AUTOMATED BOOK ORDER AND CIRCULATION CONTROL PROCEDURES AT THE OAKLAND UNIVERSITY LIBRARY Lawrence AULD: Oakland University, Rochester, Michigan 93 Automated systems of book order and circulation control using an IBM 1620 Computer are described as developed at Oakland University. Rela- tive degrees of success and failure are discussed briefly. INTRODUCTION Oakland University, affiliated with Michigan State ·university and founded in 1957, offers degree programs at the bachelor's and master's levels. By September, 1967, 3,896 students were enrolled and continuing growth is anticipated in coming years. The library had holdings of 86,755 Jlumes and 17,908 units of microform materials on July 1, 1967. Although young, Oakland's library has already encountered a host of problems common to most academic libraries. In recognizing a need to 1utomate or otherwise improve basic routines of handling book ordering •• U. circulation control, Oakland is simply another member of a growing club. The book order system developed at Oakland is noteworthy because of ·~rtain features which may be unique: a title index to the on-order file, a computer prepared invoice-voucher form, and a computer prepared voucher card which serves as input to the computer for writing payment checks. In logic the system is related, through parallel invention, to the Machine Aided Technical Processing System developed at Yale Univer- sity ( 1). The system developed with unit record equipment at the Uni- versity of Maryland is perhaps more directly related, particularly in the use of the purchase order as a vendor's report form (2,3 ). The Pennsyl- 94 Journal of Library Automation Vol. 1/ 2 June, 1968 vania State University Library design for automated acquisitions, which uses a similar purchase order, includes the capacity for an elaborate and variable method for reporting the progress of each item from initial order to completion of cataloging ( 4,5) . The IBM 357 circulation control system developed at Southern Illinois University, Carbondale, set the pattern followed by most subsequent sys- tems ( 6,7) . Oakland's circulation control system, a variation of the IBM 357 system, is more flexible than some because it uses trigger cards to control machine operations. This paper, originally distributed to a relatively small group of persons and redrafted for a more general reading, presents a case study of how one institution in modest circumstances set about solving certain problems. It describes not systems to be copied but rather a learning process which will continue for many years to come. BACKGROUND During the winter of 1964/ 65, Oakland University Library laid out the plans and began work on a program of automation of the University Li- brary. An initial four-phase plan was conceived: 1) book order, 2) circu- lation control, 3) serials acquisitions, and 4) a printed book catalog. These housekeeping routines were felt to be the foundation for develop- ing further automation in the library. Their automation would liberate the staff, clerical and professional, from such nonproductive and repetitive_ tasks as alphabetizing and re-copying of bibliographic information. An early decision to learn by doing rather than attempting to design the ulti- mate system in advance was supported by the University administration. Consensus being that a larger computer to replace the IBM 1620 would be delivered within two years, computer programs were planned to be useful for twenty-four to thirty-six months. Work on developing the book order system was begun in March, 1965; perhaps an all-time speed record was achieved when the system was put into use on July 1 of the same year. Work on a circulation control sys- tem was begun in August and on February 21, 1966, it too was ready. Phases three and four, serials acquisitions and the printed book catalog, were by then being held in abeyance until larger computer equipment should become available to the library. At Oakland University all computer and related services are pro- vided by the Computing and Data Processing Center. The computer sys- tem includes the following pieces of equipment: IBM 1620 Computer, 40K with Monitor 1 and additional instructions feature (MF, TNF, TNS) IBM 1622 card reader/ punch (240 cpm/ 125 cpm) Two IBM 1311 disk drives with changeable disk packs IBM 1443 line printer ( 240 lpm) Automated Book Order/ AULD 95 Only one of the two disk drives is available for production use because the other is committed to monitor, supervisor, and stored programs. A disk pack on the IBM 1620 can accommodate two million numeric or one million alphabetic characters. The computer language used for most of the library programs is 1620 SPS (Symbolic Programming System); Fortran is used for some computational work. Equipment within the Library consists of an IBM 026 printing keypunch which is used for the order system and an IBM 357 data collection device, including a time clock, with output via a second IBM 026 printing key- punch for the circulation system. BOOK ORDER PROCEDURE As may be inferred from a birdseye view of the order system (Figure 1), the initial input to the computer is decklets of punched cards. Output from the computer is a series of printouts: purchase orders, Library of Congress card orders, Oakland University invoice-vouchers, a complete Fig. 1. Flow Chart of Book Order System . 96 Journal of Library Automation Vol. 1/ 2 June, 1968 on-order listing with title and purchase order number indices, depart- mental listings, and budget summaries. Facu1ty and library staff submit requests for book purchases to the Acquisitions Department on a specially designed Library Book Request Form (Figure 2 ) . The 5x8-inch size provides adequate room for notes, checking marks, etc., and makes for improved legibility, which in turn makes for easier, faster, and more accurate keypunching. Kttt;e libt o ry Oo~larul Untve r1 ity Jildg. Q 11ery LIBRARY BOOK REQUEST Mutt be Typ41d &JP St orch Au th o, CIJ Tit I• P' TLA PIIP Brit . ... P~o~bli•h•r and "oce r----- No. Copie• I P, bll•h Dote l fd ition J•'· _t·· ~ Mo . Yr. Cotl. ltl!qu tttt d It D!portment Cited in r---o:;;- P'rice t ·· Dept I Vando• Clau l l C Cood N•mbe • l.C. Fig. 2. Book Order Request Form. The request form calls for the bibliographic data customarily required for book purchasing, plus date of ordering, code number for the depart- ment originating the order, and vendor number. Oakland University uti- lizes campus-wide a five-digit vendor code system; since the Library's vendor numbers are a part of the University's vendor code, this interface is one of several points where the book order system ties in with other University records and procedures. A tag number is assigned to each Library Book Request Form upon its arrival in the Acquisitions Department. After routine bibliographic identi- fication is completed, decklet cards (Figure 3) are keypunched. The in- dividual cards in each decklet are kept together by the tag number, punched into columns one through five. To keep the cards in order within decklets, column six is punched to identify the type of card as 1) author, 2) title, 3) place and publisher, or 4) miscellaneous information. Column seven indicates the card number within type of card. For exam- ple, code 11 in columns six and seven wou1d be the first author card and code 12 the second. ·- Automated Book Order/ AULD 97 I I : I •I l l I l ~~ I l I I ..... ~;,, I I I I I l l ! I I I : I .,! AUTHO• ~ ~ : 0 : I!! u ~ ••• C< ANC ~u&ll&~ biSct/11~ Cll~b:S . Each book has a machine readable book card (Figure 7). The period for which the book normally circulates is indicated with a letter code punched into column one; column two identifies the collection within the library from which the material came; column three identifies · the type of material. The call number and/ or other identifying information is punched into columns four through forty-one. Column forty-two is punched with an end-of-transmission code . 104 Journal of Library Automation Vol. 1/ 2 June, 1968 .. ::;; !!~8 g c: .. ~~z z ... . 0 p 0 -1 1111 .. !,._ ~oot ~ - .... i z:o =;.: "' !::o:r.oa 1);1 ... ;!,..< .. ... '"It"' !Ill 0 $ ;;;~_. ... ;., 0 c: S!! ~~1!!1 ,. Ill: z o:= lire:~ ... ~-~ -zr.oa s 0 n_ c: ..... :IIII I:' s; S20 z-;. z =~= 0 0-it . . c: ........ .. z !o~ ... < ... ~:~ .. "' ~ . "' PI 'I t' 1 • t •h • eijn!lf!M!i!u+•+•+•+•+•+•+•+•• .. Fig. 7 . Book Card. The IBM 357 data collection device will perform only one operation without special instructions. If it is to perform more than one operation, it. must receive instructions for each variant operation and it must receive them each time the variant operation is performed. This limitation can be met in one of three ways: by not admitting variant operations, by using a cartridge as a carrier for some information, or by providing special . instructions as they are needed via a "trigger" card. Denying the existence of a variant operation was not practical, because at Oakland the identification of a borrower constitutes a set of variant operations. The Library's clientele includes not only Oakland University students, faculty, and staff, but also residents from the surrounding com- munities, area high school students, and neighboring college students. The heaviest users are Oakland's own students and faculty, who have machine readable plastic identification cards issued by the Registrar or the Personnel Office. It has been impractical for the Library to attempt to issue similar cards to guest borrowers. Thus, the identification of a bor- rower is a set of variant operations. Use of a cartridge to gain the borrower identification number would be possible but would leave the borrower identification badge unused. This badge card constitutes an official identification card and as such should be utilized throughout the University whenever practical. - . · Trigger cards to instruct the 357 in the pedormance of variant opera- tions were developed to control the recording of borrower identifica- tion and to identify discharging and certain charging functions. The use of trigger cards provides flexibility, in that machine. instructions are carried in trigger cards and are not an integral part of the book cards. A change in machine configuration would probably not require ·repunch- ing book cards for the book collection. At the same time a wide range · of 357 machine functions are made possible through ·the use of different Automated Book Order/ AULD 105 trigger cards. In short, the adoption of trigger cards provides the greatest degree of flexibility in operating the 357. In the customary borrowing procedure the student brings a book to the circulation desk and presents it, along with his machine readable student ID card, to the desk attendant. The attendant first inserts the book card into the IBM 357 data collection device, then retrieves the book card and inserts a "student badge trigger card", which activates the badge reader on the 357. Then the badge is inserted into the badge reader, completing the transaction. By remote control this has created on an IBM 026 printing keypunch a card with the following information: typical loan period, collection from which the item came, type of material, call number, borrower type, borrower's identification number, the day of the year, and the time of day secured from an on-line clock. If the borrower does not have a machine readable badge card, an alternate method of charging a book is to use a "manual entry trigger card" which activates the manual entry unit, with which can be recorded numeric information identifvine: the borrower. With special trigger cards .bo;;ks can also be charged to reserve, bindery, or "missing". Books are discharged by passing the book card through the 357 and following it by a "discharge trigger card". Monday through Friday at closing the charge and discharge cards for the day are delivered to the Computing and Data Processing Center, where they are processed by the IBM 1620 computer system. The circulation file is maintained on a disk pack similar to that for the order. system. Three reports are received from the Computing and Data Processing Center: a daily cumulative listing of all books and materials in circula- tion (Figure 8); a cumulative weekly list of all books on long-term loan; and a weekly fines-due report. In addition, overdue notices, computer printed on mailable postcard stock, are sent weekly to the Library where they are audited before being mailed. The fines-due report is arranged by borrower, bringing together in one place all of the borrower's delinquen- cies; the books which he has neglected to return are listed here, as are the overdue books which he returned through the outdoor book return chute. For the latter the number of days overdue at the time of return is listed. Subsequent refinements introduced into this system include two addi- tional reports: a pre-notice report in call number sequence produced two days in advance of the fines-due report and a listing of books discharged each day. The pre-notice report makes it possible to search the shelves for books which have been returned but, because of time lag, may still have overdue notices generated. Normal tum-around time for the system is 24 hours, but on weekends it goes to 63 hours and at certain holiday periods even higher. The daily list of discharges documents the return and discharge of each book and is used to answer the student who says, "But I returned the book." 106 Journal of Library Automation Vol. 1/2 June, 1968 S HORT TERM BOOKS IN CIRCULATION . WEDS-JUl. 13.1966 PAG F. 1 8 CALL NUMB ER BORROWER DAY OF YR DUE ODUE 01 JC0153ol.79 01 000009B74 20B 01 JC 0179 , R723 01 000007736 209 01 J C0 179oR83-1962 01 000004838 199 01 J C0 179.R86-195 4 01 000007935 209 01 J C025 lol..27 01 000009021 20 I 01 JC0421oB8Vol 01 000000207 127 * 01 J C0 4 23oi..58Co2 04 000002393 19 9 01 JK0246oB9-1895V o 2 01 00000020 7 127 * 01 JK04 2 1oP4 01 000006266 203 01 JK0421o S7 01 000006266 209 01 JK0516oS3 01 000 003891 199 01 JK0518oH6 01 000006266 209 01 JK0524ol.38 01 000007717 2 1 4 01 JK154 1oJ27 01 000006266 182 * 01 JK1561o527 01 000003891 199 01 JK1 57 1oM8 01 000003640 208 01 J K1976 oM5-Co2 0 5 0 00002256 207 01 JK2295oM5253 01 000007397 209 01 JK2372 oH5 04 000002194 2 10 01 JK 2372oP6 04 000002194 2 1 0 01 JK2408oK4 0 1 000 00020 7 146 * 01 JN6769 oA5K622 01 00000 52 31 2 13 01 J01503 o1 912 oB7 01 000003824 209 01 J01503o1911oH72 01 000003824 207 0 1 J01512oK7 01 000 003824 207 01 J S0323oC58 01 0000 07717 209 01 JS0341oW7 0 1 00 0 00 7717 2 09 01 J X14 25 oP384 04 00000 2925 213 01 JX14 28 oC 6C5-1 964 01 000004154 199 01 JX1 977o2 oC5A73 01 000009 11 9 207 01 JX1977o2oU5577 0 1 000007371 201 Fig. 8. Example of Short Term Circulation Rep01t. Maximum file capacity will permit up to about 9,000 charges at one time. Assuming an average life of four weeks for each charge, the maxi- mum number of transactions which can be accommodated in one year is about 115,000. The circulation control system utilizes eight programs. All are written in 1620 SPS and utilize 40K storage. (An additional computational pro- gram not included in the production package is written in Fortran.) With only minor modification the programs could be made to work with 20K storage. The individual programs are described in Table 2. Tabk 2. LIB 201 LIB 202 LIB 204 LIB 205 LIB 207 LIB 209 LIB 212 LIB 213 Circulation Control System Programs To update file and to print short-and long-term reports. To print overdue notices and fines-due report. Phase 1 routine for LIB 202. Cold start program to "seed" circulation file. To restart files from one term to the next. To print pre-notice report. To print daily discharges. To print circulation file or part thereof. • Automated Book Order/ AULD 107 APPRAISAL The book order system has been described as it was originally de- signed, and the circulation control system as designed and modified. A partial update together with a critical appraisal follows. Implicit in the planning of both systems was the assumption that the IBM 1620 would eventually be replaced by a larger and faster machine and that both systems would be redesigned and augmented. However, the IBM 1620 is continuing in use for a maximum rather than minimum projected time. In July, 1965, Oakland initiated an accelerated library development pro- gram. Overnight the book budget projection for several years was avail- able and in less than three months the book order system was conse- quently overloaded. With the disk Ble filled and many orders waiting, drastic action was required. The most obvious solution seemed to be use of an additional changeable disk pack to expand the purchase order file, but this procedure would have been hopelessly unwieldy. To use a second pack would re- quire either that all transactions be run against both disk packs, roughly doubling computer time and costs, or that each transaction be addressed to a particular disk pack which would necessitate extensive systems re- design. Another proposed solution was to revert to a completely manual system, but the Order Section preferred, if at all po~sible, to retain the automated fiscal control and invoice-voucher preparation features of the order system. , The alternative finally adopted required a basic philosophical change in the system. As originally designed, the system accounted for a book from the time it was placed on order to the time it was cataloged and placed on the shelf. The disk file was one-half occupied· with items re- ceived and paid for but not yet cataloged. By purging the file of such items, an on-order file in the narrowest sense was created and a doubling of file capacity gained. Now a new problem was created. How was a book to be accounted for that had been received, paid, and purged from the on-order file, but not yet cataloged? The solution was to print a second (carbon) copy of the LC card order slip which would be hand-filed into the card catalog; there it would serve as an on-order/ in-process slip until replaced by a catalog card. Hand-filed slips replacing a machine-filed list further altered the philosophical basis of the system. Discrepancies in entry do occur, but not so often that the expedient does not work. Four months later the system was again overloaded and a routine had to be devised whereby purchase orders could be issued either manually or through the computer. However, all items were still paid via the com- puter and all invoice-vouchers computer prepared. Fiscal control was re- tained even though the rationale of the system was violated . 108 Journal of Library Automation Vol. 1/ 2 June, 1968 During the summer of 1967 a change of a different nature was imple- mented. As originally designed the system provided constant communi- cation between the Library and each faculty department through the departmental report. But, after the changes described above, the depart- mental report now included less than one-half of the items being purchased with the department's book fund allocation. It had ceased to serve any purpose and was omitted after July, 1967, with a consequent reduction of nearly two-fifths of line-printer time required for the book order sys- tem. To the question, "Would it be better to return to a completely man- ual system for ordel'ing books?" the answer by the Order Section has always been "No, retention of the automated system for fiscal control and voucher preparation is preferable, even with the patched system at hand." Nor should it be forgotten that the book order system as originally designed worked well until the demand on it exceeded its production capacity. Also to be recognized is the gain in experience and insight by the library staff during these three years. Reading about or visiting someone else's work is enlightening but day-to-day work brings an under- standing for which it is difficult to obtain a substitute. ACKNOWLEDGMENTS Four persons deserve special recognition for the roles they played in the foregoing: Dr. Floyd Cammack, former University Librarian, without · whose imagination and courage library automation at Oakland would not have been attempted; Mr. Donald Mann, Assistant Director, Computing and Data Processing Center, an outstanding systems analyst and pro- grammer; Mrs. Edith Pollock, Head of the Order Section, who likes com- puters; Mrs. Nancy Covert, Head of Circulation Department, who likes students. REFERENCES I. Alanen, Sally; Sparks, David E.; Kilgour, Frederick G.: "A Computer- Monitored Library Technical Processing System," In American Docu- mentation Institute: Proceedings of the Annual Meeting, V. 3, 1966 (Woodland Hills, Calif.: Adrianne Press, 1966) p. 419-26. 2. Cox, Carl R.: "The Mechanization of Acquisitions and Circulation Procedures at the University of Maryland Library," In International Business Machines Corporation: IBM Library Mechanization Sym- posium (Endicott, N. Y.: 1964) p. 205-35. 3. Cox, Carl R.: "Mechanized Acquisitions Procedures at the University of Maryland," College & Research Libraries, 24 (May 1965) 232-36. 4. Minder, Thomas L.: "Automation-The Acquisitions Program at the Pennsylvania State University Library," In International Business Ma- chines Corporation: IBM Library Mechanization Symposium (Endi- cott, N. Y.: 1964) p. 145-56. Automated Book Order/ AULD 109 5. Minder, Thomas L.; Lazorick, Gerald: "Automation of the Penn State University Acquisitions Department" In International Business Machines Corporation: IBM Library Mechanization Symposium (Endicott, N. Y. 1964) p. 157-63. (Reprinted from American Documentation Institute: Automation and Scientific Communication; Short Papers Contributed to the Theme Sessions of the 26th Annual Meeting ... (Washington: 1963) p. 455-59. ) 6. DeJarnett, L. R. : "Library Circulation Control using IBM 357's at Southern Illinois University," In International Business Machines Cor- poration: IBM Library Mechanization Symposium (Endicott, N. Y.: 1964) p . 77-94. 7. McCoy, Ralph E.: "Computerized Circulation Work: a Case Study of the 357 Data Collection System," Library Resources & Technical Services, 9 (Winter 1965), 59-65.