – thus supporting the social objective of informational self-determination. This is true for all systems in which consumer data will be or is already collected. For instance, it is already common practice (and a practice with widespread consumer support) for companies to use payments made by credit, bank, and loyalty cards as a means of extracting consumer data. A study conducted by the German Federal Office for Information Security (BSI) provides a detailed discussion of security aspects in the context of RFID systems: refer to the “Other Sources of Information” section. Consequently, only certain aspects are discussed in this paper. RFID systems and their security issues were being discussed in the press and publicly long before the technology was first applied in industry. However, the bulk of the discussion centered on the correlation between information on the RFID tags and personal information (or the collection of personal data) and the potential, however slim, for that information to be misused. In other words, it was really an issue for consumers. While the systems involved in current pilot projects for reporting and application testing do highlight the need for more customer information and disclosure as to how potentially personal data may be used (compare the BSI study p. 41 and p. 46 ff.), it is highly improbable that these industry-specific and logistics applications pose a direct threat to data privacy regulations. The technical security criteria currently relevant are discussed below and relate primarily to the safeguarding of critical company data and the misuse of such data. Of course, the protection of personal data might also be a consideration in this context (if RFID is used for employee ID cards or entry protection, for example). The main focus, however, is on the system security of RFID, say in logistics or other industry-specific applications. As in all areas of IT security, with RFID it is useful to distinguish between the protection targets of integrity, confidentiality, availability, and nonrepudiability. Depending on the chosen architecture, these criteria should be considered separately for RFID system operators and for users. According to holistic views of IT security, certain limitations must be defined prior to any such discussion. An RFID system is usually limited to the RFID tags, the reader, and the data transferred in this space (air interface). It might also include the threshold between the reader and the middleware that processes the data recorded. This means that when implementing an RFID system and designing its security concept, we have to ensure that the middleware and back-end systems are suitably safeguarded (see BSI study, p. 45).

37 of 50

RFID security for users The foremost protection targets for users (whether or not knowingly involved in an RFID system) concern confidentiality and nonrepudiability. Certain special cases (such as the use of RFID employee ID cards, travel documents, or means of payment) also create problems regarding the user’s “location privacy.” Access systems, in particular, rely on the exact assignment of persons (users) and RFID tags to check authorization. The process of unique identification and authentication (in the case of security revolving doors, for instance) involves collecting large amounts of personal data, which in itself needs to be treated highly confidentially. In addition, hidden RFID sensors can collect other position data that could be misused to create very accurate movement profiles. Adequate protection must therefore be provided to restrict who can access or read RFID data, in order to comply with the policy of minimal data collection. The architecture of RFID systems (the air interface!) poses an inherent problem as regards data confidentiality. Any data transferred during an RFID system activity (activation of the tag by the reader and the information emitted by the tag) could easily be the subject of eavesdropping if transferred uncoded (as is currently common practice). This possibility is, however, limited by the multitude of systems, frequencies, and codes that would call for in-depth system knowledge and considerable technical preparations in order to generate useful data from the transmissions. Anyone wishing to mount such an attack successfully would need knowledge of the systems used, as well as a suitably manipulated reader/receiver positioned unnoticed within the range of the legitimate reader-tag communication. In summary, we can say that in terms of RFID security for users, a combination of education about the use of personal data (ID cards, bank notes, and so on) and secure communication within the system (by screening, for instance, see BSI study p. 51 ff.) are vital for basic protection. In each case, additional measures may be necessary depending on the particular circumstances (especially when more complex information as opposed to simple tag data is transmitted in the RFID system).

RFID security for operators RFID system operators often map mission-critical processes such as supply chains or spare parts identification using RFID. The basic protection requirements are consequently extremely high, since company success is linked directly or indirectly to the availability of the RFID data and to its integrity and nonrepudiability. It is also in the operators’ interest to prevent third parties accessing system data, which could have drastic implications for business processes. The following aspects of RFID security need to be considered against this background:

o RFID systems must have a sufficient level of availability and operate reliably.

o The data transferred should not be easily altered.

o It must not be possible for false data to be fed into the systems.

As a wireless system, RFID is by nature easily manipulated (for instance, denial of service due to flooding of the electromagnetic field). This means that from the design stage on, care must be taken to ensure that the tag-reader communication is adequately encapsulated and shielded. The only reliable way of establishing whether a design is secure and provides sufficient availability is to conduct in-depth tests under varying conditions.

When using simple RFID tags, only downstream plausibility checks in the middleware or in the back-end system can prevent data being changed or system fraud. More powerful tags of the new generation do not impose this restriction. The use of crypto functions for one-way or two-way identification and authentication enables RFID systems to identify tags and readers as part of the system and even to encode communication within the

38 of 50

RFID system. The basic rule of thumb, however, is that simple RFID systems should store and transmit minimal amounts of data and that the application only gets its “intelligence” from the RFID data in the back-end systems.

More complex chips and readers are the only answer to some of the security problems facing RFID operators. Due to their high costs, such systems are only feasible for premium products and processes and for “closed loop” applications, in which the tags are used again and again. For more detailed information about countermeasures, see the BSI study, from p. 47, Section 7.7.

When it comes to general security for RFID systems, no one security architecture is suitable for all cases. A secure RFID solution must be designed on the basis of clearly defined stakeholders and their protection requirements. The security measures needed for a given application must be determined by the requirements of the business processes it is to support, the requirements of the stakeholders, and the requirements for extending the RFID system (both geographically and logically).

As long as only manufacturer and serial numbers are transferred and cannot be turned into usable information without a combination of detailed information from the back-end systems, the security risk posed by the use of RFID (say, in a company’s internal supply chain) can be deemed negligible.

Anti-counterfeiting

Another security issue sometimes referred to as “anti-counterfeiting,” concerns RFID tags and their unambiguity. Basic research is still needed in this area. For instance, original spare parts cannot be distinguished from counterfeit parts unless the tags are secure, that is, protected by a signature, for example. A Special Interest Group at the Auto-ID Lab at the University of St. Gallen is currently working on this topic: www.autoidlabs.ch/sigac

39 of 50

8 Sample Applications

8.1 Case Study: Logistics Processes at Hewlett-Packard GmbH With a product range exceeding 23,000 different products and trade relationships with over 110,000 suppliers, HP has been using RFID procedures for some time now. Doing so reduces stockholding and shipping costs not only within HP’s own supply chain but also for its trading and sales partners. RFID is currently deployed in 30 HP plants and logistics centers, primarily in China, South East Asia, the United States, and Brazil. HP originally installed the technology with a view to optimizing its own logistics processes. When Wal-Mart also required products to be labeled with RFID tags, HP was one of the first suppliers able to comply.

Figure 23: Goods Issue with RFID Antennas at HP’s Chester Plant Source: Hewlett-Packard RFID tagging is used for packaging units and pallets in the Sao Paulo plant (Brazil) and in the two HP logistics centers in Chester and Memphis (United States). RFID is used differently in the three locations: The plant in Sao Paolo produces printers and ships them to various logistics centers. Ink cartridges are packed and stored temporarily in Chester before being shipped to the Memphis logistics center, where RFID is used in the final packing, storage, and shipping of quality printers. In Memphis, RFID has eliminated the need for time-consuming bar-code readers and thus reduced the length of the logistics planning process for pallets from minutes to seconds. In Chester, the costs for handling packaging units and pallets within the distribution centers have fallen significantly because there are now fewer cost drivers, losses, and manual errors. And the full potential of this technology is not yet exhausted. Further benefits could be gained, for instance, by using other information in addition to the standard EPC information. The global use of RFID in the Hewlett-Packard supply chain is being systematically developed and implemented in collaboration with logistics companies, trading partners, and major customers. Thirty HP plants and logistics centers currently use RFID, a figure that is set to rise in 2005, and the company also plans to implement RFID transponders of the second EPC generation.

8.2 Case Study: Flexible Automotive Production Processes at BMW RFID is a key technology for industrial manufacturing when it comes to realizing flexible production concepts. One of its major attractions is that the data carriers (tags) are not very sensitive to harsh conditions and dirt – specialized industrial tags can be used at temperatures in excess of 200°C. The possibility of saving data on the fly also opens the way for creative, local system architectures.

40 of 50

Figure 24: Basic Architecture for RFID in Industrial Manufacturing Source: Siemens AG RFID is often used to identify support modules that (in a number of forms) transport a range of articles from single circuit boards to vehicle bodies around a particular site. An RFID tag is mounted on the supports and accompanies the article being manufactured along the entire production route. Since the support module and the tag can be reused for subsequent production runs, this is an example of a “closed loop".

Figure 25: RFID Transponder with Data Store on the Body Panel at BMW Photo: Siemens AG An extremely simple scenario could involve storing the production order identification number in the data stores, making the article automatically identifiable at the individual processing

Processing stations

Initial writing with data from MES layer

Reading & writing for checks andquality assuranceat controllevel

41 of 50

stations. A shop-floor control system (manufacturing execution system, MES) uses the number to define which production step is required at a given station. With more advanced concepts, such production instructions are stored directly on the data carrier. The advantage of this is that it enables the production line to produce goods independently (those that are fed onto the line, at least) without accessing higher-level system components. RFID technology consequently increases the availability of the equipment. BMW has successfully implemented a concept of this kind. As part of a modernization and expansion scheme at the BMW plant 1 in Munich, the new conveyor system for assembling bodies for the 3 Series was equipped with RFID data carriers and write-read devices. Each body support is fitted with an RFID data carrier, which stores all the data relevant to the production of the vehicle and makes the body identifiable at any time. At BMW, this is an important factor for production control, documentation, and quality assurance. Other advantages: The independent intelligence on every panel, coupled with seamless system integration in the Simatic automation landscape, lends the system utmost flexibility. The result is that one line can produce different models at the same time and be fed new models successively.

8.3 Case Study: Mobile Maintenance Solution at Fraport AG With global sales revenues of 1.8 billion euros, Fraport AG is a leading player in the international airport industry. It also owns and runs Frankfurt Airport. Its business areas – ground services, traffic and terminal management, communication services, and property and facility management – cover the entire range of products and services found in the airport industry. Lead-managed by SAP Consulting, an ambitious mobile RFID maintenance scenario was realized for Fraport AG. It combines SAP Mobile Asset Management (MAM) software with mobile handheld computers and RFID technology to give Fraport an innovative complete system that enables it to plan maintenance work for a wide range of different technical components, such as fire shutters, on the airport grounds – and to be sure that such work is legally watertight. The mobile RFID maintenance scenario is based on the MAM software solution and increases safety at Frankfurt Airport. It replaces working papers, provides detailed real-time data, and bridges the information gap between back-end systems and events on the ground.

Optimized paperless maintenance processes Property and facility management for Fraport AG involves servicing approximately 420 buildings and systems at Frankfurt Airport. One of this area’s key tasks is to regularly service and check technical components that are subject to legal maintenance requirements. As airport operator, Fraport is obliged to provide supporting documents. The company uses the powerful SAP Plant Maintenance solution to control its maintenance activities. This application was enhanced to include a mobile solution, Mobile Asset Management. Transponders with innovative RFID technology interact seamlessly with robust handheld computers to create an efficient complete system.

Winning combination of innovative technologies

Rather than printed work instructions, maintenance mechanics now use a robust front end, which can read and write data from and to the RFID transponder chips and communicates with the MAM application. Fraport supplies the mobile devices with due maintenance orders from the central SAP system. Following authentication and login, a controlled, step-by-step dialog begins between the MAM application, the mechanic, and the RFID labels installed in the ventilation systems. This procedure ensures that the

42 of 50

equipment is maintained in a sequence of precisely defined process steps. Any defects are recorded using a comprehensive system of damage codes. The result is a complete, electronic record of every maintenance activity along with the on-site conditions. When maintenance is complete, data is read from the mobile devices and the information is transmitted electronically to the back-end system. In all, MAM has greatly accelerated the maintenance processes, dramatically increased data quality and security, and opened new options for reporting on damage and product life cycles. For Fraport AG, the SAP solution is a future-proof, modern, and mobile maintenance solution that can be adapted to cope with any maintenance task. In May 2005, Fraport AG went live with the most recent version of the SAP MAM solution, Version 2.5, which includes RFID functions as standard.

Figure 26: Paperless Maintenance Scenario at Frankfurt Airport, Optimized by RFID Source: Claus Heinrich (SAP): RFID and Beyond (see sources of information)

Six-figure savings Fraport gained a complete electronic maintenance process ensuring that equipment is serviced correctly. Innovative technologies enable the maintenance tasks to be conducted faster than before. And it is now possible to reproduce every process step at the push of a button. Consequently, Fraport benefits from the highest possible transparency when it comes to fulfilling its duty to supply documentation to review panels and legislators. Fraport has mounted RFID tags on 22,000 fire shutters at the airport, for which it previously needed 88,000 order data sheets annually for maintenance purposes. In an interview with Computerwoche (Issue 25, June 24, 2005, page 24), Dr. Roland Krieg, CIO at Fraport AG, said that RFID had enabled the company to cut the costs for essential documentation by approximately 450,000 euro annually. For comparison, he quoted the one-time costs for RFID tags and reading devices at below 100,000 euro. This is a convincing return on investment in less than one year.

43 of 50

8.4 Case Study: Large-Scale RFID Roll Out for Metro AG Logistics Metro AG is planning an extensive RFID implementation, covering its own logistics processes through to the restocking processes in retail outlets. This calls for RFID to be implemented in many stores as well as in the associated merchandise distribution centers. Since different IT applications have to be connected to the RFID hardware, standard middleware would be advisable. The middleware checks the status of the RFID readers, controls the feedback providers, and supports the process-based tasks of the RFID readers, writers, and printers. It transmits the relevant data to the business applications. The software is based on the RFID domain model developed by IBM for processing RFID events and connecting them to business applications. In 2005, Metro AG selected the WebSphere product family, IBM’s RFID middleware, for the large-scale roll out in its stores and merchandise distribution centers. It forms the backbone for supplying the application systems with RFID events and for integration in the existing system management tool. Research showed the UHF technology to be most suited to Metro’s Future Store initiative. The manufacturers of the retail goods attach the RFID tags to the logistical units (such as pallets). These details are transmitted to Metro in a standardized EDI message. When goods are received in merchandise distribution centers or when direct shipments are made to supermarkets, the expected goods receipt can be compared with the actual delivery quantity recorded by the RFID reader. The result of the comparison can be sent to replenishment planners by means of a signal, enabling them to take immediate action if variances are identified. Other postings required for the deliveries take place in the retailing systems.

Figure 27: Layer Model for a Comprehensive Auto-ID Application Architecture Source: IBM The system would typically be structured as follows: The edge domain, the software installed in the stores, controls the RFID reader hardware. The data is transferred by means of a communication link (uplink) to the “WebSphere Premises Server,” which is the core of RFID system integration. The next level of application integration takes place in the integration layer, which – like the premises infrastructure – is operated centrally in the data center. This central integration layer enables the RFID data to be supplied directly to different applications. The WebSphere RFID middleware is connected to the different business applications by a process integration layer in order to optimize the number of IT interfaces. Similarly, the

Enterprise & Business Application

Domain

Antenna & Reader Domain

Tagged Object Domain

Security & Privacy Management

Edge Domain

Premises Domain

Business Process

Integration Domain

Systems Management Domain

Object Directory Domain

T

Tooling – Support for customer-specific logic

T T T

T

44 of 50

maintenance and operating requirements of these interfaces are significantly lower than for point-to-point couplings between the middleware and the individual target applications. The advantage of RFID technology is that it accelerates loading and unloading processes (in merchandise distribution centers, for instance) and enables data to be updated far more accurately in the business applications. In the old, bar-code-based procedures, much depends on the reliability of the employees reading the labels. RFID technology also eliminates the problems associated with dirty bar codes. Another advantage is that incorrect deliveries can be prevented or at least reduced, since employees can be given compulsory confirmation of correct loadings.

Figure 28: Configuration of RFID Antennas on Doors in the Metro Warehouse Source: IBM Kurt Salmon Associates analyzed the improvements. The study found that RFID technology had reduced losses during transit by 11% to 14%. The availability of articles in stores had increased by up to 14%. Costs in merchandise distribution centers had fallen by around 11%. The analysis also shows that substantial savings can be made if migration to new technology is supported by process changes. Metro plans to equip a total of 250 stores with this technology. Its suppliers will be introduced gradually to the tagging process, from pallet tagging progressing to case/item tagging. IBM Global Services realized the complete implementation. The following subprojects were also carried out: process analysis, description of the new process, basic tests for selecting necessary hardware and software, check on (documentation of) environmental conditions in target stores, setup of RFID hardware in target stores, configuration and integration of WebSphere RFID middleware. The system has been in operation since November 2004 and is used daily.

Dock door

RFID controller (edge domain)

Data center

Premises/integration layer and warehouse management

Trash door

Pallet/goods in storage area

Front doors with readers

All doors equipped with RFID antennas. All readers connected with the RFID controller

Warehouse

45 of 50

9 Glossary Air interface The area between the reader and tag antennas and the

communication interface required here.

ALE Application Level Events specification; the interface between the RFID middleware and application, defined by EPCglobal.

Auto-ID The process of identifying objects automatically and without contact. It is the umbrella term for procedures such as bar-coding, and the RFID research area of the Auto-ID Labs at MIT, St. Gallen, Cambridge, and so on.

Chip The data carrier in the tag.

DNS Domain Name Server: the address directory for Internet addresses, converts www addresses to IP addresses.

EICAR European Institute for Computer Anti-Virus Research; also operates an RFID task force.

EPCIS A middleware component for accessing the EPCglobal Network.

EPC Electronic Product Code: The manufacturer, product, and serial number allocated by EPCglobal.

ERP Enterprise resource planning (system)

ETSI European Telecommunications Standards Institute

Event An event in the Auto-ID environment, such as the reading of an RFID tag.

GDS Global data synchronization: The process of matching article master data between trading partners in line with global standards, developed by GS1 (previously EAN.UCC).

GTIN Global Trade Item Number: an EAN/UCC standard.

Inlay A raw version of an RFID tag for attaching to labels or for embedding in a smart card.

Inlay The connection of an antenna structure with a chip to form a transponder component. When connected to the final carrier or label, an inlay becomes a tag.

ISO The International Organization for Standardization

ITIL IT Infrastructure Library: A set of ITIL guidelines commissioned by the British government. It is the de-facto standard for service management around the world.

KPI Key performance indicator

M2M Machine-to-machine communication

MES Manufacturing execution system

Middleware A software component on the middle layer of a hierarchical software architecture.

NVE The dispatch unit number (Nummer der Versandeinheit in German) for identifying pallets, shipping cartons, parcels, packages, and so on.

ONS Object Name Service: The central directory of EPC numbers in the EPCglobal Network.

46 of 50

PLM Product Lifecycle Management

RCS Radar cross section: a fictitious surface used in radar technology to characterize the reflectivity of objects. In the area of RFID, it specifies the reflectivity of tags.

RFID Radio frequency identification

Savant A middleware concept designed by EPCglobal for RFID systems. Original plans to develop it into a marketable product were called off (see the “Infrastructure” chapter).

SCM Supply chain management (system)

SGTIN Serialized Global Trade Identification Number: a GTIN combined with a serial number to identify objects uniquely; an EAN/UCC standard.

SKU Stock-keeping unit: a packaging unit in a warehouse that bears an RFID tag.

SLA Service level agreement: a contract regarding services and assessment criteria.

SOA Service-oriented architecture: a characteristic of flexible middleware, in particular.

SOAP Simple Object Access Protocol

Tag An (electronic) RFID label comprising a chip with a data store and an antenna for transmitting data via radio frequencies (RFs). Tags can be mounted on adhesive labels, in smart cards, in glass cylinders, or in a plastic casing.

Transponder A synonym for a tag.

WSDL Web Services Description Language

XML EXtensible markup language: the Internet standard for formatting information and documents.

47 of 50

10 Other Sources of Information

Books Elgar Fleisch / Friedemann Mattern Das Internet der Dinge, Springer Verlag, Berlin, June 2005, ISBN 3540240039

Klaus Finkenzeller RFID-Handbuch, 3rd Edition, 2002, Carl Hanser Verlag München, ISBN 3-446-22071-2.

Claus Heinrich Adapt or Die: Transforming Your Supply Chain into an Adaptive Business Network. 2003, John Wiley & Sons, Inc. ISBN 0-471-26543-8

Claus Heinrich RFID and Beyond: Growing Your Business Through Real World Awareness, 2005 John Wiley & Sons, Inc. ISBN 0-7645-8335-2

BSI-Studie Risiken und Chancen des Einsatzes von RFID-Systemen (RFID – Security Aspects and Prospective Applications of RFID Systems), trends and developments in technologies, applications, and security, 2004, Federal Office for Information Security (BSI) ISBN 3-922746-r56-x. http://www.bsi.bund.de/fachthem/rfid/studie.htm

Internet Addresses Universities, organizations, bodies, cross-manufacturer Internet pages www.aim-de.de AIM e.V. www.airlines.org ATA: Air Transport Association www.autoidlabs.org Global network of Auto-ID centers www.bitkom.org Largest European association for the IT industry www.bme.de The German Association of Materials Management,

Purchasing and Logistics (BME) www.bsi.bund.de German Federal Office for Information Security (BSI) www.chep.com Global pallet and packaging logistics www.bvl.de Bundesvereinigung Logistik (German federal association

for logistics) www.gs1-germany.de GS1 Germany GmbH (previously CCG) www.epcglobal.de GS1 Germany GmbH (previously CCG) www.epcglobalinc.org EPCglobal Inc. www.future-store.org Metro Future Store Initiative www.item.ch Institute of Technology Management, University of St.

Gallen www.licon-logistics.de Logistics association, supported by Kühne + Nagel,

Siemens, and others www.logistics.about.com Information www.logistik-lexikon.de Online glossary www.m-lab.ch M-Lab (University of St. Gallen and ETH Zurich) www.myLogistics.net Logistics portal from AXIT AG, good keyword search www.rfid-handbook.de Internet page for Klaus Finkenzeller’s RFID Handbook (see

above) www.uc-council.org Uniform Code Council, United States

48 of 50

Media www.dvz.de German transport journal/German logistics journal www.logistra-net.de Specialist magazine www.eweek.com Specialist magazine focusing on RFID www.funkschau.de Specialist magazine www.ipm-scm.com Specialist magazine for supply chain management www.ident.de Specialist magazine www.logistik-heute.de BVL magazine www.rfid-im-blick.de Specialist magazine, Germany www.logistik-inside.de Specialist magazine, detailed RFID documentation www.rfidforum.de Specialist magazine/smart card forum www.rfid-forum.de AIM press organ, entitled “ident” www.rfidjournal.com Specialist magazine, United States, also organizes

conferences www.themanufacturer.com Specialist magazine, United States

49 of 50

Acknowledgements This white paper, “RFID – Technology, Systems, and Applications,” was compiled in consultation with the BITKOM Radio Frequency Identification (RFID) Project Group. Representing renowned companies from across the ITC technology world, its authors have demonstrated outstanding teamwork with fellow participants in the ITC value chain, reflecting the interaction between different technologies and applications. Our sincere thanks go to the authors, Dr. Norbert Ephan, Kathrein Werke KG, Jens Gerecke, Microsoft Deutschland GmbH, Wolf Rüdiger Hansen, Diamond Systems GmbH, Dr. Frank Gillert, Infineon AG, Michael Hegenbarth, Infineon AG, Tanja Popova, GS1 Germany GmbH, Georg Raabe, Hewlett-Packard GmbH, Sebastian Rohr, Computer Associates GmbH, Tamara Sass, Oracle Deutschland GmbH, Thomas Schuh, T-Systems GmbH, Peter Sommerfeld, Fujitsu Siemens Computer GmbH, Markus Weinländer, Siemens AG, Wolfgang Weyand, IBM Deutschland GmbH, and Dr. Alexander Zeier, SAP AG for their commitment and creative discussions. We would also like to thank the other members of the project group for their interest in the progress and for their constructive criticism. The companies provided most of the figures, with the authors selecting those which they felt best illustrated each case and agreeing on their publication in this BITKOM white paper. We are most grateful to the project leader and editor, Wolf Rüdiger Hansen, Diamond Systems GmbH, for expertly coordinating the many contributions. Special thanks go also to the deputy project managers, Georg Raabe, Hewlett-Packard GmbH and Dr. Alexander Zeier, SAP AG, to whom the final white paper owes much of its style and cohesion. Further profound thanks go to SAP AG who has conducted and sponsored the English translation of this white paper. Dr. Birgit Heinz, BITKOM e.V., edited the entire text and in the event of queries can be contacted at [email protected]. Berlin, December 12, 2005

50 of 50

The German Association for Information Technology, Telecommunications and New Media e.V. (BITKOM) represents 1,300 companies, with some 700 being direct members with sales revenues of approximately 120 billion euro and around 700,000 employees. These companies include manufacturers of terminals and infrastructure systems, as well as providers of software, services, new media, and content. BITKOM is working to improve the regulatory framework in Germany, to modernize the education system, and to promote innovative economic policy.

German Association for Information Technology, Telecommunications and New Media e.V. Albrechtstraße 10 10117 Berlin-Mitte Germany Tel.: +49 (0)30/27 576 - 0 Fax: +49 (0)30/27 576 - 400 [email protected] www.bitkom.org