Performing an Electronic Inventory using a PDA

Citation

Bennett, E. (2003), "Performing an Electronic Inventory using a PDA", Library Hi Tech News, Vol. 20 No. 6. https://doi.org/10.1108/lhtn.2003.23920faf.002

Performing an Electronic Inventory using a PDA

Erica Bennett

Located in North Orange County, California, Cypress College is a community college and administers to the academic, occupational, and special interest needs of over 16,000 commuter students. Four full-time librarians, four adjunct librarians, and six paraprofessionals staff the library.

During the summer of 2002, Cypress College Library performed a successful electronic inventory of the nearly 70,000-item collection using a personal digital assistant (PDA), several interactive programs, and a MicroSoft Access query and report. The software was designed to work in conjunction with the Endeavor Voyager integrated library management system circulation module and its Oracle database. Except for the cost of two PDAs, we acquired the software free of charge. This exciting new inventory system validated the integrity of our online catalog, and as a consequence, improved professional services to our library's diverse population of community college users. We hope that other libraries may benefit from our experience, perhaps even modifying our results to work within the requirements of their own system.

In the "olden days" two staff inventoried the collection, one classified and one student hourly personnel. It was a summer project, in the days when the community college closed for a month in the summer. The classified staff read from the shelf list, and the student hourly checked the item on the shelf. This procedure covered the entire collection over a three-year period. Missing items were searched for a period of two years before being withdrawn from the collection. Inventory was not considered an essential function of Technical Services, because the rates of the missing always stemmed at around 0.5 percent per year. Owing to a staff transfer, manual inventory effectively ceased in 1980. The library migrated to PALS, Minnesota's Project for Automated Library Services, in 1985 and by the early 1990s the library decided to give up its shelf list. However, not before each item was checked off the card and each card was compared to the computer record. One staff member accomplished this effective inventory process over a period of three years.

In 1997-1998 an infusion of the California State Funded Tele-communication and Technology Infrastructure Program (TTIP) funds brought about a significant change at the Cypress College Library. With its consortium of four local community college libraries, Cypress College decided to move to a Web-based library information system and sent out Request for Proposals to various vendors. The Endeavor Voyager system was ultimately chosen. Migration to Voyager occurred during the Winter session of 2000. Unfortunately, Endeavor had not developed an inventory module. However, at VUGM, the Voyager Users Group Meeting, in April 2000, the Cypress College staff witnessed Paul Asay's and Jeremy Shelhase's Observer Scan and Control software demonstration[1]. Their programs were designed using Visual Basic, WinBatch, and Satellite Forms SE[2]. The Asay and Shelhase presentation described the application in four parts:

1. 1.

   Observer Scan software loaded on a Symbol SPT 1500 or a heavy-duty 1700 Palm, both with an attached barcode scanner, allowing the user to scan and store up to 1000 barcodes.

2. 2.

   Next, setting the Palm in its cradle attached to a desktop computer, the Observer Control and Hotsync program pulls the data from the Palm, creating an optional database (see Figure 1).

3. 3.

   Running Observer Load and Voyager passes the barcodes automatically through the Endeavor Voyager Circulation module, changing the item records "item_status_type" to "discharged."

4. 4.

   Utilizing a MSAccess query and report, list the item's browse and usage statistics.

Figure 1 A sample of XMLMARC-lite tagged bibliographic data

The Cypress College Library staff returned from VUGM determined to find the money to purchase two Symbol SPT 1700 Palms; which are heavy-duty enough to take into account the fact that students would be performing the bulk of the scanning. Monica Doman, Systems Librarian, successfully acquired a portion of the campus' State Instructional Equipment/Block Grant funds to purchase the Palms, supplemented by TTIP funds, received from the state for the purpose of improving library automation. In April 2001 the library purchased the two 1700 Palms at $1,025.45 each. The package included 2 MB RAM, two single-shot cradles, line cords, power supplies, slip-cases, lithium batteries, and organization and desktop software.

After installing the software and linking the tables, the Systems Librarian turned the project over to Circulation. However, it was not until December 2001 that I began testing the software and reports only to discover that the browse and usage statistics produced were not the inventory module that we had imagined. I realized that while I was able to modify canned queries and reports, as well as design simple bibliographic and statistical queries and reports, I was nowhere near proficient enough to create the sort of query that we would need to perform an inventory of our collection. Not willing to give up, however, I sought out the Endeavor Support Knowledge Base, the Web site supporting Endeavor users, and performed the obligatory keyword search for "inventory" and "observer." I immediately located Christine Nelson's and Michael Eason's Observer Inventory database documents that consisted of an excellent PowerPoint presentation, database, and a Readme file that included detailed instructions about how to link tables and import the original database created by Asay and Shelhase's Observer software[3].

Nelson and Eason wrote a complex query that compared items within a specific LC call number range that were scanned or on the shelf, to what should be on the shelf in that area according to the database by item type (i.e. book, CD, video, etc.) and location; looking only at items that had the status of "not charged," i.e. ignoring items that were charged (checked out) or discharged (recently checked in), and listing those items that appeared to be missing. While this query did not perform a shelf-reading function, it gave us the ability to inventory our collection, a task that had not been accomplished at the Cypress College Library since the mid-1990s. In May 2002, I began testing our collection.

I started the project in the media collection. I chose to perform the original scanning tests myself, so that I would be in a better position to train my students. I learned that scanning the barcodes at an angle slightly below the barcode, pointing up, achieved a faster read. I scanned the A and B classifications, pulling items from the shelf to create a false-positive result, and I discovered that the system worked! The resulting report that was generated did list items that were missing from the shelves (see Figure 2). I realized that the database did not care about the order in which the items were scanned, and was looking past the items with an item_status_description of "charged" or "discharged." This became clear when I had noticed that items on my video rough-sort truck that had recently been checked in did not appear on my missing report. Consequently, I understood that it would not matter if the collection were shelf read first. I did make shelf reading a part of my procedure, however, for the main reason that when I actually went looking for missing items, I was more confident that they might be found on the shelf.

Figure 2 Part of typical catalog page rendered with an XSL stylesheet

Then something strange happened. The system locked me out, refusing to run the report. What I soon realized is that in the time it took for me to scan a section of videos, upload and discharge the barcodes, and run the report, a patron had selected items from that section of the stack and checked them out. Cindy Ristow, Circulation Supervisor, decided that I would let the Circulation staff know the section I was working on and whoever was running the Circulation desk would manually check out those items if they appeared before I moved to a different section.

The actual scanning and reporting of the main collection went fairly smoothly. While I did inventory the media collection and various classification sections in the main book collection, my staff of between five and nine work study students and library clerks performed the majority of the barcode scanning, and better yet, they enjoyed using the technology. I also discovered that my computer needed more memory, when I was locked out upload after upload. I realized that I was trying to bring in too many barcodes at a time. Closing my open windows helped, however I decided not to shoot for the maximum of 1,000, stopping between 500 and 800. This decision resulted in tying up my computer for shorter periods of time while the barcodes discharged. This decision also kept my students out in the stacks for periods between half an hour and one hour at a time, which turned out to be important in alleviating boredom, the physical stress, and the mistakes that naturally occurred as a consequence of trying to work through the collection too quickly. I found that when I was too ambitious, the students made mistakes. Sections were accidentally skipped over, which necessitated them being rescanned. "Remember, accuracy is more important than speed" became my mantra. Between my day shift and evening shift of students, we were able to perform a maximum of four 500-800 item sections a day during the periods when I actually found the time amid my other duties to work on the inventory project. The process ran throughout the summer, ending with reference in September of 2002.

In the reference collection I ran into a new challenge. Our main collection is double-barcoded, which allowed one student to scan, simply tipping a book from its place on the shelf and fixing the scanner on it. With the reference collection, however, I found only one barcode, and it was affixed along the crease on the bottom of the title page. This event necessitated me using two students, one to pull the book from the shelf and open it, and the other to use the PDA to scan the barcode. My first section in reference was again the A and B classifications. This time my report listed every book as missing. What caused the query to create a shelf list? Discussing my dilemma with the Circulation Supervisor, we realized that the items in reference could not achieve a status of discharged, due to the fact that they had been given a non-circulating status in the system administration module. The Circulation Supervisor realized that she could change the status of reference items by allowing one type of patron charge privileges. She selected a faculty patron group from a distant consortia member, whom we knew would not likely come into the library expecting to check out reference books during the time that I would be inventorying that section. And again, it worked! This procedure was electrifying in its success.

Armed with a four-inch binder filled with reports, my next step was to actually look for the missing items. Cindy determined that I would search for the items for a period of one year before sending the list to Technical Services for processing. We selected nine locations for our search:

1. 1.

   OPAC, to determine if the item had been removed from the database or had been recently checked out;

2. 2.

   manual check-out cards;

3. 3.

   stacks;

4. 4.

   mending;

5. 5.

   displays;

6. 6.

   new book trucks with librarians;

7. 7.

   new in technical services;

8. 8.

   reference, to determine if they had been misshelved; and finally

9. 9.

   after many problems had been resolved, I sent out an e-mail with an attached list of the missing to staff.

I found missing items in all places except for new in technical services and on the mending shelf (see Figure 3). I found books on the shelf with different barcodes than were reflected in the record. I found items that were correctly shelved in reference, however, their location in the record was incorrectly indicated as "stacks," and vice versa. I found duplicate records; one item in reference and one in the main collection, where we really had only one copy. Shifting a section of periodicals during another project, I found a paperback stuck behind the stack. I found videos that had fallen under the shelving. I found items that should have been shelved in the HQs that were incorrectly shelved in the HVs, same numerical location that had been missed during the shelf reading process. I found out that I had an aptitude for problem solving and investigative work, and was thrilled when other libraries in our consortium called to find out what we were doing.

Figure 3 XML OPAC - a TextML documentbase search application

I have successfully shared our inventory system with two other community colleges in our consortium, and with the Los Angeles Chapter of the American Society for Information Science and Technology. This process of reflection has brought some important conclusions to bear. The two teams of Asay and Shelhase and Nelson and Eason not only designed software that works, they provided it free of charge. Consequently, we were able to take an arduous three-year process and condense it down to less than five months. Communication between library departments was key to our success. It was also Monica Doman's determination to find the money to purchase the Palms that made the process possible. However, it was my library administration's belief that I could make the system work that made it an ultimate success. I would never have had the opportunity to develop this inventory system if I had not been afforded the time for research and experimentation. Time is a valuable resource, and hard to find in the face of the demands of front-line circulation services. However, I might never have deciphered Nelson and Eason's documentation, linked the required tables, comprehended their query, and tested the system combined with Asay and Shelhase's original software, if my colleagues at the Cypress College Library had not given me the time to make it happen. Coming at the end of my first-year as a candidate for a Master's of Library & Information Science at UCLA, this is a major lesson. The Cypress College Library is a bare-bones operation that understands that innovation does not have to suffer when the need is great and the money scarcer.

NOTES1. Paul Asay is the Systems Programmer at Indiana State University. Jeremy Shelhase was Systems Librarian at Indiana State at the time that the programs were written. He currently holds the same position at Humboldt State University in California.2. As of May 2003, the original PowerPoint presentation is located at http://paulasay.indstate.edu/observer/vugm2000/index.htm or can be linked to from the Observer homepage at http://paulasay.indstate.edu/observer/ Systems requirements, instructions for use, and the tables requiring linking can be found at the Observer Scan & Control Documentation Web site http://paulasay.indstate.edu/observer/observerdoc.htm.3. Christine Nelson is Technical Services Librarian, and Michael Eason is IT Specialist II at Everett Community College in Washington State.

Erica Bennett(ebennett@cypresscollege.edu) is currently a Library Assistant II at the Cypress College Library, Cypress, California, USA.