Sep 02, 2002Sept. 2, 2002 - It was 1998 when Patricia Mackey realized she had a problem she could no longer ignore. The librarian for Rockefeller University Library was having increasing problems tracking the books and periodicals in her collection. The research university in upper Manhattan is home to post-doctoral students working on cutting edge research. They needed access to the biomedical research collection 24 hours a day.
Budgetary cutbacks meant that Mackey couldn’t keep staff working past 5 PM, so students were entrusted to check books out themselves. The system just wasn’t working. Students often took books at night without checking them out properly. Sometimes items wouldn’t be returned for months, even years, and no one knew who had them.
The solution to Mackey’s problem came in the form of a tiny microchip attached to an antenna – a simple RFID tag. Checkpoint Systems, the Thorofare, N.J., maker of RFID systems for retail applications, was working on a system designed specifically for libraries. Checkpoint was looking to test the system in a real environment. Mackey decided to be the proverbial guinea pig.
The library has tagged some 100,000 items in its collection with 2-inch by 2-inch RFID labels and installed a self-checkout station and RFID readers at the door. Teachers and students at the university all have photo IDs with either bar codes or magnetic stripes. When someone wants to check out a book, they simple swipe or scan their card, and place the book on a platform reader. The reader can scan a dozen books at once. The data is routed to a server that runs the circulation system, and the books are registered as checked out.
Longer-range RFID readers have been placed along side the library exit. When someone tries to leave with a book, the reader picks up the RFID tag number and checks it against a list of books that have been checked out. If there is a match, the person can leave. If there is no match, a short alarm sounds to remind the person to check out the book.
The library could strengthen the self-checkout system by having the doors remain locked when someone tries to leave with a book that hasn’t been checked out. But it didn’t want to use such draconian measures. Overall, the system has improved compliance just by making it easier for students to check books out themselves. The number of people doing so has increased sevenfold since the system went into place in 2000, and the library also knows which books are missing, since the tags are read at the door.
"Our biggest concern was knowing where the books and periodicals are so that professors and students would have access to information when they need it," Mackey says. "From that standpoint, the system is working very well."
Dozens of other libraries have begun introducing RFID for inventory, circulation and security purpose. In addition to Checkpoint, 3M Library Systems offers an RFID tracking system using smart labels from Texas Instruments. And French RFID systems provider Tagsys has teamed up with VTLS, an automated library system vendor, to offer a product for libraries. But the systems don’t come cheap, and librarians have to consider the potential benefits carefully.
The tags come in a label form, so they can be stuck on inside the front or back cover of a book or periodical, or on the side of a video or audiocassette. The microchip in the tag can usually store 96 bits or more of data. The Checkpoint tags are written to once at the librarians’ station. 3M’s tags can be written to and then overwritten a number of times. Typically, the library calls up the bar code number already assigned to a book and then burns that into the RFID tag. The library may also include additional data, such as a code that identifies which branch the book belongs to.
The price of the tags is depends on volume purchased. You may pay $1 or more for 20,000 tags, or 50 cents for a million or more. Generally speaking, libraries pay about 85 cents for a 96-bit tag. Tags that can store more data and be written to over and over can cost $1.50 or more. These tags are useful for some special applications, such as storing information about the condition of rare books.
In addition to the tags, libraries need several different kinds of readers, including:
--Circulation desk reader: $1,500 to $2,500 per unit.
--Portable reader: $1,500 each.
--Book drop reader: $1,500–2,500
--Long-range exit-door reader: $3,000 each.
--Self-checkout station: $20,000 to $25,000
As significant as these costs are, they are not the only ones the library has to consider. The Rockefeller University Library had to call in extra staff to tag each book and periodical. The process took four weeks. The librarians had to be trained on how to use the system. That took just one day and was done by the vendor.
But the university’s information technology staff had to work with experts from Checkpoint to convert the existing circulation and inventory systems from bar codes to RFID. The process was fairly straightforward because unlike retail or supply chain applications, the RFID tags were used exactly like bar codes. Still, an application interface has to be written, and that has to be done either by inhouse staff or outside programmers.
The Rockefeller University Library was able to run the RFID system off of existing servers. But libraries without access to larger IT systems will have to purchase a dedicated server to handle data from the readers. Depending on the number of readers and tags, a server could run from $10,000 to $15,000.
The total cost of the system will depend on each library's need, the vendor chosen and the number of objects that need to be tagged. The U.S. Public Library Association puts the cost of a complete RFID system for a small library with 40,000 items at least $70,000. A library with 100,000 items would have to spend at least $165,000 for a comprehensive system that includes a patron self-checkout station and a book drop reader.
One of the big questions about RFID technology is whether it will provide a return on investment. That is, will the libraries save more money than they invest in the technology? It is a very difficult question to answer because it depends heavily on how the library operates. What is clear is that the libraries will be far better run if they adopt the technology.
Savings from using RFID technology comes primarily from two areas: a reduction in payroll expenses and fewer lost or stolen books. How much a library saves on payroll depends on how it operates today. Some libraries pay overtime or even hire temporary workers to come in once a year to take inventory. They will likely save on those expenses because staff can use a handheld reader to take inventory in a small library in a couple of hours. However, many libraries take inventory only every few years, so they won’t see any savings in this area (though they will be able to provide better service to patrons).
Self-checkout scanners can help reduce payroll costs. Libraries may need to have only one person on duty during slow periods because patrons can take care of themselves. And since RFID scanners at the circulation desk are much faster than bar code scanners because the book doesn’t have to be oriented a certain way to be read, fewer staff can handle more people. But the savings may not be significant. That’s because librarians that are freed up from circulation desk duty may be deployed to take on other duties, such as providing more personal service to patrons. But clearly lines will be shorter and the overall library experience will be better for patrons.
The cost of RFID systems has been dropping. The industry, however, faces a chicken-and-egg dilemma. Until libraries are using millions of tags, the price of the tags will remain fairly high. And libraries won’t tag millions of items until the price comes down. But help could be on the way.
Alien Technology has been developing a way to create RFID tags that cost less than 10 cents once it reaches full production sometime in 2005 . Alien is targeting the retail and supply chain markets. But there is no reason the same technology cannot be applied to library systems. And since Alien’s tags will carry electronic product codes – a successor to the bar code proposed by the Auto-ID Center - bookstores as well as libraries could use the tags. That means publishers would likely be willing to put the tags in the books when they are manufactured. Libraries would only have to invest in the readers and IT infrastructure.
The Alien tags will be on the market next year, but it will be a few years before the technology is widely available. It’s unlikely that libraries will be using electronic product codes to track books before 2005 at the earliest. However, if vendors like 3M and Checkpoint make their technology compatible with the Auto-ID Center’s emerging system, then a very attractive migration path opens for libraries. They could tag some of their most expensive books, or those in their reference collections, and install a few readers. As the price comes down, they could then tag their entire collection and add self-checkout readers.
Checkpoint is a member of the Auto-ID Center. The company says that it will evaluate its position once the EPC is adopted. That won’t happen until at least October 2003, so until then, libraries will be faced with the tricky question of whether to make the substantial investment in a proprietary system, or hold off for a few years until standards are adopted.