WO1992016931A2 - Archival document image processing and printing system - Google Patents

Abstract

A document processing, archival storage and printout system (Figs. 1A, 1B) such as for handling customer checking accounts. Original checks/documents are processed via document processor (8) into digital image data then stored temporarily in magnetic media (10, 20) and transferred to optical long-term archival storage (50a). The system retrieves and accumulates monthly groups of digital image data (30d), then sorts (44 and 44d) one-day's worth (1/22 of 22 business-day accumulation) by account number so that printing means can print statements (Fig. 6B), each day, covering 1/22 of the total accounts existing for that month. Massive amounts of data can be accumulated and stored, for example, for 500,000 to 1,000,000 customer accounts, while the system operates to rapidly retrieve and printout sufficient customers statements each day (Fig. 6B, 6C) so that each of the many customer will still receive a personal monthly updated statement during the appropriate month.

Description

TITLE: ARCHIVAL DOCUMENT IMAGE PROCESSING AND

PRINTING SYSTEM

FIELD OF THE INVENTION

This disclosure involves systems for storage and retrieval of documents, such as checks and financial information. An efficient sorting algorithm for the retrieval of image data is provided whereby massive amounts of statements or financial data can be retrieved and printed out in large volumes in a short space of time.

BACKGROUND OF THE INVENTION

With the present day proliferation of exceedingly high-volume databases, there is an increased desire toward automation of routine functions in the retrieval and handling of large volumes of data. This is especially the case in the work of financial institutions whereby thous¬ ands of documents such as checks, deposit slips, remittance information forms, etc. must be checked, sorted, corrected, totalled and returned to other financial institutions and where monthly statements must be prepared for individual checking accounts of thousands of customers.

While previously many financial and banking institu¬ tions were forced to maintain large staffs of people to manually handle the tedious document processing procedures, it was found efficient to provide means whereby large groups of specific amounts of data could be retrieved and printed, such as that required by a banking institutions, which found it necessary to provide hundreds of thousands of bank statements each month to its customer base. The Unisys Corporation products designated as Image and Item Processing Systems (UPS) are one example where highly automated systems were provided in order to record documents and turn them into electronic images which could be stored on magnetic disk media and retrieved at a very high rate of speed. Further, these image applications indicate how work stations could be integrated into such systems whereby system operators could quickly and easily retrieve image data regarding the documents which were placed into the system. The image and item processing system (UPS) uses a Point-to-Point Optical Link to connect an imaging module 8_X to a Storage and Retrieval Module 10. Each document processor 8 (FIG. 1A) has one imaging module 8-.. A single Image Module 8-. interfaces to a-single Storage and Retrieval Module (SRM) 10. The module 8_£ is also called the Image capture module or ICM. The SRM's (10) may be enhanced by interconnecting them to each other to form a subsystem known as the Storage/- Retrieval Subsystem (SRS). Each SRM interfaces to and supports the image storage requirements of the images being captured on the Image Capture Module (ICM) 8-_. of the document processor 8 where the storage is provided by use of high-performance magnetic disk drives.

A maximum of six SRM units may be clustered together via a high-speed data connection to provide a shared capture environment.

In addition to providing the basic image storage capability, the SRS (Storage/Retrieval System) also sup¬ ports retrieval of images from storage for transmission to the attached workstations, printers and to other SRM's. There are additional functions which are provided by each of the SRM's 10 and these include: (i) image file management (ii) workstation and printer interface manage¬ ment, and (iii) system file management, and (iv) unit management. There are a number of hardware functions provided by the SRM 10. These include:

(1) a magnetic disk write and read function;

(2) an interface controller to the disks which are controlled by the SRM which operates at a data transfer rate of 2.5 MB per second. A maximum of 4 gigabytes of disk (formatted) may be con¬ trolled by each SRM;

(3) a LAN (local area network), (IEEE 802.3) con¬ troller function used to connect workstation printers;

(4) a LAN (IEEE 802.3) controller function for communication to the host processor 6;

(5) a LAN (IEEE 802.3) protocol controller to inter¬ connect the SRM's' (6) a system bus 10_m to allow the above functions to communicate to each other within the SRM itself. It is possible to organize many different configura¬ tions in the SRM 10, each configuration of which is built from the basic configuration. The basic configuration consists of the SRM cabinet, a configuration of electronics gates and two disk drives.

Each SRM 10 uses a minimum of two hard disk drives and can support as many as eight drives. The SRM cabinet can be expanded by adding an additional module in order to support more than four drives. The software is installed at the host computer 6, and all code and dated files are released on tape for loading into the host 6. The software downloads information to the

SRM 10 from the host 6. Copies of the SRM 10 software operating systems reside on each SRM hard disk, enabling the SRM to boot the operating systems in the event of disk failure. Booting from the hard disks permits the SRM modules within the system to be initialized concurrently.

The SRM module 10 provides for the initialization of the Imaging Modules 8_lf Image Workstations 12 and Printing Workstations which are connected to it. It supports down¬ loading of software from the host computer 6 for updates to Workstations 12 and Printing Workstations. Each SRM 10 has formatted disk drives and the system software for the SRM module 10, the Imaging Module 8-., and the Image Workstations 12 and Printing Workstations. The software operating within the SRM 10 resides in an Intel 80386 Microprocessor system environment. This includes Programmable Read-Only Memory (PROM) based firmware, which boots the full system software from the hard disk during booting.

SUMMARY OF THE INVENTION

A highly sophisticated technological problem is presented when 50-100 trillion of bytes of information have been placed in archival storage, as is commonly done in a banking or financial institution, and the need arises to sort out specific portions of this data and to print them rapidly so as, for example, to be able to provide 300,000 to 400,000 individual account statements to customers during the period of a single, 30-day month. The system described herein can handle over a million account state¬ ments per month. In regard to this problem, the following high-volume image statement storage, retrieval, and printing applica¬ tion system has been developed which will permit the high- volume storage of image data, from check documents, for example, which then can be sorted and retrieved and used to print thousands of data bytes into individual statements for individual customers.

Use is made of an indexing sorting algorithm which is applied to check or document images. The volume of image data (which is used for sorting during the function of "statement printing") is very high when it is considered that the approximate check image size is 30 kilobytes (KB) or that the average document size is 50 KB. The sorting algorithm is set to perform on the "indices" rather than a given "index-plus-image-data" item so that the images are not required to be sorted, but the "index file" is the only item to be sorted.

The electronic images are stored on optical disks for archival purposes and can then be transferred to magnetic disk-buffers which can hold a 30-day supply of information sufficient to print all the required customer account information during a 30-day period. Since the optical disk transfers are generally too slow for high-speed printing application, the images are transferred to the magnetic buffer/disks before the printing. The described system provides improved features in both providing for long-term and short-term storage and retrieval of customer account data and document images. Massive amounts of data can quickly be retrieved for immediate display on a screen or for daily cycles of customer statement printout in massive numbers in rela¬ tively short, daily print cycles. Massive data storage is provided in relatively small office-equipment-space and the high rate of retrieving and sorting data permits an unusually high rate of statement printout by the printing subsystem. Additionally, rapid customer service for information, and/or replacement of document images or account state¬ ments, is effectuated in a short period of time via the rapid retrieval and printout system.

Since the stored document and image data is in binary digital form, it can also be retrieved and transferred by wire to a remote site or location for printout or for information display.

Thus the system provides solutions to the problems of long-term and short-term storage and for the long time- lags previously characteristic in sorting, retrieving and printing of massive numbers of documents. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A is an overall block diagram of the image archival storage/retrieval and printing system; FIG. IB is a block diagram showing the system arrange¬ ment for the statement printing application;

FIG. 1C shows a block diagram of the basic platform with an added archival subsystem;

FIG. 2 is a block diagram of the system for statement printing involving the least efficient (worst case) format for statement printing;

FIG. 3 is a schematic diagram showing the work flow arrangement for statement printing;

FIG. 4 is a block diagram showing the statement printing system making use of the optical drive and jukebox of optical platters;

FIG. 5 is a block diagram of a statement printing format whereby two optical drives are used for storage operations and one drive is used for retrieval operations; FIG. 6A is a chart representation of the operations of the sort algorithm, and indicates how two files are merged to create one Master Print Index File (MPIF); FIG. 6B is a flow chart showing the operational step sequence of merging information for the Master Print Index File and then sorting and retrieving image data for print¬ ing; FIG. 6C is a schematic drawing indicating how the sorted index file accesses the proper image data for printing;

FIG. 6D is a drawing indicating the Final Print Index File with new record numbers; FIG. 6E is a drawing showing how the Sorted Print Index File of FIG. 6C is burdened with many areas of "blank disk space";

FIG. 6F shows the Final Print Index File with new record numbers which eliminate the blank disk spaces in FIG. 6E.

FIG. 7 is a graph showing the sorted index file for one print cycle which can be generated from the MPIF (Master Print Index File) for every print cycle. GENERAL OVERVIEW The general system overview is seen in FIG. 1A. As seen in FIG. 1A, the central operating hub is the host processor 6 which may typically be a Unisys V Series 400 processor. Attached to the processor 6 are a number of peripherals such as a printer (PRTR), an operator display terminal (ODT), a magnetic storage disk (MSD), a tape storage unit, and an Item Data Storage unit 7 (IDS). The Item Data Storage Unit 7 (IDS) holds the MICR data (mag¬ netic ink character recognition numbers) which are on each check document to identify the bank, the check number, the account number, the type of account (checking or savings, etc.) and the amount of the check. The MICR data may also include the date of entry into the processing system, called the "capture date". The IDS also holds software for selecting various items in the MICR data. One data output bus from the host processor 6 is connected to a communications processor designated 4B. Attached to the communication processor are a number of other devices such as the Unisys B38 workstation, designed 4A. Also attached to the communication processor is a power encoder 2. The power encoders 2 are used for certain applications such as the re-entry of rejected documents and for automatically encoding items passing through the document processor 8. The power encoder 2 will pass document data through the communications processor 4B over to the host processor 6. The power encoder 2 is used to print the magnetic ink character recognition information (MICR) or else print optical character read (OCR) charac¬ ters onto the items. The document processor 8 is often designated as the Unisys DP 1800 which signifies that it can process up to 1,800 documents per minute.

As seen in FIG. 1A, the system may include up to six document processors 8 wherein each of the document proces¬ sors has an image capture module 8_ir and optionally a courtesy amount reader 8_C (CAR) . The courtesy amount reader (CAR) 8_C functions to capture the dollar amount of the numerals written on the check area which is designated for writing a numeral amount.

The document processor 8 transforms checks and other documents into electronic images via the image capture module 8_$. and then transfers it as digital data to the storage and retrieval modules 10 (SRM's). The SRM's 10 include magnetic disk media 20 for digital storage of the electronic image data.

A system operator can use a workstation such as workstation 12 to access data from the SRM 10.

For long-term storage purposes, data can be taken from the SRM 10 and processed through the archive server 30 and its magnetic disk 30_d. Then it is transmitted to a storage manager 40 which places the image data into the unit 50_j which consists of optical disk jukeboxes. The module 50_j could be called a "jukebox" unit since it consists of a multiple number of optical platters, each of which can be accessed separately. The archive server 30 and the storage manager 40 have access to magnetic disks 30_a which provide additional storage capacity (temporary) to the archive server and storage manager. Also connected to the storage manager unit 40 of FIG. 1A is a Library Unit 50_a which can be used for archival purposes of long-term storage of image and information data. This Library Unit would preferably be an optical jukebox unit holding multiple banks of optical platters for storing digital data.

Connected to the archive server 30 is a workstation 22, a remote interface 24, and CCITT gateway 26 which can transfer image data to another documentation system.

The workstation 22 permits a system operator to retrieve archival data for viewing on a window screen.

The remote interface 24 permits data from the archive server to be transmitted to a remote workstation for display to a remote operator.

The CCITT gateway 26 provides a communication link to a transport control protocol/internet protocol (TCP/IP) to ensure that data packets are delivered to their destina¬ tions in the sequence in which they were transmitted. Functionally, documents such as checks, are passed through the document processors 8. The image capture module 8i makes an image of each check as it passes through the image capture module and optically transfers the corresponding digital data bits over to the storage/- retrieval module SRM 10.

For longer term storage, the digital image data from the magnetic media disk 30_d is sent to the archive server 30 and then to the storage manager 40 for storage and place¬ ment on optical disks in the optical disk jukebox 50_j. The optical disks are used for long-term storage to provide an archive function.

The host processor 6 can be programmed to select certain account numbers on certain days of the month and to cause the monthly account data to be retrieved from the optical disk jukebox 50_j for placement onto the disks 30_d, which then can transfer to disk 44_d to rapidly disgorge this information to a print subsystem 42 which can print each of the account statements in a rapid fashion.

The following requirements may be listed as a summary of the needs and functions for a statement-printing system. For "ON-US" Statements: ("ON-US" statements are those "belonging" to the bank doing the processing.) There is a requirement that statements be printed monthly and the printing include "front" side of the image only. For each page of printed material, there will be total of eight images provided. The printer used will have a throughput in the range of 50-220 pages per minutes.

For printing of statements in the United States, an assumption will be made with respect to a mid-size bank which carries approximately 400,000 customer accounts, as well as commercial and reconciliation accounts. The subsequent calculations will be based on such a mid-size bank with the above number of accounts.

The check activity for the bank having an estimated number of "ON-US" items per day would look like the follow- ing:

There may be 400,000 customer accounts each having approximately 20 checks and covering a processing cycle period of 22 business days which could lead to involvement of 8 million checks per month. The commercial-type account might be estimated to be 50,000 accounts, each account having 20 checks per monthly statement and having a processing cycle of four days which would sum up to approximately 1 million checks per month. The reconciliation-type accounts could be estimated at 1,000 accounts which would have a processing cycle which would be daily (each day) and this would be estimated to involve 20 checks for each account which would lead to a total number of checks per month of 20,000.

Taking the total summation of checks involved in the above examples, the total number of checks would be 9,020,000 per month. Now considering one calendar banking month as being 22 days and considering the optical storage requirements for 22 days, the following calculations must be considered. For the customer-type account, these could be con- sidered to be 25 percent business checks and 75 percent convenience checks. The front of each check would involve an image having 14 KB, while the back of the check would have an image taking 9 KB, leading to a total for the check size to be 23 KB. This would require that there be pro- vided a total storage (ON-US) of 220,000 MB (megabytes).

The commercial accounts would be considered as 100 percent business checks which would require an optical image having 25 KB for the front of the check and 16 KB for the back of the check, giving a total of 41 KB for total image information for one check. This would require a total optical storage requirement of 410,000 MB (41,000 bytes x 1,000,000 checks).

In regard to reconciliation-type accounts which have to be retrieved and balanced daily, it may be considered that these would be 100 percent business type checks which, again, would require an optical image size of 25 KB for the front and 16 KB for the back, a total information storage of 41 KB per check. This would require a total optical storage of 820 MB (1,000 accounts by 20 checks each x 41 MB).

TEN-YEAR STORAGE REQUIREMENT FOR CUSTOMER, COMMER¬ CIAL, AND RECONCILIATION ACCOUNTS: The following summary will indicate a gradation of storage requirements over a period of time for archival purposes. Thus, the 9.02 million checks (ON-US) for a monthly banking period of 22 days would require a total storage of 436.36 GB (gigabytes or 1 billion bytes) of which 60 percent of this figure would come to 261.82 GB assuming that 60 percent of the total checks processed are ON-US items, and 40 percent of the total checks processed are "transit" items which are merely passing through the bank on to final destinations at another bank than the local bank. The storage requirement for one year would involve 108.24 million checks for a total storage requirement of 5,236.40 GB for all checks and 3,141.84 GB for storing ON¬ US checks. The storage requirement for ten years would involve 1,082.40 million checks (ON-US) requiring a total storage of 52,364 GB for all checks and a 60 percent figure of 31,418.40 GB for storing ON-US checks.

Thus, the total storage requirement for ten years would come to 52,364 GB which is equivalent to 52.364 tera¬ bytes, or to put it another way, this is equivalent 52,364 billion bytes. A terabyte is 1,000,000,000,000 or 1 trillion bytes.

In the following system, it will be seen that the total print window required for a complete statement printing cycle involves time required to (a) retrieve, (b) sort, and (c) print images. It is assumed that the daily reconciliation account work statement printing is done after all of the check images are transferred to the jukebox (archival storage 50_j). A total of 3 platters may be accessed to accomplish the task of account recon¬ ciliation statements on a daily basis.

In the present retrieval and printing system for monthly statement printing, there has been eliminated a very time-consuming step which was characteristic of the prior art. In the prior art, very usually, the check "image data" were sorted. In the present system, the check images are never sorted. It is only the "index file" which carries reference information (about the check images) which is sorted.

Previous sorting methods such as that developed in US patent 4,611,280, involve sorting algorithms or "variable record" lengths, and which used more than one variable length sort key. In the present situation, the sorting methods involve using data to be sorted which is of a fixed length in the sort field. Additionally, the present system involves fixed record sizes since it is only the index file that is sorted, and never the actual check image data records.

For example, the average record sizes for a check image may be in the range of 30 KB to 50 KB. This is a relatively large size record which is possibly 50 times greater than the traditional data processing record sizes which may be from 128 to 1,000 bytes (1 KB).

If it were necessary to access and sort these various lengths of check image data, a very time-consuming set of steps and functions would be involved. Thus, the present system operates only on fixed-length index file record sizes which can be sorted very quickly.

This is accomplished by transferring images (digital image data) from the archival optical jukeboxes to a magnetic media which holds data covering a period of 30 days only (to cover the 22-day-banking-cycle monthly period) . The actual check images are related to an index file, and it is only necessary to sort the index file and then retrieve the check image data in such a manner that many periods of sorting time are saved and reduced in an efficient manner. DESCRIPTION OF PREFERRED EMBODIMENT

The basic architecture for the document storage and retrieval process system is shown in FIG. IB. Here the host processor 6 is shown connected to the communications processor 4B and also to each of three document processors 8_a, 8_b, and 8_C. These may preferably be units such as Unisys DP1800 Document Processors which can process 1,800 docu¬ ments per minute by converting a paper document to optical digital image data for transmittal to a storage means.

Each of the document processors are connected to respective Storage Retrieval Modules 10 (S^riM's 10_a 10_b, and 10_c). The outputs of the SRM's 10 are connected to the archive server 30, which is part of the archive subsystem 50. Subsystem 50 is composed of the archive server 30, which is connected to the optical disk jukebox 50_j. The archive server 30 is also connected by a standard protocol communication line (IEEE 802.3) to a remote interface 24 which can convey the archival data to a remote printing station, 45.

As seen in FIG. IB, the printing subsystem 42 is composed of a print server 44 which connects by means of a small computer system interface (SCSI) to the archive server 30. The output of the print server 44 is connected to printers 46 and 48 which in this particular case, are shown as having 220 pages per minute printing capabilities. FIG. 1C shows the "Basic Platform" for document processing with the additive subsystem 50 designated "Archive Subsystem".

A check or document C_d is processed through a document processor 8 where an image is made of the document in terms of digital data (optical) which is transmitted to a Stor¬ age/Retrieval Module (SRM) where it is converted to mag¬ netic digital media storage.

A workstation 12 (Fig. 1A) may call up and retrieve the stored magnetic data for visual display if desired. Normally, the magnetic digital data will be conveyed to the Print Subsystem 42 for printing.

In order to provide for long-term (40-year) storage, the Archive Subsystem 50 can receive the digital magnetic data from the SRM 10 and convert it to optical digital data for storage on optical platters in a jukebox 50_j since the

SRM 10 is used only for short-term data storage.

The concept for this system has been considered in terms of three alternatives. The first alternative is the "worst-case configuration" of FIG. 2. Here, the document images are stored in the jukebox 50_j in the captured order. It then takes between 18 to 22 total clock hours to process and print statements for one complete daily cycle of a mid¬ size bank with 400,000 customer accounts. This assumes that the number of statements is approximately the same for each monthly print cycle. However, using the system of FIG. IB, the "printout window" can be reduced to between 9.25-10.25 hours daily by the use of four printers instead of the two printers 46 and 48 of FIG. IB. This assumes that each customer monthly printout statement report is approximately three to four pages and includes the text for printing. The second alternative configuration (Alternative No. 2) would involve a slightly different hardware configura¬ tion. In Alternative No. 2 shown in FIG. 4, a combination of an optical drive device (such as made by LMSI Company, whose address is Laser Magnetic Storage International Company, 2914 East Katella, Suite 212, Orange, California 92667), and the jukebox 50_j can be used for performance improvements. The LMSI optical storage units are designed for short-term storage. In the instant configuration, they are used for 30 calendar days (that is to say, 22 business- day statement cycles) which is enough to print all of the monthly statements required. This would require about 9- 10 hours daily with four functioning printers which provide for 3-4 pages plus text (for each customer statement report) . The third alternative configuration of FIG. 5 involves the concept of "ON-US" items which are separated from other "transit" items. The "ON-US" items are accounts which belong to the bank which is operating the storage and re¬ trieval system. These should be distinguished from data and documents which involve checks or information which belong to "other outside" banks, such as, checks which are passing through the local bank on their way to final destinations at other-owned remote banks. The locally owned "ON-US" items are stored within a fixed range of media in the jukebox 50_j of FIG. 5. This enables the system to access a minimum of number of platters (42 platters out of 120 total) for statement printing, and in so doing, increases the print speed performance. In this case, the time factor involved would be 9-10 hours daily with four functioning printers printing reports of 3-4 pages each, including the text. Thus, there is approximately one-hour's savings in time between the "worst-case" design and the alternative third design configuration. The saving in time in the third design configuration is due to the handling of a comparatively small number of platters (42 out of 120) and wherein an efficient indexing sorting algorithm is used in each of the three configurations. All document images are accessed using "pointers" (FIG. 6C), and the use of the sorting algorithm involved permits the rapid sort and retrieval and printout of the high volume of customer accounts in a relatively short space of time. Thus, while the normal time cycle to print one daily cycle of about 18,000 accounts of the base 400,000 customer accounts would normally take 18 to 22 clock hours, it can been seen that the present system will reduce this to somewhat below ten hours daily which is approximately a 50 percent reduction in the time previously required.

FIRST WORST CASE CONFIGURATION: In this situation of FIG. 2, all the check images are written to the optical storage jukebox 50_j in the "capture" order. That is to say, the sequence in which the checks are inserted into the document encoder and captured by the image capture module 8_t is the sequence order in which the checks are placed in the optical storage jukebox 50_j. Assumptions, for analysis purposes, are made in order to analytically view the operation of the system. Thus, consumer accounts are divided into N cycles to conform to the N banking days of a month. "N" is typically in the range of 20-25 days. An N value of 22 will be used for the calculations in the examples. It is also assumed that the bank has 400,000 customer accounts and that the cycles are divided as shown in Table I. Cycle NOJ Account Range

1 1i -- 1ι8o,,0υ0u0u 2 18,001 - 36,000 3 36,001 - 54,000 4 54,001 - 72,000 5 72,001 - 90,000

22 382,001 -400,000 1/28- 2/28 2/28

Thus during the 22 cycles, each cycle involves the printout of 18,000 account statements on a given day of the month so that, for example, on February 1, the system prints out 18,000 statements; on February 2, it prints out 18,000 statements, and so on, until February 28 it prints out the final group of 18,000 statements which cover the month-of-January transactions.

Under these assumptions, it would require on-line storage for 30 days in the jukebox 50_j in order to complete the statement printing for a single month for each and every one of the customer accounts involved. The jukebox 50_j has optical platters 50_p which are handled by storage drive 52 and retrieval drive 54. (Fig. 2). It is assumed that as the checks are written in the captured order, the image data could be found in any one of the optical platters 50_p of FIG. 2.

It is further assumed that each optical platter of 50_p has a capacity of 10 gigabytes, that is to say, this is 1,000,000,000 bytes (1 billion) or 10⁹. In the worst-case configuration noted in FIG. 2 , the image storage of the check document data are placed in the optical platters 50_p according to the sequence that they are captured, that is to say in the captured order. For retrieval purposes, the images for any given account range (see Table I) are retrieved from the on-line storage (jukebox 50_j) and then transferred to the print server 44 by means of the archive server 30. The work flow in this system can be better understood in reference to FIG. 3 and FIG. 6B.

RETRIEVAL ACTIVITY ANALYSIS: The goal is to retrieve all the data for any particular cycle from the jukebox 50_j and transfer it to the print server 44. As an example, it may be helpful to look at cycle 1 (Table I) with the account range: 1 to 18,000, to observe the sequential functions.

STEP 1: Loading of platters and transferring image data in binary digits to the print server 44, which is done as follows:

(al) Access the first platter (50_p to platter, FIG. 2) of the month involved by loading it into the optical driver 54, FIG. 2; (bl) Transfer all the data (by addressing the index file) for the range of account (1-18K) for that day to the magnetic disk storage (30_d of FIG. 2) of the archive server 30; (cl) Transfer all of the data to the print server 44 and store it in the magnetic disk 44_d of FIG. 2;

(dl) Load the next platter (50_p) of the optical disk jukebox 50_j, FIG 2; scan the account range (1-18K), and transfer the check image binary data to the print server 44 via the archive server 30; (^el) Continue the process until all of the first (1-18K) account range has been scanned and transferred to print server 44. At this stage, it is seen that all the data for the account range (1 - 18,000) for cycle 1 is in the magnetic storage 44_d of the print server 44. The amount of data stored in the print server 44 for "customer accounts only" would be as follows: (18,000 accounts X 17 KB (average image size, font only)

X 20 (number of checks) = 6.12 gigabytes (customer)

Then, for 12,500 "commercial accounts", each having a 25-KB image size: (12,500 X 25KB (average image size font only)

X 20 (number of checks)

- 6.25 gigabytes (commercial)

The first step as illustrated in FIG's 2 and 3, involved the loading of the optical platters and the transfer of data to the print server 44.

STEP 2: The second step, or step 2, involves the sorting of data and the printing of statements. This is done by print server 44 as follows: (a2) Sort all of the index file data (done via Archive Server 30) by the account number and by sorting the checks in sequential order. This is accomplished by the "sort algorithm" discussed hereinafter. At this juncture, it is necessary to copy the "Master Print Index file" from the archival server 30 into the print server 44; (b2) After the customer account numbers are sorted, they will require decompression by the Print Server 44 before they can be printed in full copy, since the image capture unit 8i originally compressed the image data; (c2) Then the task of printing is distributed by the software in Print Server 44 to the available printers involved. The total estimated magnetic storage in disk 44_d that might be required for the print server 44 to print cus¬ tomer, commercial, and reconciliation accounts on the very same day are indicated as follows in Table II:

TABLE II Type of Account Storage Required

Customer - 6.12 GB

Commercial - 6.25 GB

Reconciliation - 0.50 GB

Print Index File - 0.80 GB

The Print Index File requires 100 bytes for each "index record" for storage for a period of 30 days. The total maximummagnetic disk storage required may be estimated at 6.25 GB + 0.80 GB + overhead = 7.05 GB. Added overhead would be required for the Print Server 44.

In regard to doing a performance analysis wherein the "front" and the "back" of a document are in one file, the following assumptions are made:

1. The "ON-US" statement printing is done during the night.

2. The window time for printing is 10 to 12 hours. 3. Printing is done in off-business hours and in a batch mode.

4. The images of the "front" side of the check are printed in the actual statements.

5. The printer speed is set at 90 pages per minute (PPM) .

6. The time required to change platters via optical retrieval drive 54 is 6 seconds.

7. The capacity of each platter 50_p is 10 giga¬ bytes (GB) . 8. The "front" and the "back" images are stored together in the same files in the captured order. The front images are only retrieved for printing purposes while the back images are skipped. Now, using as an example, a CYGNET 1800 Jukebox with a Hitachi drive, the following projections can be made. The CYGNET 1800 Jukebox is manufactured by CYGNET SYSTEMS, INC. of Sunnyvale, California, whose address is 601 West California Avenue, Sunnyvale, California 94086. The Table III hereinbelow indicates the various factors involved in the first (worst case) configuration regarding the CYGNET 1800 Jukebox and the forthcoming higher capacity jukebox drive. TABLE III

FORTHCOMING

Specifications CYGNET 1800 CYGNET DRIVE

Media capacity 2.6 GB/Platter 10 GB/Platter Average Seek Time 200 ms 50 ms Average Latency 50 ms 25 ms Transfer Rate 440 KB/sec 800 KB/sec Seat & Spinup Time 4.5 sec 1.5 sec Spindown Time 3.5 sec 1.5 sec

Using the above-mentioned assumptions, an analysis can be made which would indicate that the total storage required for 22 days (a banking month) would come to 436.36 GB. The total "ON-US" items storage for one month would be 60 percent of 436.36 GB and this would come to 261.81 GB. Thus, the total number of optical platters required for "two months storage" would be approximately 46, which is to say that using a 0.95 effectiveness factor times required storage of 436.36 GB, times 10 would equal 45.89 or approx¬ imately 46 optical platters needed for two months. Thus using platters each holding 10 GB, then 46 platters would provide a total storage of 460 GB which could handle the required 436.36 GB for the 22-day bank month.

It is estimated that the effective usage for each platter is at the level of 95 percent. The system operates such that it would always be accessing 28 platters for every given cycle of monthly statement printing.

It has been estimated that the average document (customer account) image size (front and back) comes to 27.5 KB. Likewise, the estimate for the average image size for statement printing of the "front" image (customer account) only would be 17 KB. The total number of images stored on one platter would be 345,454, while the total number of images stored on "one side only" of a platter would be 172,727.

The total number of "ON-US" images on one platter would be estimated at 207,272, while the "average" ON-US images/cycle/platter would come to 9,421. Thus the average ON-US images/cycle/one-side of platter would come to 4,710. The ratio of the total number of images/images- retrieved per platter would come to 36.6 and this means that approximately one in every 37 images will be re¬ trieved. Now, in order to "retrieve" one image (on an average basis), it is necessary to move 37 images, to wait for latency, and to read the desired image from the optical platter. This requires approximately 10 milliseconds, + 25 milliseconds + (17 X 1000/800) milliseconds. To retrieve images from one side of a platter (4,710 images), it is necessary to drive the platter, seat the platter and use spin-up time + spin-down time + read time which comes to an estimate of 1.5 seconds + 1.5 seconds + 0.05625 X 4710 which equals 267.93 seconds. Thus dividing 267.93 by 4710 results in a time of 0.0568 seconds per image on an average basis. Thus, it would take 56.8 milliseconds per image to retrieve an image from one side of a platter.

Thus, in order to print a daily cycle of 18,000 "customer account" statements, the print time required would be 18,000 (customer accounts) X 20 (checks per account) X 0.0568 seconds (image retrieval) which comes to

20,448 seconds or 5.6 hours.

Likewise, for "commercial accounts", the estimated time to retrieve one image would come to 31.25 milliseconds and to retrieve images from one side of the platter (4710 images) would take 315 seconds and this would involve the average time value of 66.88 milliseconds per image re¬ trieved. The total time required to retrieve 12,500 images covering a monthly accounting period of the prior 22 days would involve 16,720 seconds, or 4.6 hours.

Sometimes it is necessary to retrieve image data in order to reconcile certain discrepancies in account data. For this "reconciliation" operation, the following analysis would indicate that to receive images from one side of the platter would take 25.27 seconds. Assumption is made that 1,000 image items (covering the prior month) located on three sides of platters would be involved. Thus, 1,000/3 equals 333 images on one platter side and the total number of platters/cycle would be equal to 28/22 which equals 1.27 thus involve the use of 3 sides of the platters. Thus, 1.5 + 1.5 + 0.06688 X 333 equals 25.27 seconds. Then 25.27 seconds divided by 333 images equals 0.0758 seconds re¬ trieval time per image.

Thus, the total time required to "retrieve" (on a daily basis) 1,000 reconciliation accounts covering the prior month period, would come to 1,000 X 20 X 0.0758 which comes to 1,516 seconds and which is equal to 0.421 hours, or approximately one-half hour.

It is assumed that there is no additional time required to transfer all image data from the archive server 30 over to the print server 44 as the images are "trans¬ ferred" to the print server 44 in the 5.6 hour "retrieval window" which is also used for the archive server.

In this situation for the "reconciliation" accounts, the time to "sort" 18,000 accounts covering the prior month, could occupy from a few minutes to approximately 0.5 hour.

PRINT PERFORMANCE ANALYSIS: After retrieval, the system then functions to execute the printout cycle. Here the following assumptions are made: PI: Compressed check-sized images will be printed.

P2: Each printout page can hold up to eight images.

P3: Printing speed will operate at a speed of 90

PPM. P4: The printer is assumed to operate on an 80 percent duty cycle.

It will be noted from the previous analysis, the average "retrieval time" per image for customer checks would be 56.8 milliseconds (ms). Thus the "retrieval time" for a total of 18,000 accounts, each having 20 checks within them, would come to 18 X 20 X 0.0568 which comes to 5.6 hours. SITUATION 1: FOR PRINT OUTS AVERAGING THREE PAGES PER ACCOUNT AND INCLUDING TEXTS:

The following analysis would occur under this first situation where the total number of pages required to be printed comes to 54,000 pages. Since there are 18,000 accounts handled per daily cycle multiplied by three pages for each account, this comes to 54,000 pages. The time required to print 54,000 pages in only "one" printer would come to 12.5 hours, that is to say, 54 divided by (90 X 60 X 0.8 duty cycle) equals 12.5 hours.

The time required to print 54,000 pages with two printers would be 6.25 hours, and the time required to print 54,000 pages with three pages would come to 4.16 hours; while using four printers, this would come to 3.125 hours.

Thus, by combining the 5.6 hours required for "re¬ trieving" 18,000 accounts with 20 checks each, plus the 12.5 hours required to "print" 54,000 pages on one printer, plus the one-half hour (0.5) required to sort 18,000 accounts, this would lead to the total clock hours for the complete daily print cycle to be 18.6 hours. This would complete one cycle of Table I so that 18,000 statements would be completed on February first.

For commercial accounts where the total number of pages requiring printing would come to 37,500 pages, that is to say, 12,500 accounts by three pages each, the calcu¬ lated total print cycle time would come to 8.6 hours which is 37,500 divided by (90 X 60 X 0.8).

Likewise, for reconciliation functions where the total number of pages would be 3,000 pages, or 1,000 items X 3 pages, the total print cycle time would come to 41.6 minutes.

SITUATION CASE 2: TOTAL AVERAGE OF FOUR PAGES PRINTED PER ACCOUNT INCLUDING TEXT: In this situation, the total number of pages to be printed in ON-US customer statements would be 72,000 pages and with the use of one printer, this would take 16.6 hours. With two printers, it would be 8.33 hours; and with 3 printers, this would involve 5.53 hours, while with four printers, this would only take 4.15 hours. Here the total clock hours required for printing (on one printer) would be 22.7 hours which would mean the use of 5.6 hours for retrieving 18,000 accounts with 20 checks each plus 16.6 hours which is the time required to print 72,000 pages on one printer, plus one-half hour (0.5) which is the time to sort 18,000 accounts. Likewise, using four pages of printing in commercial account statements, then for "commercial" accounts, the total number of pages required would be 50,000 pages which is 12,500 X 4 and the total print time would come to 11.57 hours. Likewise for the "reconciliation" account function, then for the retrieval of 1,000 documents printed on 4 pages, this would come to 4,000 pages and the total print time would be 55.55 minutes.

Referring to FIG. 3, it will be seen that the check image archives are kept in the optical jukebox 50_j and the check images are stored in their capture order.

For one statement cycle (which covers 22-banking days), all of the check images are transferred to the print server 44 from the jukebox 50_j by means of the archive server 30.

The print server 44 gets a "list" of items to be printed from the host processor 6 which is also called the demand deposit account host or DDA host 6. The print server 44 retrieves check images for any statement cycle by scanning each of the platters 50_p in the optical jukebox 50_j for a 30-day period.

Then as seen in Block 44₈ of FIG. 3, the sorting of check images is done according to their sequential "account number" . Data is provided to print the statements required for the given printing cycle. Thus, in a typical, medium- size, modern bank, there can be provided a "daily print cycle" which retrieves and prints statements (of data from the past month of 22 banking days) on a daily basis to print some 18,000 account statements per day. Thus, over a 30-day month of 22 work days (bank work days), the system would be capable of printing out 396,000 account state- ments, or more.

Referring to FIG. 3, the "ICPS" is the Image Check Processing System providing software for various capa¬ bilities. The SRM images 10-_. are available for reading and sorting. Likewise, the images in the SRM can be accessed for amount entry, for image data correction, and for balancing accounts.

The embodiment of the storage/retrieval and print out system of the present disclosure makes use of a sorting algorithm which is graphically represented in FIG's 6A, 6B, 6C and 7.

The sort algorithm involves the following steps: A. Creating an index file: This involves the following steps:

(a) Create a "Print Index file" in disk 44_d to keep important information about the checks which have been processed. This is placed in the magnetic disk 44_d of FIG. 2. This is done by using the extraction method of the IPS, (image processing system) using the IDS disk 7

Fig. 1A. The index file in disk 44_d will have fields such as:

Date (of capture of document into the system); Account Number; Check Number; Amount of Check. (The capture date is placed on the original document via magnetic ink encoding. )

(b) Copy the modified Print Index file from disk 44_d to the Archive Server 30. Then add two more fields to that particular account file which will correlate the (i) platter number and (ii) record number. These can be received as a return value after writing the document data into the optical platter in the 5 Jukebox 50_j. This file is designated as the "Master Print Index File" (lower half FIG. 6A). (c) Build this file up to a capacity of 30 days by adding daily extractions to the

10 original index file. In this system, it is contemplated to use only 30 days of indexing information (onemonthly state¬ ment cycle) for statement printing. A brand-new print index file may be

15 created after 30 days to be used for the next month's statement printing cycle. A total of 12 Master Print Index Files would be created for the yearly period.

Table IV, shown hereinbelow, indicates the appear¬ ance of the Master Print Index file from a complete cycle.

TABLE IV Mer ed Master Print Index File

(d) Before any printing occurs, the complete index file with 30-days of information data will be transferred to the print server 44.

B. Sorting the index file: The next sequence of steps for "sorting" the index file (by account number) operates as follows:

(e) The index file is now sorted (via the software in server 44) by its "account number" before any digital image data will be transferred from the archive server 30 to the print server 44. After the index file has been sorted, the newly "sorted" Master Index File in server 44 will show the account information in the order as seen in Table V. TABLE V "SORTED MASTER INDEX FILE fIN SERVER 44.

The "Sorted Master Index File" on the print server 44 will follow the pattern as shown in Table V since it is transferred from the archive server 30 to the print server 44. The digital image data retrieved from the jukebox 50_j, according to the platter sequence, is trans¬ ferred to the print server.44. The print server 44 (with its magnetic disk buffer 44_d) has a large file in the same platter sequence order just as is set up in the jukebox 50_j.

The large file 44_d in the magnetic media of the print server 44 may take an appearance similar to the FIG.

7 upper left block. As seen in FIG. 7 in the upper left block, there are a series of account numbers such as A/C- 1, CK 5, which indicates that this record is the fifth check in the Account 1. As another example, the 18,000th account is designated as A/C-18,000 and the nineteenth check is designated as CK-19. It will be noted that the various account numbers and check records are allocated to various "logical platters" so that the upper most group is stored in the logical platters 1 through 5, and similarly the lower groups of data are stored in logical platters 21- 42 which corresponds to areas of the jukebox 50_j. The lower portion of FIG. 7 shows the logical sequence of data according to the account number, the check number, the logical platter number, and the record number on the platter, which would then correlate with the amount of the check and the date of the check. The logical platter number in FIG. 7 is software information which relates the "physical" platter location to the check image data in magnetic disk 44_d. This is the "sorted index file" for only one daily print cycle of 18,000 customer accounts. By using the Master Print Index File, this sorted index file can be created for every single print cycle.

In FIG. 6C, the digital image data retrieved from the jukebox 50_j by platter sequence is transferred to the print server 44. The magnetic disk buffer 44_d of the print server 44 has a large file in the platter sequence order just as is done in the jukebox 50_j. When a record is searched by platter number, a "pointer" (via software in print server 44) will be moved to the appropriate area of the magnetic disk buffer file 44_d in order to access that particular record.

The magnetic buffer 44_d of print server 44 (FIG. 2) will contain check images (digital data) as indicated in FIG. 6C where the Sorted Print Index File shows the optical platter sequence for each account number. However, due to the situation indicated in FIG. 6E, the Sorted Print Index File of FIG. 6C is burdened with many areas of "blank disk space".

In order to eliminate the time-consuming factors that this would entail, it is desirable to develop a Final Print Index File, such as indicated in FIG. 6F and Fig. 6D, which eliminates the blank disk space involved. This is ac¬ complished by software in the Print Server 44 which op¬ erates to eliminate the blank areas (compression of data), and in so doing, replaces the "old record number" with a "new record number" as indicated in FIG. GD.

There is no time penalty involved in this conversion since this compression of data from the S -ι:ed Print Index File to the Final Print Index File is executed during the same time period as the retrieval of data is occurring from the optical juke box 50_j to the magnetic disk 44_d of the print server 44. FIG. 6B indicates the step of the Compression Routine, done via software whereby blank record areas are eliminated so that the old record number in FIG. 6D is replaced with a "new" record number to form a "Final Print Index File" as shown below in Table V-A.

TABLE V-A

Account Check Record Number Number Number

1 1 1 1 2 4

• • •• ••

2 1 3

2 2 2

• • • 2 20 5

The record numbers are new and reassigned for each check number and account number.

While the archive server disk 30 requires 0.80 GB to hold 30 days of data in the Master Print Index File, the print server disk 44_d only requires 1.8 MB (Table IV) for holding "one day's worth" of Master Print Index File data, to enable one day's work of statement printing.

It may be noted that the platter numbers shown in FIG. 7 have already been mapped to the magnetic disk 44_d (FIG. 5) of the print server 44, which data is shown in FIG. 7 upper left block.

(f) At this stage, the digital image data for the given monthly print cycle is now available in the magnetic buffer 44_d of the print server 44. (g) The next step is the reading of the "correct record" as printing gets started. The record access is directly accomplished by the use of pointers (FIG.6C) operated through software in the print server 44. The time to search one record will be approximately

15 milliseconds and the time to search all 18,000 accounts will be approximately 90 minutes. This can be seen as

18,000 accounts X 20 checks (average) per account which gives us 360,000 records. Then the 360,000 records X 15 milliseconds results in 5,400 seconds, or 90 minutes, for searching all 18,000 accounts.

Since the printing time, using four printers, requires approximately 8 to 10 hours, it is possible to search records for image data and print at the same time

(except at the very beginning of the printing process) . It is necessary to pre-process and keep the print file ready before it is possible to go ahead with the printing. After a few minutes of pre-processing, it is possible to continue to do both the pre-processing and printing at the same time, i.e., the searching and accessing of image data records can occur concurrently with printing operations.

The effective time required for sorting is con- sidered to be approximately 30 minutes.

FIG. 6A, in the upper portion, shows the initial print index file which is extracted from the image process¬ ing system database in IDS 7 Fig. 1A. This is done on a daily basis wherein the corresponding platter and record number are correlated.

Thus, as seen in the upper part of FIG. 6A, a correlation link is started between a given capture date, the account number, the check number which goes with the account number, and the amount of money involved. This is correlated with a platter number such as Platters 1 and 2 for the "date" of 3/14 (FIG 6A) . Then additionally, each platter number is "associated" with a given "record number" to indicate the location of that digital image check data. This is done by software in archive server 30. For ex- ample, Platter 1 has the first record at position 1 and the second record at position 2 (FIG 6A) . Then Platter 2 has the image data for check 3 of account number 13,000, located on Platter 2 at record position 2.

The lower portion of FIG. 6A shows a duplicate of the information arranged similarly to the upper portion of

FIG. 6A, but the two files are "merged" to Archive Server

30 (FIG 6B) to create one Master Print Index File. The top two files of FIG. 6A are created daily and merged to create the Master Print Index File which is then placed in disk 44_d. These data are maintained throughout the monthly statement cycle. Subsequently, this master file can be deleted after one monthly statement cycle has been printed; and then a new one may be created.

In the overall summary, check documents are imaged into digital image data and placed in the Storage/Retrieval Modules 10 (SRM), then transferred to optical platters 50_p in the jukebox 50_j via Archive Server 30.

The system creates an "Initial Print Index File" (FIG. 6B) from IDS disk 7 (FIG. 1A) with detailed checking data in digital form.

The Archive Server 30 "merges" its Archive Index (of each dated digital image entry with its location of platter number and record number) with the Initial Print Index File to develop the "Merged Master Print Index File" in the

Archive Subsystem 50.

The Archive Server 30 and Archive Disk 30_d accumu- late data in the Merged Master Print Index File to cover a 30-day monthly period (of 22 business banking days).

The Archive Server 30 extracts and copies one day's worth of index records, of its accumulated 30 days of data, into the Magnetic Disk 44 of the Print Subsystem 42 (FIG. IB).

The Print Server 44 "sorts" (FIG. 6C) the account numbers in the Merged Master Print Index File (holding one day's worth of the 30 days of data) to develop a set of sequential account numbers for each date of the 30-day cycle together with correlated data indicating (for each check document) the location of the platter and record number of image data now residing in disk 44_d (which was copied from the jukebox 50_j.

Then the Print Subsystem 42 prints out N statements each day where N is equal to 1/22 of the total number of customer accounts. FIG. 4 indicates a second alternative configuration which uses a specialized drive called the LMSI drive together with the jukebox 50_j. The LMSI 6100 Series may include ten drives and 50 platters which involve one drive and five platters per unit. The LMSI is a unit manufac¬ tured by Laser Magnetic Storage International (LMSI) Company located in at 2914 East Katella Avenue, Orange, California 92667.

In the second alternative configuration of FIG. 4, the jukebox 50_j showing 120 optical platters is used for archival storage when it is necessary to retain and main¬ tain data for over 30 days. The LMSI 6100 Series drives and platters, designated as 57 and 58 of FIG. 4, are used for the storage and retrieval of data which will only be held for a 30 day period.

Since the LMSI drives have a higher transfer rate and shorter access time, it is possible to speed up the retrieval/ sorting and printing processes involved. Thus the check images will be stored in the LMSI based optical platters 57 and 58 (FIG. 4) for a period of 30 days. Subsequently, the information on these platters will be transferred to the jukebox 50_j after the statements have been printed.

The assumptions for this configuration are the same as those for the first-mentioned design configuration, except that no time period is required to change platters, since the platters here are always connected with the drives, and the capacity of each platter is 10 GB. Addi¬ tionally, in this second design configuration (FIG. 4), the drive can seek/read from one side of a platter at a time, even though it has two seek-heads on two sides of the drive.

The following Table VI indicates the specifications that are projected and based on the CYGNET 1800 Jukebox which has a Hitachi Drive. SPECIFICATIONS

Media Capacity Average Seek Time Average Latency Transfer Rate Seat and Spinup Time Spindown Time No. of Seek Heads

PERFORMANCE ANALYSIS: An analysis similar to that previously done for the first configuration would indicate, in this second configuration (FIG. 4), that the total storage required for 22 days (banking month) would be 438.36 GB. The total "ON-US" item storage for one month would come to 261.81 GB, and the total number of optical platters required for two months storage would be 46 platters.

The total number of images stored on one platter would be 345,454 and the average "ON-US" images per platter would be 207,272 (60% of 345,454).

The average "ON-US" images per daily (1/22 of

207,272) cycle per platter would come to 9,421 images. The ratio of the total number of images to the images retrieved would come to 36.6. This would mean that approximately one in every 37 images would be retrieved.

In order to retrieve one image (on the average) , it would require that the system move 37 images, wait for latency, and read the front image. This would take 53.8 millisecond per image when reading the front image.

In order to retrieve images from one side of a platter (4710 images), the read time would involve 0.0538 X 4710 images which would come to 253.3 seconds, or 53.8 milliseconds per image. For "customer account" statements, assuming 18,000 accounts each having 20 checks, and each taking a read time of 53.8 milliseconds, this retrieval would come out to 19,368 seconds or 5.3 hours. Likewise, for "commercial accounts", the total time to retrieve the data for 12,500 accounts would come to 4.3 hours.

For the "reconciliation" accounts, the time to retrieve image data from one side of the platter would come to 20.87 seconds for retrieving 333 images, and this would come 0.0627 seconds per image. Thus, the total time required to retrieve, on a daily basis, 1,000 recon¬ ciliation accounts would come to 0.34 hours.

There is no significant time required to transfer all the image data from the archive server 30 to the print server 44.

PRINT FUNCTION ANALYSIS: Here the retrieval time for a total of 18,000 accounts with 20 checks each would come to 5.3 hours. SITUATION—OF THREE PAGES PRINTED PER ACCOUNT: this would require the total printing of 54,000 pages and using one printer, this would take 12.5 hours. However, the total clock hours required for the printing would involve 5.3 hours for retrieval time, plus 12.5 hours for printing the 54,000 pages, plus the 0.5 hours to sort the 18,000 ac- counts. This would result in a total of 18.3 hours.

Likewise, for "commercial" accounts in this second configuration, the total print time would take 8.6 hours, and the "reconciliation" accounts would take a total print time of 41.6 minutes. SITUATION—OF FOUR PAGES OF PRINTING REQUIRED PER ACCOUNT: here the total number of pages required to be printed would be 72,000 pages, and using one printer, this would take 16.6 hours. However, with two printers, this would take 8.33 hours; with three printers, this would take 5.53 hours; and with four printers, this would take 4.15 hours. The total clock hours required for printing the four-page-per-account situation would involve a total of 22.7 hours.

For commercial accounts requiring four pages per account, this would require the printing of 54,000 pages and the total print time would take 11.57 hours.

Likewise, for the reconciliation accounting using a four-page account print out, the total number of pages would be 4,000, and the total print time would be 55.55 minutes.

FIG. 4 shows how the SRM document image data in 10i are transferred to the archive server 30 which has addi¬ tional storage on a magnetic disk 30_d. The archive server can transfer the data to the LMSI drives and platters 57 and 58 which can hold 30-days worth of data. Data which is to be held for longer than 30 days would be transferred to the 120 platters shown in the jukebox 50_j.

In FIG. 4, the archive server 30 has a small computer systems interface connection (SCSI) to the print server 44 which has an auxiliary magnetic disk 44_d. Attached to the print server is an image workstation 12 through which an operator can access image data for dis¬ play. The print server 44 provides and transfers its data to the printer 46 and 48 for eventual printing of the required account information.

FIG. 5 indicates the third format or design con¬ figuration for the storage/retrieval and printout system involved herein. This third design configuration involves the storing of the ON-US images and other document images "separately" in the jukebox platters 50_p, FIG. 5.

As seen in FIG. 5, the jukebox 50_j includes two optical storage drives 52 and 54 for storage operations and a third operating drive 56 for retrieval operations. In FIG. 5, the jukebox platters 50_p are shown numbered from platter 1 to platter 120. The-archive server 30 connects to the storage drives 52, 54 and the retrieval drives 56. The platters, in addition to being connected to the print server 44, provide output to two image printers, printers 46 and 48.

The storing of documents in the archive subsystem media 50 of FIG. 5 may be done as follows: (a) The checks are segregated (by code number to identify the group from MICR data) into three groups after the amount entry and the date of correction is done in the ICPS (Image Check Processing System) applications. This is done by the print server software in 44. The three groups into which the checks are sorted are: (i) ON-US items (personal and commercial accounts); (ii) Reconciliation account items; (iϋ) Transit items (items from other banks which are passing through the local bank);

(b) The complete images of all documents will be sorted in the archive subsystem 50 shown in FIG. 5, within jukebox 50_j, in the following order:

(1) ON-US items will be stored in the jukebox via one of the drive units such as 52. The ON-US items are separated from the other items and are stored only in the first 28 optical disk platters which involve the image data only required for a period of 30 days. A total number of 28 disks (60 percent of 46) are required as was previously indicated in the first "worst case" designed configuration. This situation permits increase in performance, as a lesser number of platters are required to handle the sorting of check images. (2) Reconciliationaccount items and transit items are stored in the jukebox via a second drive such as drive 54, FIG. 5.

The remaining 40 percent of information data (these are non "ON-US" items such as items related to other outside banks) will be stored in platters numbered 29 through 46 of the jukebox 50_j.

PERFORMANCE ANALYSIS FOR THIRD DESIGN CONFIGURATION: Here the front and back images are located in one file. The following assumptions are made:

1. ON-US statement printing is done at night.

2. The window time for printing is 10 to 12 hours.

3. Printing is done in off-business hours and in a batch mode.

4. The images of the "front" side of the check are printed in actual statements given to the customer.

5. The printer speed is maintained at 90 PPM (pages per minute).

6. The time required to change the optical platters is 6 seconds.

7. The capacity of each of the optical platters is 10 GB. 8. The "front" images are stored together in the same files in the captured order. The front images only are retrieved for printing purposes while the back images are held in storage.

The following Table VII indicates the specifications projected, for the third design alternative of FIG. 5, which are based on a CYGNET 1800 Jukebox with Hitachi drive.

Thus based on the above "forthcoming" specifica¬ tions, it is possible to estimate the following performance figures.

PERFORMANCE ANALYSIS: The total storage required is for 22 days (a banking month) and this would require 436.36 GB. The total ON-US item storage for one month is 60 percent of this which comes to 261.81 GB. The total number of optical platters 50_p (FIG. 5), required for one month would be 27 optical platters (using a 95 percent effective rate and assuming the effective usage of each platter is 95 per- cent).

The system will be accessing 54 platters (worst case situation) for every single daily cycle of the state¬ ment printing.

The average image size (for the front and back) is 27.5 KB, while the average image size for statement print¬ ing of the "front" only would be 17 KB. The total number of images stored in one platter (using the average of 17 KB) would be 345,454, while the total number of images stored on "one side" of the platter would be 172,727. The total number of ON-US images on one platter 50_p would be 345,454 while the average of ON-US images per daily cycle per platter would come to 15,702. (This is 345,454 divided by 22 banking days equal to 15,702.) This would average out to 7,851 ON-US images per daily print cycle on one side of the platter. The ratio of the total number of images to the number of images retrieved would be 22.0 per platter. This indicates that approximately one in every 22 images will be retrieved. In order to retrieve one image, on the average, it is necessary to move 22 images, wait for the latency period, and read the front image. This would involve a total of 5 milliseconds plus 25 milliseconds plus 21.25 milliseconds which comes to 51.25 milliseconds per image read. This means that the system moves 19.5 images per second from the disk 44_d to the printers 46, 48.

In order to retrieve images from one side of a platter (holding 7,851 images), it is necessary to drive, seat, and spin-up the drive in addition to spin-down the drive, plus the read time,which leads to an estimate of 0.0516 seconds per image or an average of 51.6 milliseconds per image (retrieval time from one side of a platter).

For customer account statements, the retrieval would involve 18,000 accounts X 20 checks per account X 0.0516 seconds per image which would entail 5.1 hours.

For commercial accounts, the time to retrieve one document image would average out to 61.25 milliseconds per image. This means that 16.32 images per second are re- trieved. Now to retrieve images from only one side of the platter (4,710 images), the time involved would be 61.5 milliseconds per retrieval of an image, on the average. Thus, the total time to retrieve 12,500 commercial accounts each having 20 checks and at the rate of 61.5 milliseconds per image—this would come to 4.2 hours.

For the reconciliation accounts, the process of retrieving images from one side of the platter (1000/3 equals 333 images) and where the total number of platters per cycle (equals 28/22) is 1.27 and involves 3 sides to be accessed, this would come to a total of 23.47 seconds for retrieving 333 images—which comes to 0.0704 seconds per image. Thus, the total time required to retrieve, on a daily basis, 1,000 reconciliation accounts would come to 1,000 accounts X 20 items per account X 0.0704 seconds which comes to 0.39 hours.

The time required to transfer all the images from the archive server 30 to the print server 44 is negligible and not counted. Thus no additional time is required for images to be transferred to the print server since they are transferred to the print server in the same 5.1-hour retrieval window used for the archives server (during customer account retrieval) .

Thus, the time to sort the 18,000 accounts would be only one-half hour.

PERFORMANCE SUMMARY

DESIGN NO. 1 WORST CASE CONFIGURATION (See FIG. 2) Case 1: Average 3 pages including text TIPTR = Image Printer!

Account Type Per Cycle Daily

Consumer

Commercial

Reconciliation

Case 2: Average 4 pages including text

Consumer 22.7 Commercial 16.67 Reconciliation 1.47

DESIGN NO. 2 USING LMSI DRIVE CONFIGURATION (See Fig. 4)

Case 1: Average 3 pages including text

Print Print Print Print

Account Window Window window Window

Type with with with with

Per Cycle 1IPTR 2 IPTR 3 IPTR 4 IPTR

Daily in Hours in Hours in Hours in Hours

Case 2: Average 4 pages including text

Consumer 22.4 Commercial 16.37 Reconciliation 1.39

DESIGN NO. 3 STORING ON-US AND TRANSIT ITEMS SEPARATE IN THE JUKEBOX (See FIG. 5)

ase No. 1: Avera e 3 a es includin text

Described herein has been a versatile and flexible document image storage, retrieval and print system suitable for users involved with massive amounts of transactions which have to be stored, retrieved, displayed, corrected and amended, and printed out on a regular basis. One typical archival storage and retrieval system described herein can store, for example, 400,000 accounts (mid-size bank) and on each day of the banking month (of 22 days) sort the stored image data from 18,000 accounts, retrieve them, and then print them in multiple-page state- ments for each account. Each account will manage trans¬ actions of checks involved^' in each account for over the last 30-day period. This-sorting, retrieving and printing can be accomplished with time frames such as: For one printer: the complete cycle would take 18.6 hours per day for a three-page statement, and 22.7 hours per day for a four-page statement; with two printers, the complete transaction would take only 12.35 hours for a three-page statement and 14.4 hours for a four-page statement; with four printers, the entire cycle would be accom¬ plished in 9.25 hours per day for a three-page state- ment and 10.25 hours per day for a four-page state¬ ment.

A major advantage of this system is that formerly used sorting methods would have required 10-20 hours alone just for the sorting while the present system requires one-half hour or less for the sorting. This is due to the fact that no check images are sorted, but rather only the index numbers of the images are sorted, thus saving long, drawn- out time periods such as were required for the old sorting systems. The Print Index Files are short, fixed records and very easy to sort as compared with older systems which have variable records and involve large database records for each check image.

While the described system is capable of variable configurations, it should be understood that the invention is encompassed by the following claims:

Claims

WHAT IS CLAIMED IS:

1. In a document processing system for capturing and storing digital images of documents being processed, and for retrieving and printing copies of selected documents, a method of organizing digital data and printing out humanly recognizable statements, comprising the steps of:

(a) converting images of captured original docu¬ ments into digital image data for storage on optical platters;

(b) correlating locational optical platter data of each one of said document digital image data with identification data associated with each document;

(c) sorting said correlated locational and iden¬ tification data into a sequential index list;

(d) printing out copies, in index list sequence, of selected document digital image data in humanly recognizable form.

2. In a document processing system for capturing and storing digital images of documents being processed, and for retrieving and printing copies of selected documents, a method of organizing digital data and printing out humanly recognizable statements, comprising the steps of:

(a) converting images of captured documents into digital image data;

(b) storing said digital image data on optical platters in a jukebox unit; (c) storing, in an archive server means, loca¬ tional data of platter number and record number for each document digital image data to form an Archive Server Index File;

(d) storing information-and-identification indicia for each document processed to form an Initial Index File;

(e) merging said Initial Index File with said Archive Server Index File to form a Master Print Index File which correlates the location of each document's digital image datawith its information-and-identification indicia;

(f) sorting said Master Print Index File to develop a Sequential-list Index File based on said information-and-identification indicia of each document processes;

(g) retrieving a selected portion of said document digital image data based on said Sequential-list Index File for storage in a print server means;

(h) printing out said selected portion of said document digital image data to provide humanly recognizable copies of original documents that were processed.

3. A storage/retrieval and printout system for copying original documents having indicia of capture date and customer account number, the combination comprising:

(a) first means for converting images of docu- ments into digital image data;

(b) second means for long-term optical platter storage of each of said document digital image data at a specific platter number and record number location to provide location- data;

(c) third means for correlating said identifica¬ tion-data of each document with its loca¬ tion-data to form a Master Print Index File;

(d) fourth means for sorting and storing said Master Print Index File in a numerical sequence of said document identification data;

(e) fifth means for printing copies of said original documents of said sorted numerical sequence of said Master Print Index File.

4. The combination of claim 3 wherein said second means includes:

(a) jukebox means having multiple optical platters for storage of said digital image data.

5. The combination of claim 4 wherein said jukebox means includes:

(a) first drive means for storing said digital image data on said optical platters; (b) second drive means for retrieving said digital image data from said optical plat¬ ters.

6. The combination of claim 3 wherein said fourth means includes:

(a) print server means for developing a sequen¬ tially sorted Print Index File holding a fractional portion of 1/22 of the total document image data accumulated over a 22- day work period.

7. The combination of claim 6 wherein said print server means includes:

(a) print server magnetic disk means for tem¬ porary storage of said fractional portion of document image data.

8. The combination of claim 3 wherein said third means includes:

(a) correlation means including:

(al) system disk means for storing an Initial Index File of information- identification data for each document processes; (a2) an Archive Server File which holds the platter number-record number location for each document image data stored in said optical platters;

(a3) merging means to combine said

Initial Index File with said Archive Server File to develop a

Master Print Index File, for transfer to said fourth means.

9. In a document image capture and storage system, which transforms original documents into digital document image data, an archival storage, retrieval and printing subsystem comprising: (a) archival storage means for long-term storage of digital document image data; (b) retrieval means for retrieval of selected items of said stored digital document image data; (c) temporary storage means for temporary storage of selected 1/N portions of said selected items where N represents the total number of document images stored during a prescribed time period; (d) sorting means for organizing said selected

1/N portions into a sequentially ordered file; (e) printing means for printing copies on a daily basis of digital image 1/N portions listed in said sequentially ordered file.

10. The subsystem of claim 9 wherein said archival storage means includes:

(a) optical drive and platter means for long- term storage of said digital image data; (b) archival server means for transmitting digital image data to said optical drive and platter means and also for receiving se¬ lected digital image data retrieved from said platter means for transmittal to a print server means.

11. The subsystem of claim 10 wherein said print server means includes:

(a) means for receiving said selected retrieved digital image data and sorting said digital image data into a sequential format, said means including:

(al) printer means for taking said sorted digital image data and printing it out to reproduce said original document.

12. The subsystem of claim 9 wherein said sorting means includes software means for correlating each digital document image data item with a location in said archival storage means to form a sorted Master Print Index File in an ordered sequence for said printing means.

13. A storage/retrieval and statement printout system for document digital images stored on optical platter media, said system comprising:

(a) system magnetic disk means for storing information-and-identification-data as¬ sociated with each document processed, to form an Initial Index File;

(b) archive server means for maintaining loca¬ tion-data, in an Archive Server Index File, which specifies the optical platter number and record number of each document's digital image data location, said archive server means including:

(bl) merger means for correlating said Initial Index File and said

Archive Server Index File into a Master Print Index File; (b2) means to transfer said Master Print Index File to a print server means;

(c) said print server means for temporarily storing said Master Print Index File and including:

(cl) sorting means for sequentially listing said Master Print Index

File into a Numerical Sequence Index File;

(c2) retrieval means for storing 1/N items of document digital image data in the sequential listing order of said Numerical Sequence Index File where 1/N is a frac¬ tional portion of the total (N) number of document images stored; 35 (c3) print means for printing humanly readable copies of said document digital image data for document items in said Numerical Sequence Index File.

14. The system of claim 13 wherein said print means prints document copies, on a daily basis, of 1/22 of the total number (N) of documents processed and stored during a 22-day work period.

15. In a document processing and imaging system, a method for storing, retrieving and printing copies of docu¬ ments comprising the steps of:

(a) placing digital document image data onto optical platters for archival storage;

(b) listing identification indicia for each document with its digital image location at a platter number and record number;

(c) retrieving document digital image data for a 30-day (monthly) period;

(d) placing said retrieved document digital image data on an archive server magnetic disk;

(e) organizing, for a fractional portion of monthly document image data, a daily Master

Print Index File which correlates each document digital image data item with a platter and record number;

(f) sorting said Master Print Index File to provide a sequentially dated list of docu¬ ment items indexed for use in printing;

(g) printing daily a group of document copies representing said digital image data listed in said sorted Master Print Index File.

16. A system for original document image capture and storage in digital document image form, said system com¬ prising:

(a) document processor means for capturing an electronic image of an original document and converting it to optical digital image data;

(b) storage/retrieval means for reception of and temporary storage of said digital image data; (c) archival subsystem means for transferring said digital image data from said storage/- retrieval means to a long-term archival storage means, said archival subsystem means including: (cl) means to retrieve on a daily cycle, selected elements of said digital image data to eventually cover a specified period of time, such as an Q-day cycle of business days, where Q might be set for a monthly period of 20-25 days, said selected elements including data representing (i) date of capture of original document; (ii) customer account number; (iii) sequential document number (iv) numeric financial quantity involved in each document selected; (c2) means to transfer and store a one-day- portion of said selected elements of said digital image data into a printing subsystem means; (d) said printing subsystem means for sorting on a daily basis, by specified indicia, said selected elements of said digital image data

35 and enabling a printing means to print, for each account number, an itemized statement of financial document transactions covering the period of a given Q-day cycle.

17. The system of claim 16 wherein said archival subsystem means includes:

(a) temporary storage means for holding 2 Q days worth of data on said selected elements; (b) means for retrieving, from said temporary storage means, 1/Q days worth of data, for use of said printing subsystem means.

18. The system of claim 16 wherein said storage/- retrieval means includes:

(a) means for converting said optical image data to magnetic image data for temporary storage and then for transfer to said archive subsystem means.

19. The system of claim 16 wherein said long-term archival storage means includes:

(a) jukebox means holding a plurality of optical disk platter for storage of digital image data of original documents.

20. The system of claim 19 which includes:

(a) storage management archival server means for providing data transfer and communication to/from said jukebox means and an archival library storage means, said archival server means including;

(al) means for merging identified platter and record numbers for each one of said selected elements; (b) said archival library storage means for providing multiple years storage of digital image data beyond that capable by said jukebox means.

21. The system of claim 19 wherein said storage/- retrieval means transmits Q days worth of selected elements for data accumulation in said archive subsystem means and said jukebox means while said storage management archival server means includes:

(a) means for transferring only 1/Q worth of said selected elements accumulated to said printing subsystem means for printout of said itemized statements during a daily print cycle.

22. The system of claim 16 wherein said archival subsystem means includes:

(a) means to retrieve, accumulate and store selected elements of digital image data selected for retrieval which encompasses a specified period of time to enable a one- day print cycle;

(b) means to create a Master Print Index File which correlates each retrieved selected element of document image data with its location position in said long-term archival storage means;

(c) means to transfer said Master Print Index File to said printing subsystem means.

23. The system of claim 16 wherein said printing subsystem means for sorting includes:

(a) means for arranging said Master Print Index

File into a sequential order determined by the numerical value of said specified indicia on each,document.

24. The system of claim 16 wherein said document processor means includes:

(a) host computer means for controlling said document capture and including: (al) means for retrieving and storing document item data (MICR data) including document capture date, account number, bank ID number and monetary amount;

(a2) means for supplying said document item data to said archival subsystem means to enable said archival subsystem means to merge the document item data with the associated location of its digital image data in said long-term archival storage means to form a Master Print Index File in said archival subsystem means.

25. The system of claim 24 wherein said printing subsystem means includes:

(a) means to receive and sort said Master Print

Index File into a specified sequential order.

26. The system of claim 17 which includes:

(a) power encoder means for placing additional MICR information indicia on selected docu¬ ments being processed.

27. The system of claim 16 wherein said archival subsystem means includes:

(a) interface means to transmit said selected digital image data to a remotely located print subsystem.

28. A storage/retrieval and printing system for capturing digital image data and information data from documents being scanned in an image document processor and said data being managed by a host computer for storage in storage subsystems and for retrieval and utilization by workstation subsystems, said storage/retrieval/printing system comprising:

(a) storage/retrieval module means coupled to said image document processor and to said host computer;

(b) line controller means in said host computer and said storage/retrieval module means coupled by bus means for managing and controlling said storage/retrieval module means;

(c) imaging module means in said image document processor;

(d) point-to-point link controller means in said image module means and said storage/- retrieval module means coupled by bus means for supplying said image data and said information data directly to said storage/- retrieval module means;

(e) archival subsystem means for receiving and storing said image data and information data in a long-term archival storage means, said archive subsystem including:

(el) means to retrieve selected ones of said digital image data and information data encompassing a specified short-term period; (e2) means to transfer data in said short- term period of said digital image and information data to a printing sub- system means; (f) printing subsystem means for sorting said short-term period of digital image and information data into a prescribed sequen¬ tial order and including: (fl) printer means for printing copies of said scanned documents.

29. The storage/retrieval/printing system as set forth in claim 28 wherein said storage/retrieval module means includes a plurality of storage/retrieval modules each being coupled to said image document processor via said point-to-point link controllers.

30. A storage/retrieval printing system as set forth in claim 29 wherein each of said plurality of storage/- retrieval modules is coupled to said host computer via line controllers.

31. A method for storage and retrieval of document image data from optical media for sorting the printout of monthly checking account statements, comprising the steps of: (a) processing financial check documents to convert document information into digital image information; (b) storing said digital image information on optical platters in a jukebox unit; (c) creating, via an image data storage disk, an

"initial index file" for deposit into an archive server means having fields of data representing:

(i) date of capture of checking document; (ii) customer account number;

(iii) check number; (iv) amount of check;

(d) merging, in said archive server means, two additional information fields to the said initial index file so that each of said check numbers is associated with a: (i) platter number in said jukebox unit; (ii) a record number on the identified platter; thus to form a Master Print Index File in said archive server means;

(e) accumulating data into said Master Print Index File in said archive server means, such that all check documents received in optical digital storage during a 30-day monthly period will include the said four fields of the "initial index file" plus the added platter and record number data; (f) copying said Master Print Index File to a 35 print server means and transferring said 30- day accumulation of Master Print Index File data into said print server means and its magnetic buffer print disk;

(g) sorting, via print server software, said 40 Master Print Index File to sequentially arrange said Master Print Index File accord¬ ing to the customer account number ascending from lower account numbers to higher account numbers;

45 (h) transferring the image data retrieved from the optical platter jukebox unit to said print server magnetic buffer print disk according to said sorted Master Print Index File;

50 (i) accessing, via software, said print server means and print disk by means of software to deliver co-related, digital image data to a printer means; (j) printing out, during each working day of the

55 month, a number of checking account state¬ ments equal to 1/22 of the total number (N) of customer accounts.