Acsis White Paper

ACSIS, INC.

Lessons From The Front Line

RFIDIntegration  Acsis, Inc. 3000 Lincoln Drive East • Suite C Marlton, NJ 08053 Phone 856.489.4900 • Fax 856.489.1007

Table of Contents Introduction: Acknowledgments: The Basic Pieces Tags: Readers: Antenna Host Application The Environment RFID features and benefits Time for a sanity check Major areas of concern: Item Environmentals: System Environmentals: Data Requirements: Tag structure: Tag mounting: Reader capabilities: Antenna selection and placement: Software Application notes Summary:

RFID Integration

3 4 5 5 5 5 5 5 6 6 6 7 8 8 9 9 9 9 10 11 13

Page 2

Introduction: Purpose:

It is my hope that this document will provide a framework from which you can evaluate the use of RFID technology for your company. Background:

Acsis is a company that specializes in enterprise data collection solutions for global companies utilizing SAP R/3 software. Acsis, through it’s Data-Link based solutions, is called upon to provide world-class mission critical data collection utilizing the latest technologies. Acsis began working with RFID technologies many years ago in order to fulfill our customer’s requirements. Our initial investigation provided us with more questions than answers. This started us down a path of discovery. Competent education was hard to find, and in many cases required traveling halfway around the world from our US headquarters. Having learned a great deal of information in formalized settings, the real learning was about to begin. The majority of knowledge in RFID is gleaned from real word testing. This does not include empty cardboard boxes pushed at .0042 mph through a perfectly aligned portal. It comes from tagging real products in real scenarios. Setting up equipment at client sites and experiencing real in-field production issues is invaluable. Quite frankly most material handlers encountering the technology for the first time view it as a threat and will not be very helpful in making it a success. Also the differences between sites, areas inside the same site, and the limits presented by partner facilities can create havoc and put us back to the drawing board for a solution. The industry itself has in many ways been it’s own worst enemy. It has pushed the idea of massive volumes and productizing offerings that can’t be delivered at this time, if ever. Everyone wants to believe that RFID can be handled in the same fashion as a barcode. It has many similarities, but a greater number of differences. It takes all of the know-how used in barcode systems coupled with an entire new set of skills to provide a solution for a particular application. At this time most RFID implementations are more dissimilar than similar. The discussions of standards, penny tags, and all-powerful readers that can interrogate all forms of tags have paralyzed the market. Meanwhile the single real criterion of a positive ROI has taken a back seat in this dot-com like environment. Too many pundits, who have never actually implemented and supported a real production RFID application, tend to profess the requirements and system capabilities with no real experience to support their statements. RFID is a powerful and wonderful technology when used correctly and can provide functionality not available in any other way. Many of the systems we have been involved in have tremendous returns on the investment made. After all, that should really be the goal of almost all technical projects. Dave Harty Director of Research and Development Acsis, Inc. [email protected] www.Acsisinc.com

RFID Integration

Page 3

Acknowledgments: While I cannot possibly recall all the individuals that have had some level of impact on this paper, Acsis has had the pleasure of working with or having been exposed to the following high quality organizations. Please accept my apology for any company, which has been overlooked Amtech AWID Escort Memory Systems Gemplus Global ID HID Identec

RFID Integration

Intermec KSW Lucatron Matrics Omron Phillips PSC

Page 4

RFID, Inc Savi SCS Symbol Texas Instruments Transcore Zebra

The Basics of RFID

T

o integrate an RFID system successfully a strong understanding of the basic elements and their behavior is a true requirement. While some very simplistic systems, such as basic access control and vicinity card systems, require drastically less RFID systems knowledge all implementations are subject to the same issues and a thorough knowledge of these will help all who work with the technology.

The Basic Pieces The items have been discussed in great detail in many documents and brochures, but we can reduce them to their simplest forms for this discussion. Tags:

Readers:

Antenna

RFID tags are a good starting point and while there are a great many different types, active vs. passive hard-tag vs. insert (or smart label) they all serve the same purpose. They are electrical devices, using some form of antenna to communicate. They are the item, which is used to identify something to an RFID tag reader. RFID tag readers are the next logical items as we have referred to them in the previous section. A reader is just what it implies, a device that reads RFID tags. As with tags there are a great many different types with various capabilities. We can leave the definition to the reading of RFID tags at this point.

RFID tag readers use an antenna to communicate to the RFID tag through the tag’s antenna. Some readers have integral antenna while other can have various types and sizes of antenna fitted to them.

Host Application

Now this is a real broad category. While we can have embedded programs in some readers let us define the host application as something that tells the reader when to read RFID tags and then do something with the data that it receives.

The Environment

The single greatest item you will have to deal with in implementing RFID is the environment. Again we need to narrow the scope, however this one can not be so simplified as the other components simply for the fact that it is the least understood, yet can have the greatest impact on the system, usually negatively. For our purposes lets define the environment as the physical (including unseen items such as radio and magnetic waves) items between and surrounding both the tags themselves and the area in which we are trying read tags. This would include the items that the tags would be placed upon as well.

We will probe deeper into each of these categories and how they affect decisions such as systems architecture, actual RFID components to be used (types of tags, frequency, and readers), application logic, physical environment and business processes as we progress.

RFID Integration

Page 5

RFID features and benefits The list of benefits and features afforded by an RFID system is quite impressive. They can include applications that remove human intervention completely, allowing for the changing of data in the field, operation where line of sight is not possible, and acquiring of information from multiple items at high speed, and the ability to identify items nested inside of packaging without opening the packaging.

Time for a sanity check The benefits and features provided from RFID technology are impressive, however, one must remember that they are potential features and benefits, not guaranteed ones. It is the understanding of what can truly be delivered and utilized in the individual implementation that determines the project’s success or failure. First we start with some basics. Why do you want to tag it in the first place? Many projects should stop at this point. If we aren’t solving a problem then do not proceed. There is no need to try tagging something simply to prove you can. We will address that shortly. However, I think we can all agree that if the problem can be equated to having a cost to the company of 1 million dollars a year and implementing RFID technology to solve the problem costs 2 million dollars a year, we should not implement A positive ROI is the driving force to any successful project. Please keep in mind total system cost in calculating your ROI. A great number of projects have concentrated on some magical tag cost (usually based on nothing) only to find out the infrastructure and software development required to support those tags has a price equal to the gross national product of a small nation. There are plenty of systems with high dollar tags with great ROIs, and even more systems with low cost tags that have a horribly negative ROI when the total system cost is calculated. Simply put if the total cost of the system is more the value of the benefits then STOP!

Major areas of concern: This leads us to categorizing some major areas, which we need to address in order to achieve success in our implementation. We can break our system requirements into several major areas of concern. While there is no steadfast rule in how to categorize these or the completeness of this list our experience has shown each of these to be significant factors to consider. •

Item Environmentals

•

System Environmentals

•

Data Requirements

•

Tag Structure

•

Tag Mounting

•

Reader Capabilities

•

Antenna Selection and Placement

•

Software

RFID Integration

Page 6

Item Environmentals: In this area we will discuss the item to be tagged. This may be a pallet, tote, carton, case, individual item, truck, railcar or a tube of toothpaste. The item itself really doesn’t matter. The issues are all the same. What is the item made of? This question is tricky because in it is a literal question pertaining to the item to be tracked in its entirety. I have given many presentations in which I use the following example. Company ABC wants to track cartons with RFID tags. Ok, cartons of what? Company ABC informs us their product is soda. Ok, bottle or cans or both? Company ABC only produces cans. If we wanted to tag a carton of canned soda, we would have the following three major materials (we will ignore ink, paint, etc at this point) to deal with. 1. The cardboard of the carton. 2. The metal of the can. 3. The soda fluid itself. In the above example we are effectively tagging a block of metal with a small amount of cardboard and air gap between the tag and the metal. In this case, as metal is a reflector; the internal contents really don’t have an effect worth considering. So as you can see we went from having to track cartons, then soda, to finally tracking a block of metal wrapped in cardboard. Not as simple and clean as the original statement from Company ABC that they simply wanted to track cardboard cartons. Now that we understand what sort of materials we are trying to tag, we can think about how to properly tag the item. We need to ask in what ways will the tag be read? One at a time or in groups, what distances will we need to read, what environment will the item be exposed to as far as temperature, mixed pallet storage, damage potential, handling equipment, etc. All of these factors and more will affect the tagging decision. The harder questions are the following: What, if any, business processes need to be changed in order to work with the technology? What, if any, physical changes to the carton or its dimensions need to be adjusted? These are traditionally the make or break areas for a great number of projects. The technology can work if certain conditions are acceptable. Do workers need to keep one particular side of the carton exposed all the time for the tags to be read? Does the tag stick out 15mm from the side of the carton? Understanding the confines of the item’s environment is critical to success. Knowing the material composition of the item we want to tag, the business processing involved in handling the item, and the physical issues in material handling equipment, storage strategies, storage environment, physical size, shape, and appearance all affect the correct tag decisions. We will discuss tag mounting and structure in later sections.

RFID Integration

Page 7

System Environmentals: In this area we will discuss the area in which tagged items will be utilized. This may be a manufacturing facility, warehouse, store, yard-scale, shipping dock or some third party location. What are the items that will impact the systems capabilities? Is there a great deal of machinery, PLC, CNC Mills, etc that can contribute to some white noise and affect our system? Will the item(s) to be dealt with in an area where some level of isolation occurs? What kind of distances are involved? Do the items stay put for any period of time or do they pass by at full forklift speed on their way out the door? Are there pallets of the same types of items sitting in the doorway adjacent to the one we want to work with? What are the physical surroundings? The greatest challenge to many of these questions is the variation of the response based on different locations for the same company. In most cases they are different at most of the locations for many large companies and requires grouping similar sites together.

Data Requirements: There are three areas of data requirements to be considered. They can be classified into the following categories: 1. Tag Data Storage 2. Data Transmission Requirements 3. Back-End Data Storage Many RFID tags have the ability to store and edit user data on the tag itself. This, while an extremely attractive feature, is not implemented very often. In the cases where it has been used, a small number of implementations actually edit the information after the initial writing of data to the tag. If data is to be stored and edited on the tag then there are a few things to consider. First size does matter in storage with RFID tags. Most people say things like “I want to keep the service history on the tag.” Or “The delivery details including hazardous goods information for all items on the pallet.” The problem arises when people want to store 2000 characters on a tag that can hold about 100 characters. Data integrity is another area. While one should never consider using the tag as the system of record for any data, storing the right data in the right situation can be an extremely powerful feature. A backup or copy of the data needs to be maintained somewhere for use when the tag is not available, not working, lost, or destroyed. Security in systems is always an issue. While the small bits of data that would be stored on a tag would be undecipherable to virtually anyone, it is the data loss or corruption, intentionally or not, that is of greatest concern. This effectively leaves one with an un-labeled item. The data transmission requirement is one of the items that usually rears it’s ugly head near the end of the project. Most RFID implementations provide for a great deal of data to be transmitted at very high speeds. The impact of this increased data needs to be addressed in order to support the solution. In the cases where the tags are only a unique id and all data must be processed by a host system, if there are a high number of tags entering and leaving read zones the data transmission requirement can be quite high. The data we transmitted above probably is going to be recorded.

RFID Integration

Page 8

In most systems we have been involved with, reads were recorded if for no reason other than an audit trail of activity. In most cases it had some complex logic applied to determine patterns of activity. Therefore the storage requirements, including the ability to supply responses to the system at high speed, are critical.

Tag structure: Tag structure refers to the physical structure of the tag. Tags can be in an inlay such as a smart label, credit card style, or what is known as a hard tag. Many factors must be considered in tag structure selection. First and easiest is the true smart label. If the tag needs printed information to be applied on demand then there really is only one format, the smart label. There are many sizes, which may affect the antenna options, affecting performance. Hard tags are available in many formats, encased in molded plastic, protected against harsh chemicals, in rubber discs, etc. The item environment discussed earlier drives many of the requirements for the tag structure. The other area affecting tag structure is the system environmentals, which affects the antenna requirements for the tag.

Tag mounting: Tag mounting is another overlooked area. The tag must be affixed to the item in a manner that insures that it survives the system environment. It must protect the tag; possibly provide a critical level of offset, and provide a required angle possibly needed by the system design.

Reader capabilities: Readers are available with many different capabilities. Most have a great number of configuration options available, however, a great many of these are sparsely documented and usually are gleaned after a great deal of experimentation. Readers may control multiple antennas. It is important to note that these are almost always switched so the greater number of antennas a reader must drive the greater the delay before any individual antenna can be polled again. Advance capabilities to utilize the reader’s processor to provide some logic in the unit can make the difference between success and failure. There are multiple connectivity options and one must make sure they allow for the support needed in an enterprise solution.

Antenna selection and placement: Another major area to address is the antenna system. We are specifically discussing the reader’s antenna having left the tag antenna to the tag structure section. Some readers have integrated antenna therefore eliminating this as an area with any flexibility. For the rest of the reader community the selection of the right type of antenna, cable and their placement is an area of almost endless possibilities. Careful consideration needs to given to allow for the area of coverage desired, while attempting to control the read zone to avoid spurious readings that are in affect not valid. In dock door portals, a balance between controlling the RF field and distance can be a challenging goal. Utilizing reflective and/or absorptive materials to control the field may be what is needed to provide a robust solution. Business process control and environmental changes may be required in many instances to provide a successful implementation.

RFID Integration

Page 9

Software We have spent time discussing the physical items involved including the tags, readers, antennas, and physical surroundings. Now we move to the software arena. This is the second most overlooked area in RFID systems consideration. As we have discussed there are a great number of issues that arise in RFID systems that need to be resolved. Failure to automate the resolutions to these issues will result in additional workload needed to make adjustments, eliminating any benefit from the system. The software must be robust enough to handle the multiple reads supported by RFID technologies, the understanding of the expected number of items that are to be in the read zone, the coordination of timing and material flow, the user feedback features, event management, backend systems updates, and integration to other associated control systems. Without providing these types of support functions, the system is an isolated island of information and will rarely provide a positive return on investment. The logic of understanding the condition of readers in an area in relation to users, operations, feedback, and various other systems requires a high level of understanding in advance data collection techniques. RFID is a form of automated data collection. By itself it has limited use. Coupled with an application that compensates for it’s weaknesses, and at the same time fully utilizes its incredible power, a solution can be designed that provides incredible benefits to the company; cost reduction, increased accuracy, improved workforce efficiency, streamlining business process, and improving the company’s ability to execute. In the days of running supply chains leaner and at the same time more responsive, RFID in many cases provides the needed functionality to support these elusive goals.

RFID Integration

Page 10

Application notes Now that we have defined the major components and the areas of concern in designing an RFID system we can discuss some applications and their challenges. These are derived from real experiences in RFID systems design and implementations. Line of sight:

Line of sight is one of the extremely powerful features of RFID. It can also create some real challenges at the same time. The idea of not needing to acquire line of sight, as in a barcode system, is ideal in many situations. There are, however, many times when it introduces problems that do not exist in systems that require line of sight or isolation of a data capture point. The following are two examples of line of sight complications. BIN PICKING SCENARIO:

Figure 1 A1

A2

B1

C1

RFID Integration

B2

C2

A3

B3

C3

In the example in figure 1, the warehouse storage area has a series of bins. The pallets in the bins have RFID tags on them. The material handler approaches the rack and wants to scan the pallet in the center to confirm removing this item to fulfill a pick request. The handler removes the pallet from the center bin labeled B2. The handler triggers a read in either a handheld reader or a forklift mounted reader using a high frequency or ultra high frequency system. The reader returns four (4) different tags. Which tags were returned? Can we assume the first tag returned is closest to the reader? We cannot be sure which tags were returned. The structure of the bins, the items on each pallet, the angle of the antenna of the reader when triggering the read, the mounting of the tags to the pallets, and the attribute of each individual tag make this impossible to decipher with certainty. One might first think that time could be the deciding factor. Surely the closest tag to the reader was the first one returned. For all of the above reasons and the fact that systems that handle multiple tags apply an anti-collision mechanism, time cannot be used in this scenario. Note this does not apply to triangulation used by RTLS system, that is a completely different scenario.

Page 11

DOCK DOOR PORTAL SCENARIO

Figure 2 F O R K LIF T W ith 4 P allets

F O R K LIF T W ith 4 P allets

In figure 2 there are multiple products about to be loaded onto two different trucks for two different orders. There are antennas in either side of each doorway. The company wants to know which pallets went to which order and they want as little user interaction as possible. What are the potential issues? First there is a good chance that both doorways will read some of the tags from the adjacent doorway. Some tags may not be read at all, from being obscured, broken, etc. Some might only be read from the adjacent doorway. The handler may run in and out of the read zone several times positioning his forklift resulting in multiple reads. The handler may also need to un-load a pallet or more as the truck becomes full. While shielding, feedback lights or buzzers, and software logic can overcome these issues in many cases, you start to see things are more complicated then first envisioned.

It becomes clear that line of sight has shortcomings in applications and that there are some things that are better served by technologies that actually require line of sight to provide the level of control and user feedback. The first part of RFID is “Radio”. This is a transport, which we are limited in our ability to control and focus. In a barcode environment we can physically see the laser illuminate the barcode and usually receive an audible beep with a valid scan. Not having a direct parallel in typical RFID technology, one must develop equivalents where they make sense or different methods of assuring the right things are happening in our system to insure it’s success. Multiple reads: HIGH DENSITY SCENARIO

Another powerful feature of RFID technology is the ability to identify a large quantity of items in a short time span. As companies strive to increase efficiencies in their supply chains this capability holds great promise. This also presents areas of concern. As in the dock door portal example above, if we are reading a large number of tags simultaneously we somehow need to be assured that we actually read them all. Also, did we read more than we expected? How to recognize that we have exceeded our expected number of reads and a method for dealing with that scenario needs to be defined. There also looms the possibility of saturation. We have seen scenarios where simply too many tags were in the read zone at a given time. There are limits to everything and RFID is no exception. RFID Integration

Page 12

System Environments: INTERFERENCE SCENARIOS

As RFID technology is based on radio communications, there are naturally interference issues to be dealt with. Shop floor equipment such PLCs, motors, switches, etc. may create enough “white noise” to adversely affect the system’s ability to perform properly. Physical items in the area such as loadlevelers, cranes, forklifts, and other equipment may create additional interference. The wide variation in locations and between parts of the same facility can make system design an extremely elusive issue. These inconsistent environments can rarely be served by a single configuration. Some level of customization per read zone may be required. These may be small, but they add to the complexity of the system design and the raise support costs.

Summary: While a great deal of the information portrayed in this document might be viewed as critical, this is not the purpose. It has been constructed from a critical viewpoint of people who have to deliver systems and stand behind them. All of the issues raised in this paper can be and have been overcome. The solution costs and dependencies sometimes are not as palatable once the hurdles have been surmounted. This is no different than many other technical projects. The goal of the paper is to ground your thinking as to the reality of RFID technology as it exists today and for the near future. Clearly some people will read this paper and decide not to pursue a project they may have been considering. That is probably a good thing. Others will simply apply the ideas and concepts presented here to their own situation and this will drive them to clarity. They will move ahead and implement an RFID based solution that is designed reasonably and provides a positive return to the company and, as a whole, the RFID industry will be better for it. RFID is a wonderful technology. It can provide great functionality. It can offer unique solutions that no other technology can provide. Be judicious in your use of it and you will be rewarded with successful implementations and all that accompany such successes. Blindly apply it, as a miraclelike solution to all problems and failure will be an all to familiar term. As in all technical projects striving for the simplest approach and the least amount of effort that provides the greatest return will serve you well. We at Acsis wish you great success in all your RFID endeavors.

RFID Integration

Page 13