Management of RFID in Libraries

By Karen Coyle

Preprint version of article published in the Journal of Academic Librarianship, v. 31, n. 5, pp. 486-489

Much has already been written about the use of RFID in libraries, such as the excellent article by Laura Smart [1] , and the works by Richard W. Boss [2] [3] for the American Library Association. These articles give details on the use of RFID in libraries and the products that are available. I won’t repeat those details in this article, nor will I cover the privacy issues (in libraries and for consumers in general), which are well-covered elsewhere. [4] [5] [6] Instead, I wish to spend some time contemplating the management questions that RFID surface for libraries that are considering this technology.

Briefly, the RF in RFID stands for "radio frequency"; the "ID" means "identifier." The tag itself consists of a computer chip and an antenna, often printed on paper or some other flexible medium. The shortest metaphor is that RFID is like a barcode but is read with an electro-magnetic field rather than by a laser beam. The similarity ends there. RFID is an advanced technology compared to barcodes. The RFID tag does not have to be visible to be read; instead, it can be read even when it is embedded in an item, such as in the cardboard cover of a book or the packaging of a product. It can also carry a more complex message than a barcode, which is limited to an identification number. The chip that is part of the RFID tag can carry many bytes of information, which means that it has the potential to carry not only the item number used by a library but also information such as the title of the book and/or its call number. The size of the information payload of RFID chips is one of the features that will undoubtedly expand as future technology advances allow the creation of smaller and more powerful chips.

A key thing to understand about RFID is that it isn’t a single technology; there are hundreds of different RFID products on the market today, and new ones appearing constantly. There are the RFID tags that are used for automated toll-taking for cars that can be read from many feet away as cars speed along highways. There are those that are in the card keys that many of us use to gain entry to our office buildings by swiping the card within a few inches of a pad by the office door. There are chips that are used to track animals on farms or identify lost pets, and others that help warehouses manage the inventory of pallets of goods. [7] The Food Drug Administration is considering the use of RFID to identify drugs and prevent counterfeiting,[8] and there may be a use for RFID in DVD’s to prevent movie piracy.[9] These are all very different technologies that work on the same principle. What varies is the amount of information the tag carries, the range in which it can be read, the frequency of its radio waves, its physical size, and of course its cost. The tags used in libraries today are among the lower priced tags, with short read ranges and limited functionality, yet even within a single library the technology can vary based on the need at that particular station. For example, where RFID is used to read shelves a narrow range is needed so that the reader doesn’t pick up items on shelves above or below the one being read; yet a circulation check-out station will be designed to handle a stack of books at a single read.

As this is a rapidly developing technology, it is not possible to say what capabilities may be available in one, two, or five years, much less further out. There have been significant advancements of this technology in just a few years, with no sign that limits of its potential have been reached.

Should Libraries Use RFID?

Because of the privacy issues, some librarians and library users question whether libraries should consider using RFID at all. While we can ask this question today, we may be facing RFID in our future regardless, especially if RFID becomes the successor technology to barcodes. Should barcodes and barcode readers go the way of vinyl records and turntables, libraries needing new or replacement technology will have little choice but to purchase RFID-based systems.[10] [11] Because of this possibility, we cannot afford to ignore this new technology, even if we do not embrace it today.

In considering the introduction of any technology into the library we need to ask ourselves "why?" What is the motivation for libraries to embrace new technologies? The answer to this question may be fairly simple: libraries use new technologies because the conditions in the general environment that led to the development of the technology are also the conditions in which the library operates. In the case of RFID, anyone managing an inventory of physical objects needs to do item-level functions, such as sales or lending, more efficiently and with less human intervention.

RFID is a highly advantageous technology for a wide variety of inventory tracking situations. It is also coming into its own for payment systems, including the ever elusive "micro-payment," the holy grail of non-cash transactions. Whether or not libraries embrace RFID, it will probably continue to replace barcodes in the retail supply chain. And it will contribute to the general speeding up of our world, which affects libraries as well as other institutions. A key fact is that library circulation, the primary function where RFID can be used, is increasing while library budgets and purchasing power are losing ground.

RFID and Library Functions

I have likened RFID to the barcode, which is an apt analogy. As an identifier, it is particularly suited to inventory functions, and a library has a strong inventory component. There is, however, a key difference to the library’s inventory as compared to that of a warehouse or retail outlet. In the warehouse and retail supply chain, goods come in, and then they leave. Only occasionally do they return. The retail sector is looking at RFID as a "throw-away" technology that gets an item to a customer and then is discarded. Yet the per item cost of including an RFID tag is much more than the cost of printing a barcode on a package. In libraries, items are taken out and returned many times. This makes the library function an even better use of RFID than in retail because the same RFID tag is re-used many times.

Second only to circulation, libraries look to RFID as a security mechanism. [12] [13] The RFID tags can facilitate security in a variety of ways. In one method, the tag that is used has a special "security bit" that can be switched from "checked-in" to "checked-out." The exit gates at the library read each tag as the user passes out of the library and sounds an alarm if the bit is not in the "checked-out" state. The check-in function resets the bit. Another method is for the tags themselves to remain the same; as the user passes through the exit gate the system reads the tags in the books in the user’s arms or bag and queries the library database to be sure that the items have been checked out.

Although RFID can be used in library anti-theft systems, this doesn’t mean that it is a highly secure technology. What libraries don’t tell their users, and none of us should probably say very loudly, is that RFID tags can be shielded by a thick layer of Mylar, a few sheets of aluminum foil, or even an aluminum gum wrapper, so they won’t be detected by the reading device. In addition, today’s tags are not hidden in the spine of the book, like security tape, but are often found on the inside of the book cover, barely concealed by a library label, and can be removed. This is not a condemnation of the technology nor even a reason not to use it in the library security system; the reality is that library security has never provided more than a modicum of security for library items. The gates and their alarms are as much social deterrent as they are actual prevention. The reason to use RFID for security is not because it is especially good for it, but because it is no worse than other security technologies. There is, however, some potential savings because a single tag serves many different functions. The library saves some time in processing new items because it only has to affix one technology to the item. It may also save some money due to the integration of circulation and security with a single vendor and into a single system. Some future-positive thinkers in the library world see the potential to have a combined exit-gate/check-out station that allows patrons to walk about of the library with their books in hand and their library card in their pocket. That brings up other questions, especially privacy ones, but the notion is intriguing.

As well as being an inventory technology, barcodes also serve the point of sale (or lending). The need to have a direct line of sight on the barcode makes it difficult, however, to perform functions on more than one item at a time. RFID systems can read multiple tags at once, allowing you to check out a stack of books with a single transaction. Barcodes also have some disadvantages when taking an inventory of the library. The line of sight requirement means that each book must be tipped out far enough to read the barcode if it is on an outside cover, or removed entirely from the shelf if the book or item must be opened to see the barcode. This is an area where RFID can provide great advantages because the tags can be read while the books sit on the shelf. Not only does the cost of doing an inventory of the library go down, the odds of actually completing regular inventories goes up. This is one of those areas where a new technology will allow the library to do more rather than just doing the same functions with greater efficiency. Library experience with RFID is still in its early stages, but already some librarians are getting ideas for additional uses of this technology. RFID could be used to gather statistics on the re-shelving of books in the stacks area, by equipping shelvers with hand-held readers. Vendors of RFID systems for libraries are already offering automated sorting of returned books into a handful of bins that facilitate the re-shelving of books that are checked in. A fully automated library could potentially know exactly where an item is, down to the very book truck or bin, during the return process. In theory, a library could "know" when a book leaves the shelf, and could trace the progress of the book through the library to check-out. In reality, it is already possible to find a requested video in a jumbled browsing section that gets out of order due to high use.

Justification and Return on Investment

In the commercial world, all things are measured by "return on investment" or ROI. A company invests in a new technology that speeds up its delivery of widgets and cuts costs warehousing costs. The cost of the new technology is compared to the increase in profits, and in the best of cases profits rise enough to both pay for the new technology and then some. This is the return on investment. When libraries measure their success, profit isn’t part of the equation. Like other institutions that provide services, such as schools and city governments, the bottom line is less quantitative than the business case. Libraries "spend" their ROI on new services or on beefing up existing services. They also spend their ROI to respond to budget cuts or to loss of buying power when budgets do not keep up with inflation. This makes it hard to demonstrate that an investment in technology is worth the cost.

Laura Smart’s Library Journal article on RFID [14] lists fourteen areas of library operations where one might measure gains in time and materials costs. The obvious gains are in checking out and checking in books, although the gains vary by the degree of automation. Additional investment can be made on a book return system that automatically checks in items as they pass along a conveyor attached to a book drop. This system can be attached to an automated sorting machine that sorts the items into bins based on their call numbers. All of this saves time, but there is one thing to remember about this efficiency: the items still need to be shelved. The increase in circulation, which is often one of the few quantitative measures that libraries have to show that RFID has made their operations more efficient, has to be balanced against the cost of re-shelving more items.

It is generally agreed that the greatest return on investment depends on turning over the check-out functions to patrons [15], in essence practically eliminating the need for circulation staff. Some libraries intend to become 100% self-check-out. Others are content to allow patrons to choose between the staffed check-out desk (which should operate more quickly), and the self-check stations. There are arguments for and against these approaches. On one hand, for many people who frequent libraries the circulation desk staff is the only staff with whom that they have any interaction. Some librarians fear that self-check could eliminate what little human factor that libraries have for these patrons. On the other hand, the act of running patron cards and library items through a check-out station for hours at a time is mind-numbingly dull, and probably not the best use of staff time. It’s also the source of many staff repetitive stress injuries.

What the ROI statements that I have seen uniformly fail to take into account is user satisfaction. In a service business like that of the library, satisfying your users is one of the few measures of success that you have. You can intuit satisfaction from an increase in use statistics, but an actual survey of users, ideally both before and after a change is made in library operations, would be the best evidence the library has that it is fulfilling its mission. Self-checkout could be seen by users as a mere shifting of the burden of check-out from the library to the users themselves, who will feel that they are being asked to do the work of the library. Or it could become the "ATM" of the library world, the fast service with few lines that users come to prefer to old way of doing things. Just adding self-check stations will not be enough; libraries must be sure that the stations serve the needs of the users and perhaps even provide additional services to win over their hearts.

Some Problems Remain

It seems clear, at this moment in time, that RFID will become a widespread technology, replacing barcoding for a variety of industries. It also is already leading to entirely new functionality, such as the "touchless" payment systems for debit cards and highway toll payments. But this is a very new technology and there are "gotchas" and limitations that we need to recognize. For libraries, RFID is very promising for items with a certain physical "substance," such as books and media in cases and boxes. But as we move to less sturdy items, RFID tags pose problems. Libraries must decide if they need to attach RFID tags to magazines, pamphlets, sheet music, and a host of other items that may not have a good location for a somewhat bulky two-inch square tag, and that are so numerous that the tag cost is significant. If it is determined that RFID will not work for these items, then an alternative check-out system needs to be maintained, which has real costs as well as a certain nuisance factor both for patrons and the library. It also may not be possible to accurately check out a stack of items that are particularly thin, such as journals or even children’s books. The RFID tags on these materials can be so close together physically that the signals essentially cancel each other out.

Items with odd shapes and metal components, such as CDs and DVDs, are stretching the creativity of vendors of RFID systems for libraries. RFID tags for optical discs have been developed that fit around the hole of the disc. This requires a reduction in the size of the antenna and thus reduces the read distance of the tag. But discs with an extra thick metal layer can still block the reading of the tags.

Another concern for libraries is that there are two main directions for the RFID industry at this time: one is toward the use of tags in the retail chain, the other is toward higher end tags for payment systems. In the retail environment, tags are only needed up to the point that the item is sold. This means that there will be some emphasis on inexpensive and not entirely durable tags, and ones that can be written at the time of manufacture. For the payment systems, tags carry more information but are more durable and more secure. Libraries need a slightly different technology from either of these. Libraries need tags that are durable, since they be placed on an item that will be used repeatedly. The library function will probably have the longest-lived tags of any other sector, since books can remain on shelves and circulate over decades, while retail products have a short shelf life and even debit cards rarely are issued for more than four or five years. Libraries may also need tags that, although inexpensive, can be reprogrammed and may even need to have more than one "lifetime," for example in the case of an item that is moved into storage or is sent to another library through interlibrary loan. These functions may be available, but perhaps not at the rock-bottom prices that libraries want to pay for the tags and the related technology. [16]

Conclusion

Whether your library is using RFID today, is thinking of using it in the future, or has pre-determined that RFID is not suitable for libraries, we cannot ignore this technology. It is going to be incorporated into products that libraries purchase and into items that users bring into the library, such as smart cards and hand-held electronics. It is being considered for passports and is already in use as a payment system. Now is the time to develop both policies, such as the American Library Association’s statement on RFID in libraries, as well as sets of best practices that give the library some clearly stated goals for the decisions that will inevitably have to be made as RFID becomes a common, if not ubiquitous, technology.[17]


[1]Smart, Laura "Making Sense of RFID," Netconnect (Fall, 2004): 4-14
[2] Boss, Richard W., "RFID Technology for Libraries" Library Technology Reports (Chicago: American Library Association, 2003) 58 p.
[3] Boss, Richard W. "RFID Technology for Libraries" ALA Tech Notes (Prepared May 14, 2004) Available: http://www.ala.org/PrinterTemplate.cfm?Section=technotes&Template=/ContentManagement/HTMLDisplay.cfm&ContentID=68138. (June 11, 2005).
[4] Molnar, David and David Wagner. "Privacy and Security in Library RFID: Issues, Practices, and Architectures" in: ACM Workshop on Visualization and Data Mining for Computer Security (Washington, D.C.), October 25-29, 2004. (Association of Computing Machinery, 2004) Pp. 210-219.
[5] Electronic Frontier Foundation, RFID. Available: http://www.eff.org/Privacy/Surveillance/RFID/ (June 13, 2005)
[6] Federal Trade Commission. Radio Frequency Identification: Applications and Implications for Consumers. March, 2005. Available: http://www.ftc.gov/os/2005/03/050308rfidrpt.pdf (June 11, 2005)
[7] Want, Roy, "A Key to Automating Everything," Scientific American (January, 2004): 56-65.
[8] U.S. Food and Drug Administration, Radiofrequency Identification Feasibility Studies and Pilot Programs for Drugs; Guidance for FDA Staff and Industry. November, 2004. Available: http://www.fda.gov/oc/initiatives/counterfeit/RFID_cpg.html. (June 11, 2005)
[9] O'Conner, Mary Catherine, "Group Studies RFID to Stop Digital Piracy" RFID Journal (May 12, 2005) Available: http://www.rfidjournal.com/article/articleview/1/1/ (June 13, 2005)
[10] Margulius, David L., "The Rush to RFID" Infoworld (April 12, 2004) pp. 36-41.
[11] Manhattan Associates, Inc. RFID: The UPC of the 21st Century; Small Technology, Enormous Impact. (2003) 10 p.
[12] Infopeople, California Library RFID Survey. (August-September, 2004) available: http://www.infopeople.org/resources/RFID_survey/ (June 13, 2005)
[13] In his 2003 Library Technology Report (Boss, 2004, p. 44), Richard Boss reported that at the time 3M Corporation was not offering RFID exit sensors since the company felt that RFID was not a suitable replacement for security strips (which that company, not coincidentally, produces). Since that time 3M has added RFID security to its product.
[14] Smart, 2004
[15] "Primary Research Group has Released a New Study: Best Practices of Public Library Information Technology Directors" eMediaWire (March 14, 2005) Available: http://www.emediawire.com/releases/2005/3/emw216328.htm (June 13, 2005)
[16] Although the greatest cost in the purchase of an RFID system today is still in the tags, there is also significant investment in workstations that write and read the tags, as well as auxiliary functions like sorting and inventory. See Boss, 2004, for some discussion of system costs.
[17] Some RFID in library resources: The "RFID in Libraries" blog, available: http://www.libraryrfid.net/wordpress. Also, there is a "libraries" section of the RFID Gazette blog, available: http://www.rfidgazette.org/libraries/. You can subscribe to the RFID in Libraries discussion list at http://slisweb.sjsu.edu/ecommunication/listsubscriptions.html.


The copyright in this article is NOT held by the author. For copyright-related permissions, contact Elsevier Inc.