WO1999005632A1 - System for enterprise-wide work flow automation - Google Patents

System for enterprise-wide work flow automation Download PDF

Info

Publication number
WO1999005632A1
WO1999005632A1 PCT/US1998/015565 US9815565W WO9905632A1 WO 1999005632 A1 WO1999005632 A1 WO 1999005632A1 US 9815565 W US9815565 W US 9815565W WO 9905632 A1 WO9905632 A1 WO 9905632A1
Authority
WO
WIPO (PCT)
Prior art keywords
work
work flow
work object
partition
flow system
Prior art date
Application number
PCT/US1998/015565
Other languages
French (fr)
Inventor
Luen Kimball Poindexter
Glenn R. Seidman
Stephen R. Timm
Bruce A. Waddington
Original Assignee
Filenet Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Filenet Corporation filed Critical Filenet Corporation
Priority to CA002297065A priority Critical patent/CA2297065A1/en
Priority to AU86649/98A priority patent/AU8664998A/en
Priority to EP98938035A priority patent/EP0996916A1/en
Publication of WO1999005632A1 publication Critical patent/WO1999005632A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management

Definitions

  • the present invention relates generally to systems for automating document processing in a distributed computing environment. Specifically, the present invention relates to methods and apparatus for implementing work flow systems on an enterprise-wide scale.
  • Paperwork is a fact of business life. To schedule a vacation or purchase office supplies a form must be filled out and processed in accordance with appropriate procedures, including receiving authorizations by various personnel. Similarly, information and data must be disseminated throughout a business organization in the form of sales reports, accounting projections, market surveys, and the l ke. In some organizations, elaborate written guidelines have been developed to specify what paperwork gets sent to whom and under what conditions.
  • the "paperless office” was conceived as a way to combat the ever increasing volume of paperwork by replacing paper forms with electronic documents stored in a computer.
  • "Work flow” systems such as the WorkFlo ® Business System and Visual WorkFlo ® sold by FileNet Corporation, Costa Mesa, California, provide the means to create the paperless office by substituting computer based objects for their paper- based counterparts.
  • Such work flow systems provide distributed processing and distribution of data and information in accordance with procedures specified by the business analyst.
  • FileNet' s WorkFlo ® Business System provides a queue-based system for use in a client/server architecture in which objects are sequentially processed to accomplish a business procedure by scripts stored at the client stations.
  • FileNet' s Visual WorkFlo ® enables the business procedure to be accomplished using work objects that are processed at client workstations in accordance with centrally stored Instruction Sheets that specify the steps to be performed to accomplish the business procedure.
  • Previously known work flow systems have been successful in automating document management for organizations located at a single geographic site or area. However, these work flow systems are not presently scalable from a system that serves a single site to a system that serves an organization spanning multiple geographically separate sites. In view of the foregoing, it would be desirable to provide a work flow system that may be scaled up from a single site system to a multi-site system.
  • Previously known work flow systems also may become excessively burdensome to administer and maintain when the number of users becomes relatively large. It would therefore be desirable to provide methods and apparatus for dividing a relatively large work flow system into multiple smaller cooperating partitions. Moreover, to provide seamless operation of the work flow system, it would be desirable to provide methods and apparatus for allowing intercommunication of multiple work flows wherein the communications technology is transparent to the work flow system.
  • a work flow system may be partitioned into multiple self-contained work flows. Each work flow may advertise certain services for export to other work flows.
  • a gateway mechanism is provided that enables a work flow to seamlessly span multiple computer networks.
  • a partitioned queue may be provided for efficiently distributing data objects to work flows located on remote networks.
  • FIG. 1 depicts an exemplary local area network (LAN) suitable for use as a component of a work flow system constructed in accordance with the present invention
  • LAN local area network
  • FIG. 2 is a schematic illustration of a previously known single site, single partition work flow
  • FIG. 3 is an illustrative portion of an Instruction Sheet suitable for use in a work flow system constructed in accordance with the present invention
  • FIG. 4 is a diagram of a single site, multiple partition work flow
  • FIG. 5 is a schematic of a single site, multiple partition work flow in accordance with the principles of the present invention
  • FIG. 6 depicts an exemplary enterprise-wide, global scale computer network
  • FIG. 7 is a schematic representation of a single work flow spanning multiple physical sites
  • FIG. 8 is an exemplary partition table arranged in accordance with the principles of the present invention
  • FIG. 9 illustrates the use of a load balancing technique to provide a virtual queue in an enterprise-wide work flow system
  • FIG. 10 is an illustrative work object transmission sequence configured in accordance with the present invention. Detailed Description of the Invention
  • a work flow system typically comprises a LAN, such as LAN 10 schematically represented in FIG. 1, and software that controls the flow of information through the LAN.
  • LAN 10 comprises workstations 14 and server 16.
  • Cable 12 which may be coaxial cable, is routed throughout a site, such as an office, building, or campus, and is coupled to workstations 14 and server 16.
  • LAN 10 provides a means for workstations 14 and server 16 to communicate with each other according to various LAN protocols and technologies, such as, for example, an Ethernet LAN. It is to be understood, however, that the principles disclosed herein are not limited to a particular LAN technology or protocol and may be practiced on other than Ethernet based networks.
  • work flow system denotes a combination of hardware and software that makes up a system to automate a business procedure, and includes work objects that are processed by the system and Instruction Sheets that specify the sequence of steps in the process.
  • work flow refers to the run-time software components and data that control the flow of information through the underlying distributed processing system.
  • a work flow system preferably employs an object-oriented model of the information and communication flow within an organization.
  • object-oriented techniques and tools such as for example, the above-mentioned Visual WorkFlo ® product sold by the assignee of the present invention.
  • object-oriented programming techniques such as for example, the above-mentioned Visual WorkFlo ® product sold by the assignee of the present invention.
  • the software portion of a work flow system is preferably written in an object-oriented programming language.
  • the C++ programming language may be advantageously used to construct a work flow system in accordance with the principles of the present invention.
  • each class of object is represented by a C++ class.
  • the C++ class definitions encapsulate data attributes belonging to the class objects and methods or functions for working on those objects.
  • the Java language developed by Sun Microsystems, Inc., Mountain View, California is another object-oriented, platform independent programming language that may be advantageously employed in practicing the various features of the present invention.
  • a class is an abstract representation of an entity in a work flow system, including data contained within the entity and methods of accessing or manipulating that data.
  • An object is a specific instantiation of the entity described by a class definition. Accordingly, a work object is an instance of a work object class; a work performer is an instance of a work performer class; a work queue is an instance of a work queue class; and a work flow object is an instance of a work flow class.
  • a work object class defines a representation of an entity to be processed within a work flow system. For example, a work object class may define the representation of a process or transaction, such as approving a loan or making a credit card purchase.
  • a work object class may include various kinds of data and may also reference other documents such as word processing documents, data files, responses to database queries, images, videos, sounds, or any other entity that can be stored and processed by a computer.
  • a work performer class defines a service or set of services to be applied to a work object.
  • a work performer class may represent a task requiring human intervention, such as displaying some information contained within a work object to a user.
  • a work performer class may represent an automated task, such as updating a database with information contained within a work object.
  • a work queue class defines a queue for work objects waiting to be serviced by a work performer.
  • each work performer is associated with a specific instance of a work queue class.
  • a work flow system must also include run- time information about the objects within the system. Specifically, a work flow system must be able to determine which classes exist in the work flow system, how many of each class of object have been instantiated, and where each object is located. This information is provided by one or more globally visible configuration data structures. The configuration data structures are initialized at work flow system startup to include the necessary configuration information. Subsequently, the work flow system updates the configuration data structures as necessary when new objects are created or when old ones are destroyed.
  • a work flow system comprises three major components: (1) the definitions of all classes that may be instantiated within the work flow system; (2) configuration information describing specific instantiations of those classes and their relation to one another; and (3) a run-time engine.
  • Work flow system 20 includes work object 22, work performers 26a, 26b, 27, and 28, and work performer queues 24, 25, and 29.
  • Work object 22 resides as a database object in a server, and may be copied to a workstation where some of its data fields are operated .upon by the work performers; the revised work object preferably may then be copied back to the server upon completion of the processing step.
  • Work performers 26a and 26b each represent an instance of the same work performer class (i.e., either can perform a specified step or steps of a business procedure) .
  • Work flow system 20 also includes a run-time engine that coordinates, monitors, and controls the processing of work objects in the work flow system.
  • the run-time engine is responsible for creating work objects, monitoring and updating their status as they are serviced by work performers, executing instruction sheets and dispatching work objects, or data extracted therefrom, accordingly.
  • the run-time engine also monitors completion of the steps of the business procedure, and destroys an object when it is no longer needed.
  • the concept of the instruction sheet is employed in the above-mentioned Visual WorkFlo ® product and comprises a linked-list of the steps to be performed to accomplish the business procedure.
  • the instruction sheet may be graphically represented as a series of steps of the business procedure to be performed, as shown in FIG. 3.
  • the instruction sheet is converted to a linked-list of centrally-stored work orders .
  • an instruction sheet may comprise any of a number of forms.
  • an instruction sheet could be written in a suitable computer programming language, including general purpose and special purpose scripting language. Languages such as C++, Java, or BASIC may be suitable languages for developing work object class instruction sheets.
  • an instruction sheet is programmed graphically, wherein icons represent various programming steps, and arrows or lines represent the flow of execution.
  • an instruction sheet 30 may embody procedural intelligence, such as distribution and approval policies of an organization.
  • instruction sheet 30 reflects policies regarding the levels of approval required for a business loan.
  • Input icon 31 represents a step of obtaining pertinent loan information from an applicant, and may consist of a data entry operator transferring data from a handwritten application. The application information is stored in data fields contained within the work object. A basic check, indicated by assignment icon 32, is then performed to determine whether certain minimum criteria required for loan approval are met, and the result is stored in a data field of the work object.
  • branch icon 33 may test to determine whether the application passes certain minimum criteria and the loan amount exceeds $10,000. If both conditions are met, execution branches to the step indicated by input icon 34; otherwise, execution continues to the step indicated by branch icon 37.
  • a loan officer reviews the application. Since this branch is only taken for applications over $10,000, branch icon 33 implements a policy that loans meeting certain minimum criteria and for amounts over $10, 000 require loan officer approval.
  • the step corresponding to branch icon 35 ensures that branch manager approval is obtained for loans exceeding $100,000.
  • Each of the icons of FIG. 3 represents an operation, service, or task, to be performed on loan work object 22.
  • the skilled artisan will recognize that the icons shown in FIG. 3 represent only a fraction of the useful icons and that icons representing other statements, expressions, and control structures may also be incorporated into a instruction sheet. Indeed, in a preferred embodiment of the present invention, new icons representing custom tasks and processes may be added.
  • each work object such as work object 22
  • a run-time version of one or more instruction sheets (a list of work orders) that indicates to the run-time engine, in linked-list fashion, the sequence of services required in processing the work object.
  • a work class also may have multiple instruction sheets associated with it, corresponding to multiple states or conditions of a work object.
  • a work class may have an associated instruction sheet that gets executed when a work object is first instantiated to ensure that the work object is properly initialized.
  • a second instruction sheet may provide instructions for the normal processing of an instance of the work class.
  • a third instruction sheet may specify actions to be taken when an instance of the work class is being deleted or the work flow system is being shut down, and, for example, may cause a work object to be archived or otherwise disposed of.
  • a work object dispatching sub-system of the work flow system run-time engine identifies the work queue associated with the desired work performer class and enqueues a pointer to the work object accordingly.
  • a work performer associated with the class becomes available, e.g., by completing a work order for a previous work object, the work performer polls its associated queue to determine if there is another work object in need of servicing.
  • a work object typically includes a number of data fields that hold various information about the work object and the data it contains.
  • a work flow system for use in a bank branch office may include classes of work objects representing various banking transactions such as account deposits and withdrawals, or loan applications.
  • a work object may, therefore, be a specific instance of a work object class representing business loan applications, and might include data fields such as amount and approval representing, for example, a desired loan amount and whether the loan has been approved, respectively.
  • the values of data fields are provided by the originator of the work object, such as by filling in a computer-based form, but may also be calculated by a work performer, retrieved from an on- line data base or automatically maintained by the work flow system run-time engine.
  • work performers 26a and 26b may each represent a task of filling out an application form.
  • a potential loan customer or bank employee may fill in a computer-based loan application form with relevant loan information, such as the amount and duration of the loan, as well as the customer' s credit information.
  • relevant loan information such as the amount and duration of the loan, as well as the customer' s credit information.
  • Other data such as the amount of each installment payment, may be calculated later by other work performers, such as work performer 27 in work flow system 20.
  • work object 22 comprising the loan form is submitted to the run-time engine of work flow system 20 and processed in accordance with the instruction sheet associated with the work object class of which work object 22 is an instance.
  • work object 22 is dispatched to work queue 24 which is associated with the class of work performers 26a and 26b.
  • work performer 26a When work performer 26a becomes available, it polls work queue 24 and determines that work object 22 is the next item requiring service. Accordingly, work performer 26a dequeues work object 22, obtains a copy of work object 22 from the server, or the relevant subset of data from within the work object, and begins processing it. However, if queue 24 is empty, work performer 26a enters an idle state and periodically polls queue 24 until another work object has been enqueued. Similarly, work performer 26b also services work objects from work queue 24.
  • Multiple work performers instantiated from the same work performer class may be included in a work flow system to improve system response times, or to provide redundancy for important or high volume services.
  • various algorithms or heuristics may be used to determine which work performer is to service a work object.
  • the work object is submitted to the work flow system run-time engine which determines further processing pursuant to run-time series of work orders for objects of that class.
  • Work flow system 20 of FIG. 2 is appropriate for a single work flow system that can encompass an entire organization without becoming unmanageable. As the work flow system increases in size, however, it may become difficult to design, administer, and maintain. Applicants have discovered it is desirable to be able to divide, or partition, a large work flow system into multiple smaller, cooperative, work flow systems, so that, for example, separate work flow systems may be implemented for each department of a corporation. While partitioning a large work flow system may ease the design and maintenance burden of the work flow system as a whole, the presence of multiple work flow systems presents other difficulties, for which the present invention provides solutions .
  • work flow system 40 may comprise a loan department procedure of a bank while work flow system 42 comprises an accounting department procedure of the same bank.
  • work objects 44a and 44b may each represent a purchase order for new computer equipment for their respective departments, which must be approved by the bank' s management information services (MIS) department. Because both work flow systems 40 and 42 require the task of obtaining MIS approval, each must include a work performer, such as 46a and 46b, that performs that task.
  • MIS management information services
  • modifying an object in one work flow system may cause objects in another work flow system to cease functioning properly.
  • the maintenance and administrative burden of a work flow system is reduced while reliability and security are enhanced by providing limited access across work flow system boundaries .
  • This is accomplished by allowing a work flow system to advertise specific services that may be accessed by external objects.
  • a special work performer class therefore is created that encapsulates those services and provides a protected, well-defined interface for accessing those services.
  • a C++ or Java class representing a work flow system is created wherein the publicly visible member functions provide access to those services that the work flow system advertises for export.
  • the exporting class definition therefore, appears to be a work performer class definition.
  • another work flow system desiring to access the advertised services need only include the work performer class definition of the exporting work flow system in its own list of object classes .
  • work flow systems 50 and 52 may be set up for the accounting departments of an organization. Work flow systems 50 and 52 may then export, or make public, various services, such as get_purchase_approval or update_accounts_receivable for use by other work flow systems.
  • a work performer class is created that is incorporated into other work flow systems.
  • An instance of a work flow system work performer class functions as a proxy for a work performer in the work flow system advertising the service.
  • the proxy work performer has access into the exporting work flow system through the mechanism of an agent work object.
  • work flow system 50 alternatively may belong to the research department of an organization, and work object 53 may correspond to a purchase request for new computer equipment. At some point in the processing of work object 53 within research department work flow system 50, it may be necessary to obtain accounting department approval of the purchase request. Rather than explicitly incorporate an accounting department process step in work flow system 50, work flow system 52 (corresponding to the accounting department of the organization) exports a service get_accountin ⁇ _approval . This service is incorporated into work flow system 50 via proxy work performer 54. When work object 53 requires the service get_accounting_approval, the work flow system run-time engine determines that the service is provided by proxy work performer 54 and dispatches work object 53 accordingly.
  • Proxy work performer 54 dequeues the pointer to work object 53 from work queue 55, obtains a copy of the entire work object (or a subset of its data) from the repository and determines that work object 53 is requesting service from work flow system 52. Proxy work performer 54 then causes the creation of agent work object 57 in work flow 52. Any required parameters of work object 53 are transferred to agent work object 57, the identity of the originating work flow and work object (i.e., work object 53 in work flow 51) are saved, and a pointer corresponding to agent work object 57 is enqueued to work queue 59 to await the first available service in work flow system 52 of the kind requested (e.g., the service provided by work performer 58). While agent work object 57 is being processed, originating work object 53 is locked to prevent inconsistencies between the values of data in the agent work object and originating work object 53.
  • agent work object 57 When agent work object 57 has finished executing the work orders associated with the exported service, a special terminate instruction is invoked. The terminate instruction causes the identity of the originating work flow and work object to be retrieved. Any return values may then be transferred back from agent work object 57 to originating work object 53. Finally, agent work object 57 is disposed of, the list of work orders associated with work object 53 is updated, and work object 53 is unlocked and returned to the run-time engine of work flow system 50 for further processing.
  • each work performer queue is visible to all work performers servicing that queue.
  • a global queue for a particular work performer class may be created and maintained on a globally visible network server.
  • enterprise network 60 comprises LANs 61a, 61b and 61c located at disparate sites around the globe.
  • a work flow spanning these sites should appear to the user as a single network, i.e., there should be little discernable difference to a user on LAN 61a between a work object being serviced by a work performer in LAN 61a and one being serviced by a work performed in LAN 61c.
  • Interlinked networks may be classified by type according to speed and availability.
  • speed refers to the rate at which data can transmitted over the link.
  • a T3 line can carry data at a rate of about 45 million bits per second, whereas a dial-up modem link is limited to about 28 thousand bits per second.
  • Availability refers to whether access to the link is permanent or only exists on a transient basis. For example, a leased line may be available continually whereas a satellite link may only be available during certain times of the day.
  • a transient link may require a mechanism to store a transmission and then forward the transmission when a link becomes available, and may introduce substantial delay in transferring data to a remote site.
  • LAN 61a may be connected to LAN 61b using dedicated line 62 or other means to provide continuous, high-speed, inter-LAN communications, in which case the connection between LANs 61a and 61b would be classified as both fast and permanent.
  • LAN 61a may be connected to LAN 61c on an as-needed basis by a dial-up type modem connection via link 64. Such a connection would be a slow and transient connection.
  • gateway work performers 72a and 72b are provided at either end of communication link 74 interconnecting work flow partitions 70a and 70b.
  • Gateway work performers 72a and 72b serve as local receivers for work objects or extracted data that are destined for remote work queues. For example, when work object 76 is destined for remote work queue 77, it is sent to gateway work performer 72a for transmission to the remote site. Local gateway work performer 72a provides an acknowledgment to the run-time engine of work flow system 70, thus allowing the run-time engine of work flow system 70 to provide another work object to work performer 73 for servicing. Gateway work performer 72a then transmits the work object to the remote site in a manner appropriate for the type of communication link. For example, if link 74 is a fast and permanent connection, gateway work performer 72a may immediately transmit the work object to gateway work performer 72b at the remote site.
  • gateway work performers 72a and 72b may be enhanced to provide additional capabilities such as gathering statistics for determining inter-site transit times, to aggregate multiple work objects or subsets of data destined for a remote site to enable transfer in a single transmission, performing data compression to reduce transmission times, or encrypting data to improve data security while transmitting over open lines, such as telephone lines or satellite links.
  • communications links other than fast, permanent links introduce other concerns.
  • typical dial-up analog connections are limited to transmission rates in the neighborhood of three thousand bytes per second. While digital connections using ISDN may increase the data rate by a factor of four over analog connections, the data rate is still very low compared to a rate of 10 million bits per second for a typical Ethernet LAN.
  • the dial-up link is always connected, additional time is expended in establishing the link.
  • a work performer exists on LAN 61c, of FIG. 6, while its class queue is located on LAN 61a, it may take several seconds for a dial-up link to be established each time work performer polls the queue.
  • a large organization may have many work performers and corresponding queues distributed across multiple LA s. Even when high speed dedicated digital links are employed, the overhead associated with polling a class queue on another LAN may negatively impact system throughput and performance.
  • a single global queue paradigm is retained through the use of a structure referred to herein as a partitioned logical queue .
  • each work performer class is associated with a logical queue.
  • a logical work queue provides the appearance of a single global queue, but internally it is divided into partitions, wherein each partition is associated with a physical queue.
  • a logical work queue provides an additional layer of abstraction between the work flow system runtime engine and the work performers in the work flow system.
  • a logical work queue accepts work objects destined for a work performer class queue and assigns the work object to one of the logical partitions of the logical queue based on various properties of the work object.
  • FIG. 8 represents screen display 80 of a partitioned logical queue as table 81, with each row 81a, 81b, . . . 81n of table 81 corresponding to a logical partition.
  • a first attribute of a queue partition is its partition expression 82, which may depend on information within data fields of the work object.
  • partition expression 82 for each partition queue 81a...81n is evaluated.
  • the work object is then assigned to a work queue based on the results of evaluating partition expressions 82. If only one partition expression is true, the work object is assigned to the corresponding queue partition.
  • various heuristics may be used to select a queue partition, for example, heuristics relating to throughput statistics of the various queues, locality, etc.
  • an exception is declared and an exception work order for the work object being enqueued is invoked by the work flow system run-time engine.
  • the logical queue associated with the Loan Officer class of work performers has three partitions, each represented by a row (81a to 81c) in table 81. If a work object to be enqueued has a loan amount greater than $203,500 partition expression 82 in the first partition 81a is evaluated as true and the work object is assigned to the Loan Officer queue of the first partition. However, if the loan amount is less than or equal to $203,500 the first partition equation is evaluated as false, and the partition expressions in the second and third rows, 81b and 81c, are true.
  • each partition, 81a...81n, of a partitioned queue has an associated weight 83 as shown in the second column of FIG. 8.
  • the work object is dispatched in accordance with a weighted probability distribution based on weights 83 assigned to each partition 81a...81c.
  • weights 83 assigned to each partition 81a...81c For example, in FIG. 8 partitions 81b and 81c have the same partition expression, but in partition 81b weight 83 has a value of 2, while in partition 81c weight 83 has a value of 1.
  • two- thirds of loan applications for less than $203,500 are enqueued to the queue partition associated with row 81b of the table, and the remaining one-third to the third queue partition (81c) .
  • partition expressions in a work performer queue partition table may be updated to reflect non-availability of a service (due to personnel vacation, etc.).
  • work objects may have associated with them information for identifying specific work performers which are to process the work object, thus overriding the normal work object dispatch function of a partitioned work performer queue.
  • these potential solutions may be undesirable due to increased administrative burden and/or processing overhead.
  • the foregoing problems are solved by periodically redistributing work objects from one partition of a work performer queue to another partition.
  • Various criteria such as elapsed time in a queue or queue size, may be specified to indicate when work objects should be redistributed. Redistribution is accomplished through the mechanism of a load balancing work object and load balancing work performers.
  • each queue partition has an associated load balancing work performer.
  • a load balancing work object travels in a circuit that visits all load balancing work performers associated with queue partitions having the same partition expression.
  • the load balancing work object carries with it information concerning work objects matching the corresponding partition expression.
  • When a load balancing work performer services the load balancing work object it determines the status of each work object in the work queue for the local work flow system and as well as information for those work objects carried by the load balancing work object.
  • the load balancing work performer Based on the status of the work objects, the load balancing work performer selectively transfers work object pointers from the load balancing work object to the queue of the local work flow system and/or from the queue to the load balancing work object. The load balancing work object is then dispatched to the next load balancing work performer in the circuit, which repeats the process just described.
  • work flow 90 is divided into partitions 90a, 90b, and 90c, including load balancing work performers 92a, 92b, and 92c, respectively. Let it be assumed that based on the partition expressions for partitions 90a to 90c, work object 94 is to be enqueued to work performer 96a in work flow 90a.
  • a pointer for work object 94 is, therefore, enqueued to load balancing work performer 92a for transfer to work flow 90b.
  • Load balancing work object 98 circulates between load balancing work performers 92a, 92b, and 92c. Information relating to work object 94 is inserted into load balancing work object 98 the next time it arrives at load balancing work performer 92a. Subsequently, when load balancing work object 98 arrives at load balancing work performer 92b, the pointer for work object 94 is removed from the load balancing work object and enqueued for work performer 96b. At the same time, load balancing work performer 92b may insert information for other objects (not shown) bound for partitions 90a and 90c into load balancing work object 98 to be carried back to their respective load balancing work performers.
  • load balancing work performer 92a may delete it from work performer queue 95 and insert a pointer for work object 94 into load balancing work object 98. Then either of load balancing work performers 92b or 92c, in response to the excessive age of work object 94, may transfer the information for work object 94 from load balancing work object 98 and enqueue a pointer for work object 94 to work performer queues 96b or 96c, respectively.
  • criteria other than length of time enqueued may be used to determine when the partition assigned to a work object should be overridden, such as, the length of the destination queue, the time of day, or a heuristic based on average service times at work performers 96a, 96b, and 96c.
  • a load balancing technique to virtualize a global queue may provide additional benefits such as dynamically balancing queue work loads, providing work performer redundancy, and ensuring all work objects are serviced in a reasonable length of time.
  • a source site may first obtain a record lock for the record being transmitted and, using a remote procedure call, may obtain a corresponding record lock at the destination site. The record is then transmitted. After the destination site acknowledges receipt of the transferred record, the source site lock is removed, and then the destination site lock is removed. However, if the source site were to shut down or fail between obtaining the source lock and releasing the destination lock recovery of the transaction may be difficult or impossible.
  • FIG. 10 illustrates transmission of work objects
  • the source marks the work object to be transmitted as completed, i.e., processing at the current site is finished.
  • the source site enqueues the work object and creates a marker record at the destination site (step 106) .
  • This operation is completed as an atomic step.
  • the source then deletes, backs up, or otherwise archives its copy of the work object, at step 108.
  • the source site work flow engine uses a remote procedure call to delete the marker record at the destination site. Note that once the work object is enqueued and the marker record created, the destination site may begin processing the work object. In contrast, when using a record or file lock, the destination site may not begin processing until the lock is removed.
  • the transmission system of the present invention avoids the use of file or record locking. Furthermore, the status of a work object can easily be determined at any time. For example, when a work flow system is restarted after a shutdown or failure, the presence of a marker record indicates that a work object was received, but that the source site may still need to delete its copy of the completed work object. Analogously, the presence of a work object marked complete indicates that it was in the process of being transmitted. By checking for the presence of a marker record at the destination, the source site may quickly ascertain whether the work object was successfully transmitted and may retransmit or delete the work object and marker record as appropriate.

Abstract

Methods and apparatus are provided for an enterprise-wide work flow system that may encompass multiple geographically separate sites. The sites may be either permanently or transiently linked. A single computer network may accommodate multiple work flow systems and a single work flow system may be distributed over multiple local area networks. The system maintains the paradigm of one global queue per service and provides for individual work flow systems to export services to one another in an enterprise.

Description

SYSTEM FOR ENTERPRISE-WIDE WORK FLOW AUTOMATION
Background of the Invention The present invention relates generally to systems for automating document processing in a distributed computing environment. Specifically, the present invention relates to methods and apparatus for implementing work flow systems on an enterprise-wide scale.
Paperwork is a fact of business life. To schedule a vacation or purchase office supplies a form must be filled out and processed in accordance with appropriate procedures, including receiving authorizations by various personnel. Similarly, information and data must be disseminated throughout a business organization in the form of sales reports, accounting projections, market surveys, and the l ke. In some organizations, elaborate written guidelines have been developed to specify what paperwork gets sent to whom and under what conditions.
Often paperwork must be transmitted to or processed by multiple organizational units of a company. For example, hiring a new research scientist may require approval of the research and development department, the accounting department, and the personnel department. In organizations, such as modern multi-national corporations, these organizational units may be located in different buildings, in different cities, or even in different countries.
The "paperless office" was conceived as a way to combat the ever increasing volume of paperwork by replacing paper forms with electronic documents stored in a computer. "Work flow" systems, such as the WorkFlo® Business System and Visual WorkFlo® sold by FileNet Corporation, Costa Mesa, California, provide the means to create the paperless office by substituting computer based objects for their paper- based counterparts. Such work flow systems provide distributed processing and distribution of data and information in accordance with procedures specified by the business analyst. FileNet' s WorkFlo® Business System provides a queue-based system for use in a client/server architecture in which objects are sequentially processed to accomplish a business procedure by scripts stored at the client stations. FileNet' s Visual WorkFlo® enables the business procedure to be accomplished using work objects that are processed at client workstations in accordance with centrally stored Instruction Sheets that specify the steps to be performed to accomplish the business procedure. Previously known work flow systems have been successful in automating document management for organizations located at a single geographic site or area. However, these work flow systems are not presently scalable from a system that serves a single site to a system that serves an organization spanning multiple geographically separate sites. In view of the foregoing, it would be desirable to provide a work flow system that may be scaled up from a single site system to a multi-site system.
Previously known work flow systems also may become excessively burdensome to administer and maintain when the number of users becomes relatively large. It would therefore be desirable to provide methods and apparatus for dividing a relatively large work flow system into multiple smaller cooperating partitions. Moreover, to provide seamless operation of the work flow system, it would be desirable to provide methods and apparatus for allowing intercommunication of multiple work flows wherein the communications technology is transparent to the work flow system.
In addition, the construction of previously known work flow systems may require detailed a priori analysis, so that the resulting work flow system does not experience bottlenecks that reduce overall throughput of the system. Accordingly, it would be desirable to provide methods and apparatus for dynamically balancing the flow of work through the system.
While previously known work flow systems generally involve only a single site, it would be desirable to provide methods and apparatus for connecting multiple work flows over a variety of communications links. Thus, it would be desirable to provide methods and apparatus for efficient and transparent inter-site communications while supporting many different connection mechanisms, in which the physical location of a work flow service is transparent to both the user and the work flow definition itself. Summary of the Invention
In view of the foregoing, it is an object of the present invention to provide a work flow system that is easily scaled up from serving a single site to serving multiple sites.
It is also an object of this invention to provide a system in which the physical location of a work flow service is transparent to both the user and the work flow definition itself. It is an additional object of the present invention to provide efficient and transparent inter- site communications while supporting many different connection mechanisms.
It is a further object of the invention to provide a system in which work flows may be grouped and partitioned.
These and other objects of the present invention are provided by a work flow system that may be partitioned into multiple self-contained work flows. Each work flow may advertise certain services for export to other work flows. In addition, a gateway mechanism is provided that enables a work flow to seamlessly span multiple computer networks. Optionally, a partitioned queue may be provided for efficiently distributing data objects to work flows located on remote networks.
Brief Description of the Drawings
The above and other objects and advantages of the invention will be apparent upon consideration of the following detailed description, taken in conjunction with the following accompanying drawings, in which like reference characters refer to like parts throughout, and in which:
FIG. 1 depicts an exemplary local area network (LAN) suitable for use as a component of a work flow system constructed in accordance with the present invention;
FIG. 2 is a schematic illustration of a previously known single site, single partition work flow;
FIG. 3 is an illustrative portion of an Instruction Sheet suitable for use in a work flow system constructed in accordance with the present invention;
FIG. 4 is a diagram of a single site, multiple partition work flow; FIG. 5 is a schematic of a single site, multiple partition work flow in accordance with the principles of the present invention;
FIG. 6 depicts an exemplary enterprise-wide, global scale computer network; FIG. 7 is a schematic representation of a single work flow spanning multiple physical sites;
FIG. 8 is an exemplary partition table arranged in accordance with the principles of the present invention; FIG. 9 illustrates the use of a load balancing technique to provide a virtual queue in an enterprise-wide work flow system; and
FIG. 10 is an illustrative work object transmission sequence configured in accordance with the present invention. Detailed Description of the Invention
A work flow system typically comprises a LAN, such as LAN 10 schematically represented in FIG. 1, and software that controls the flow of information through the LAN. LAN 10 comprises workstations 14 and server 16. Cable 12, which may be coaxial cable, is routed throughout a site, such as an office, building, or campus, and is coupled to workstations 14 and server 16. LAN 10 provides a means for workstations 14 and server 16 to communicate with each other according to various LAN protocols and technologies, such as, for example, an Ethernet LAN. It is to be understood, however, that the principles disclosed herein are not limited to a particular LAN technology or protocol and may be practiced on other than Ethernet based networks. As used herein, the term "work flow system" denotes a combination of hardware and software that makes up a system to automate a business procedure, and includes work objects that are processed by the system and Instruction Sheets that specify the sequence of steps in the process. The term "work flow" as used herein refers to the run-time software components and data that control the flow of information through the underlying distributed processing system. In accordance with the principles of the present invention, a work flow system preferably employs an object-oriented model of the information and communication flow within an organization. Although many techniques and tools may be used to implement a work flow system, it is preferable to implement a work flow system using object-oriented techniques and tools, such as for example, the above-mentioned Visual WorkFlo® product sold by the assignee of the present invention. This preference for use of object-oriented programming techniques in a preferred embodiment of the present invention is reflected throughout this disclosure by the use of object-oriented terminology, as defined hereinbelow.
Although any general purpose programming language may be used, the software portion of a work flow system is preferably written in an object-oriented programming language. The C++ programming language may be advantageously used to construct a work flow system in accordance with the principles of the present invention. In programming a work flow system, each class of object is represented by a C++ class. The C++ class definitions encapsulate data attributes belonging to the class objects and methods or functions for working on those objects. Alternatively, the Java language developed by Sun Microsystems, Inc., Mountain View, California, is another object-oriented, platform independent programming language that may be advantageously employed in practicing the various features of the present invention.
As used herein, a class is an abstract representation of an entity in a work flow system, including data contained within the entity and methods of accessing or manipulating that data. An object is a specific instantiation of the entity described by a class definition. Accordingly, a work object is an instance of a work object class; a work performer is an instance of a work performer class; a work queue is an instance of a work queue class; and a work flow object is an instance of a work flow class. A work object class defines a representation of an entity to be processed within a work flow system. For example, a work object class may define the representation of a process or transaction, such as approving a loan or making a credit card purchase. A work object class may include various kinds of data and may also reference other documents such as word processing documents, data files, responses to database queries, images, videos, sounds, or any other entity that can be stored and processed by a computer.
A work performer class defines a service or set of services to be applied to a work object. A work performer class may represent a task requiring human intervention, such as displaying some information contained within a work object to a user.
Alternatively, a work performer class may represent an automated task, such as updating a database with information contained within a work object.
A work queue class defines a queue for work objects waiting to be serviced by a work performer. Typically, each work performer is associated with a specific instance of a work queue class.
In addition to the foregoing class definitions, a work flow system must also include run- time information about the objects within the system. Specifically, a work flow system must be able to determine which classes exist in the work flow system, how many of each class of object have been instantiated, and where each object is located. This information is provided by one or more globally visible configuration data structures. The configuration data structures are initialized at work flow system startup to include the necessary configuration information. Subsequently, the work flow system updates the configuration data structures as necessary when new objects are created or when old ones are destroyed. Thus a work flow system comprises three major components: (1) the definitions of all classes that may be instantiated within the work flow system; (2) configuration information describing specific instantiations of those classes and their relation to one another; and (3) a run-time engine. Referring now to FIG. 2, an illustrative processing sequence of work object 22 in exemplary work flow system 20 is described. Work flow system 20 includes work object 22, work performers 26a, 26b, 27, and 28, and work performer queues 24, 25, and 29. Work object 22 resides as a database object in a server, and may be copied to a workstation where some of its data fields are operated .upon by the work performers; the revised work object preferably may then be copied back to the server upon completion of the processing step. Alternatively, only selected variables may be extracted from the work object and provided to the workstation, with updated variables and the results of the processing task being copied back to the server upon completion. Work performers 26a and 26b each represent an instance of the same work performer class (i.e., either can perform a specified step or steps of a business procedure) . Work flow system 20 also includes a run-time engine that coordinates, monitors, and controls the processing of work objects in the work flow system.
Specifically, the run-time engine is responsible for creating work objects, monitoring and updating their status as they are serviced by work performers, executing instruction sheets and dispatching work objects, or data extracted therefrom, accordingly. The run-time engine also monitors completion of the steps of the business procedure, and destroys an object when it is no longer needed.
The concept of the instruction sheet is employed in the above-mentioned Visual WorkFlo® product and comprises a linked-list of the steps to be performed to accomplish the business procedure. During authoring time, in which the work flow class is defined, the instruction sheet may be graphically represented as a series of steps of the business procedure to be performed, as shown in FIG. 3. At runtime, when the work flow system is instantiated, the instruction sheet is converted to a linked-list of centrally-stored work orders .
The skilled artisan will recognize that an instruction sheet may comprise any of a number of forms. For example, an instruction sheet could be written in a suitable computer programming language, including general purpose and special purpose scripting language. Languages such as C++, Java, or BASIC may be suitable languages for developing work object class instruction sheets. However, in a preferred embodiment of the present invention, an instruction sheet is programmed graphically, wherein icons represent various programming steps, and arrows or lines represent the flow of execution.
Referring to FIG. 3, an illustrative authoring form of an instruction sheet 30 is described. In addition to controlling the performance of various tasks, such as calculations or database lookup, an instruction sheet may embody procedural intelligence, such as distribution and approval policies of an organization. For example, instruction sheet 30 reflects policies regarding the levels of approval required for a business loan. Input icon 31 represents a step of obtaining pertinent loan information from an applicant, and may consist of a data entry operator transferring data from a handwritten application. The application information is stored in data fields contained within the work object. A basic check, indicated by assignment icon 32, is then performed to determine whether certain minimum criteria required for loan approval are met, and the result is stored in a data field of the work object.
A branch icon, such as icons 33, 35, and 37, provide an alternate execution path, depending upon the outcome of the test associated with each branch icon. For example, branch icon 33 may test to determine whether the application passes certain minimum criteria and the loan amount exceeds $10,000. If both conditions are met, execution branches to the step indicated by input icon 34; otherwise, execution continues to the step indicated by branch icon 37. At the step corresponding to input icon 34, a loan officer reviews the application. Since this branch is only taken for applications over $10,000, branch icon 33 implements a policy that loans meeting certain minimum criteria and for amounts over $10, 000 require loan officer approval. Similarly, the step corresponding to branch icon 35 ensures that branch manager approval is obtained for loans exceeding $100,000.
Each of the icons of FIG. 3 represents an operation, service, or task, to be performed on loan work object 22. The skilled artisan will recognize that the icons shown in FIG. 3 represent only a fraction of the useful icons and that icons representing other statements, expressions, and control structures may also be incorporated into a instruction sheet. Indeed, in a preferred embodiment of the present invention, new icons representing custom tasks and processes may be added.
Associated with each work object, such as work object 22, is a run-time version of one or more instruction sheets (a list of work orders) that indicates to the run-time engine, in linked-list fashion, the sequence of services required in processing the work object.
A work class also may have multiple instruction sheets associated with it, corresponding to multiple states or conditions of a work object. For example, a work class may have an associated instruction sheet that gets executed when a work object is first instantiated to ensure that the work object is properly initialized. A second instruction sheet may provide instructions for the normal processing of an instance of the work class. And a third instruction sheet may specify actions to be taken when an instance of the work class is being deleted or the work flow system is being shut down, and, for example, may cause a work object to be archived or otherwise disposed of.
When the services of a particular work performer class are needed, a work object dispatching sub-system of the work flow system run-time engine identifies the work queue associated with the desired work performer class and enqueues a pointer to the work object accordingly. When a work performer associated with the class becomes available, e.g., by completing a work order for a previous work object, the work performer polls its associated queue to determine if there is another work object in need of servicing.
A work object typically includes a number of data fields that hold various information about the work object and the data it contains. For example, a work flow system for use in a bank branch office may include classes of work objects representing various banking transactions such as account deposits and withdrawals, or loan applications. A work object may, therefore, be a specific instance of a work object class representing business loan applications, and might include data fields such as amount and approval representing, for example, a desired loan amount and whether the loan has been approved, respectively. Typically, the values of data fields are provided by the originator of the work object, such as by filling in a computer-based form, but may also be calculated by a work performer, retrieved from an on- line data base or automatically maintained by the work flow system run-time engine. For example, work object 22 of FIG. 2 may represent a computerized loan application form and work performers 26a and 26b may each represent a task of filling out an application form. Thus, a potential loan customer or bank employee may fill in a computer-based loan application form with relevant loan information, such as the amount and duration of the loan, as well as the customer' s credit information. Other data, such as the amount of each installment payment, may be calculated later by other work performers, such as work performer 27 in work flow system 20.
Referring still to FIG. 2, after the form is completed, the work object 22 comprising the loan form is submitted to the run-time engine of work flow system 20 and processed in accordance with the instruction sheet associated with the work object class of which work object 22 is an instance. Illustratively, work object 22 is dispatched to work queue 24 which is associated with the class of work performers 26a and 26b. When work performer 26a becomes available, it polls work queue 24 and determines that work object 22 is the next item requiring service. Accordingly, work performer 26a dequeues work object 22, obtains a copy of work object 22 from the server, or the relevant subset of data from within the work object, and begins processing it. However, if queue 24 is empty, work performer 26a enters an idle state and periodically polls queue 24 until another work object has been enqueued. Similarly, work performer 26b also services work objects from work queue 24.
Multiple work performers instantiated from the same work performer class may be included in a work flow system to improve system response times, or to provide redundancy for important or high volume services. In the event that both work performer 26a and 26b are available, various algorithms or heuristics may be used to determine which work performer is to service a work object. When a work performer finishes servicing a work object, the work object is submitted to the work flow system run-time engine which determines further processing pursuant to run-time series of work orders for objects of that class.
Work flow system 20 of FIG. 2 is appropriate for a single work flow system that can encompass an entire organization without becoming unmanageable. As the work flow system increases in size, however, it may become difficult to design, administer, and maintain. Applicants have discovered it is desirable to be able to divide, or partition, a large work flow system into multiple smaller, cooperative, work flow systems, so that, for example, separate work flow systems may be implemented for each department of a corporation. While partitioning a large work flow system may ease the design and maintenance burden of the work flow system as a whole, the presence of multiple work flow systems presents other difficulties, for which the present invention provides solutions .
From a theoretical point of view, principles of object oriented design, including data abstraction and information hiding, imply that a work flow system, such as work flow system 20 of FIG. 2 should be semantically self-contained. In a pure object-oriented work-flow system, each class definition and each object within the boundaries of work flow system 20 should refer only to other classes and objects contained within work flow system 20. This implies that objects within the boundaries are invisible, and hence inaccessible, to objects outside the boundaries and, conversely, objects outside the boundaries of work flow system 20 are invisible to objects within work flow system 20.
In an environment consisting of a single work flow system, the semantic impenetrability of a work flow system boundary is not a concern because all objects, whether work objects, work performers, or work queues, are contained within the work flow system. Thus, a single instance of a work performer class may service all work objects in the work flow system. However, in a multiple work flow environment, such as that shown in FIG. 4, a single work performer cannot service all work objects. Rather each work flow system 40 and 42 must instantiate at least one work performer of each work performer class to be used in that particular work flow system.
For example, in FIG. 4, work flow system 40 may comprise a loan department procedure of a bank while work flow system 42 comprises an accounting department procedure of the same bank. Furthermore, work objects 44a and 44b may each represent a purchase order for new computer equipment for their respective departments, which must be approved by the bank' s management information services (MIS) department. Because both work flow systems 40 and 42 require the task of obtaining MIS approval, each must include a work performer, such as 46a and 46b, that performs that task.
Providing instances of work performer 46a and 46b in each of work flow systems 40 and 42 is an inefficient use of system resources because all of the class information associated with work performers 46a and 46b must be duplicated in each work flow system. Furthermore, such duplication violates principles of abstraction and data hiding, since modifying the behavior or characteristics of a work performer class in one work flow system would necessitate re-coding each copy of that work performer class in all work flow systems that use a work performer of that class. It may also be difficult to automatically gather and report information on an organization-wide level. For example, it may be difficult to gather comprehensive bank-wide statistics on computer purchases. However, allowing unrestricted access across a work flow system boundary is undesirable with respect to maintenance concerns. For example, modifying an object in one work flow system may cause objects in another work flow system to cease functioning properly. Also, for security reasons it may be disadvantageous to allow external objects to access internal objects of a work flow system. For example, free access to a work performer that debits a bank account may allow a rogue external object to accidentally or intentionally post a debit to an incorrect bank account.
In accordance with the principles of the present invention, the maintenance and administrative burden of a work flow system is reduced while reliability and security are enhanced by providing limited access across work flow system boundaries . This is accomplished by allowing a work flow system to advertise specific services that may be accessed by external objects. A special work performer class therefore is created that encapsulates those services and provides a protected, well-defined interface for accessing those services.
In a preferred embodiment of the present invention, a C++ or Java class representing a work flow system is created wherein the publicly visible member functions provide access to those services that the work flow system advertises for export. The exporting class definition, therefore, appears to be a work performer class definition. Thus, another work flow system desiring to access the advertised services need only include the work performer class definition of the exporting work flow system in its own list of object classes . Referring now to FIG. 5, work flow systems 50 and 52 may be set up for the accounting departments of an organization. Work flow systems 50 and 52 may then export, or make public, various services, such as get_purchase_approval or update_accounts_receivable for use by other work flow systems. To export these services, a work performer class is created that is incorporated into other work flow systems. An instance of a work flow system work performer class functions as a proxy for a work performer in the work flow system advertising the service. The proxy work performer has access into the exporting work flow system through the mechanism of an agent work object.
Still referring to FIG. 5, work flow system 50 alternatively may belong to the research department of an organization, and work object 53 may correspond to a purchase request for new computer equipment. At some point in the processing of work object 53 within research department work flow system 50, it may be necessary to obtain accounting department approval of the purchase request. Rather than explicitly incorporate an accounting department process step in work flow system 50, work flow system 52 (corresponding to the accounting department of the organization) exports a service get_accountinσ_approval . This service is incorporated into work flow system 50 via proxy work performer 54. When work object 53 requires the service get_accounting_approval, the work flow system run-time engine determines that the service is provided by proxy work performer 54 and dispatches work object 53 accordingly.
Proxy work performer 54 dequeues the pointer to work object 53 from work queue 55, obtains a copy of the entire work object (or a subset of its data) from the repository and determines that work object 53 is requesting service from work flow system 52. Proxy work performer 54 then causes the creation of agent work object 57 in work flow 52. Any required parameters of work object 53 are transferred to agent work object 57, the identity of the originating work flow and work object (i.e., work object 53 in work flow 51) are saved, and a pointer corresponding to agent work object 57 is enqueued to work queue 59 to await the first available service in work flow system 52 of the kind requested (e.g., the service provided by work performer 58). While agent work object 57 is being processed, originating work object 53 is locked to prevent inconsistencies between the values of data in the agent work object and originating work object 53.
When agent work object 57 has finished executing the work orders associated with the exported service, a special terminate instruction is invoked. The terminate instruction causes the identity of the originating work flow and work object to be retrieved. Any return values may then be transferred back from agent work object 57 to originating work object 53. Finally, agent work object 57 is disposed of, the list of work orders associated with work object 53 is updated, and work object 53 is unlocked and returned to the run-time engine of work flow system 50 for further processing.
In the preceding discussion it has been assumed that each work performer queue is visible to all work performers servicing that queue. In a work flow system running within a local area network environment, a global queue for a particular work performer class may be created and maintained on a globally visible network server. As long as communication traffic on a LAN is not excessive, communication delays are expected to remain relatively short and periodic polling of queues by the associated work performer objects is not expected to significantly impact communication throughput.
However, it is contemplated that the foregoing design may begin to fail when a work flow system expands beyond a single network site. The separation of a networked enterprise into multiple independent, mostly-autonomous, physical sites is expected to pose a formidable problem in work flow system design. It is desired that portions of a single work flow system may be distributed across multiple physical sites located anywhere in the world, but seemingly must cooperate with each other as if there were no separation. For example, in FIG. 6, enterprise network 60 comprises LANs 61a, 61b and 61c located at disparate sites around the globe. However, a work flow spanning these sites should appear to the user as a single network, i.e., there should be little discernable difference to a user on LAN 61a between a work object being serviced by a work performer in LAN 61a and one being serviced by a work performed in LAN 61c.
Interlinked networks may be classified by type according to speed and availability. As used herein, speed refers to the rate at which data can transmitted over the link. For example, a T3 line can carry data at a rate of about 45 million bits per second, whereas a dial-up modem link is limited to about 28 thousand bits per second. Availability refers to whether access to the link is permanent or only exists on a transient basis. For example, a leased line may be available continually whereas a satellite link may only be available during certain times of the day. A transient link may require a mechanism to store a transmission and then forward the transmission when a link becomes available, and may introduce substantial delay in transferring data to a remote site.
In enterprise-wide work flow system 60 of FIG. 6, LAN 61a may be connected to LAN 61b using dedicated line 62 or other means to provide continuous, high-speed, inter-LAN communications, in which case the connection between LANs 61a and 61b would be classified as both fast and permanent. Conversely, LAN 61a may be connected to LAN 61c on an as-needed basis by a dial-up type modem connection via link 64. Such a connection would be a slow and transient connection.
Typically, when a work object or subset of data is dispatched to a work performer, the run-time engine of the sending work flow system waits to receive an acknowledgment from the receiving work queue before proceeding to service another work object. If the runtime engine of the sending work flow system and the destination work queue are linked by a fast, permanent connection the wait is short and system response times remain acceptable to the end user. However, when the sending work flow system and receiving work queue share a slow or transient link, waiting for an acknowledgment may result in unacceptably slow system response times. Accordingly, referring to FIG. 7, another aspect of the present invention is described. In FIG. 7, gateway work performers 72a and 72b are provided at either end of communication link 74 interconnecting work flow partitions 70a and 70b. Gateway work performers 72a and 72b serve as local receivers for work objects or extracted data that are destined for remote work queues. For example, when work object 76 is destined for remote work queue 77, it is sent to gateway work performer 72a for transmission to the remote site. Local gateway work performer 72a provides an acknowledgment to the run-time engine of work flow system 70, thus allowing the run-time engine of work flow system 70 to provide another work object to work performer 73 for servicing. Gateway work performer 72a then transmits the work object to the remote site in a manner appropriate for the type of communication link. For example, if link 74 is a fast and permanent connection, gateway work performer 72a may immediately transmit the work object to gateway work performer 72b at the remote site. By comparison, if communication link 74 is a transient connection, work performer 72a may defer transmission until the link is available, or may accumulate multiple work objects so that they may be transmitted in bulk to the remote site. One skilled in the art will recognize that gateway work performers 72a and 72b may be enhanced to provide additional capabilities such as gathering statistics for determining inter-site transit times, to aggregate multiple work objects or subsets of data destined for a remote site to enable transfer in a single transmission, performing data compression to reduce transmission times, or encrypting data to improve data security while transmitting over open lines, such as telephone lines or satellite links.
In addition to delays in receiving acknowledgments, communications links other than fast, permanent links introduce other concerns. For example, using current technologies, typical dial-up analog connections are limited to transmission rates in the neighborhood of three thousand bytes per second. While digital connections using ISDN may increase the data rate by a factor of four over analog connections, the data rate is still very low compared to a rate of 10 million bits per second for a typical Ethernet LAN. In addition, unless the dial-up link is always connected, additional time is expended in establishing the link. Thus, if a work performer exists on LAN 61c, of FIG. 6, while its class queue is located on LAN 61a, it may take several seconds for a dial-up link to be established each time work performer polls the queue. In addition, a large organization may have many work performers and corresponding queues distributed across multiple LA s. Even when high speed dedicated digital links are employed, the overhead associated with polling a class queue on another LAN may negatively impact system throughput and performance.
In accordance with further principles of the present invention, a single global queue paradigm is retained through the use of a structure referred to herein as a partitioned logical queue . Instead of having a physical work queue, as described hereinabove with respect to FIG. 2, each work performer class is associated with a logical queue. A logical work queue provides the appearance of a single global queue, but internally it is divided into partitions, wherein each partition is associated with a physical queue. In essence, a logical work queue provides an additional layer of abstraction between the work flow system runtime engine and the work performers in the work flow system. A logical work queue accepts work objects destined for a work performer class queue and assigns the work object to one of the logical partitions of the logical queue based on various properties of the work object.
FIG. 8 represents screen display 80 of a partitioned logical queue as table 81, with each row 81a, 81b, . . . 81n of table 81 corresponding to a logical partition. A first attribute of a queue partition is its partition expression 82, which may depend on information within data fields of the work object. When a work object is dispatched to the logical queue, partition expression 82 for each partition queue 81a...81n is evaluated. The work object is then assigned to a work queue based on the results of evaluating partition expressions 82. If only one partition expression is true, the work object is assigned to the corresponding queue partition. If more than one partition expression evaluates as true, various heuristics may be used to select a queue partition, for example, heuristics relating to throughput statistics of the various queues, locality, etc. Lastly, if no partition expression is true, an exception is declared and an exception work order for the work object being enqueued is invoked by the work flow system run-time engine.
For example, in FIG. 8, the logical queue associated with the Loan Officer class of work performers has three partitions, each represented by a row (81a to 81c) in table 81. If a work object to be enqueued has a loan amount greater than $203,500 partition expression 82 in the first partition 81a is evaluated as true and the work object is assigned to the Loan Officer queue of the first partition. However, if the loan amount is less than or equal to $203,500 the first partition equation is evaluated as false, and the partition expressions in the second and third rows, 81b and 81c, are true.
In a preferred embodiment of the present invention, each partition, 81a...81n, of a partitioned queue has an associated weight 83 as shown in the second column of FIG. 8. When more then one partition expression evaluates to true, the work object is dispatched in accordance with a weighted probability distribution based on weights 83 assigned to each partition 81a...81c. For example, in FIG. 8 partitions 81b and 81c have the same partition expression, but in partition 81b weight 83 has a value of 2, while in partition 81c weight 83 has a value of 1. Thus, two- thirds of loan applications for less than $203,500 are enqueued to the queue partition associated with row 81b of the table, and the remaining one-third to the third queue partition (81c) .
As discussed hereinabove, it is desirable to provide the paradigm of a single work performer queue per class of work performer in a work flow system. Due to performance issues, it is also desirable to actively direct work objects to remote queue partitions instead of requiring remote work performers to poll their associated work queues across a low bandwidth communication channel. However, it is possible for a work object to be directed to a remote queue partition and never be serviced. For example, a document needing approval may be dispatched to a remote queue partition that is served by a person who happens to be on vacation. Additionally, it would be desirable to provide a load-balancing capability that ensures that some work performers are not underutilized and sitting idle for long periods.
The foregoing issues may be addressed in several ways. For example, partition expressions in a work performer queue partition table may be updated to reflect non-availability of a service (due to personnel vacation, etc.). Alternatively, work objects may have associated with them information for identifying specific work performers which are to process the work object, thus overriding the normal work object dispatch function of a partitioned work performer queue. However, these potential solutions may be undesirable due to increased administrative burden and/or processing overhead.
In a preferred embodiment of the present invention, the foregoing problems are solved by periodically redistributing work objects from one partition of a work performer queue to another partition. Various criteria, such as elapsed time in a queue or queue size, may be specified to indicate when work objects should be redistributed. Redistribution is accomplished through the mechanism of a load balancing work object and load balancing work performers.
In accordance with this aspect of the present invention, each queue partition has an associated load balancing work performer. A load balancing work object travels in a circuit that visits all load balancing work performers associated with queue partitions having the same partition expression. The load balancing work object carries with it information concerning work objects matching the corresponding partition expression. When a load balancing work performer services the load balancing work object, it determines the status of each work object in the work queue for the local work flow system and as well as information for those work objects carried by the load balancing work object. Based on the status of the work objects, the load balancing work performer selectively transfers work object pointers from the load balancing work object to the queue of the local work flow system and/or from the queue to the load balancing work object. The load balancing work object is then dispatched to the next load balancing work performer in the circuit, which repeats the process just described. For example, referring to FIG. 9, work flow 90 is divided into partitions 90a, 90b, and 90c, including load balancing work performers 92a, 92b, and 92c, respectively. Let it be assumed that based on the partition expressions for partitions 90a to 90c, work object 94 is to be enqueued to work performer 96a in work flow 90a. A pointer for work object 94 is, therefore, enqueued to load balancing work performer 92a for transfer to work flow 90b. Load balancing work object 98 circulates between load balancing work performers 92a, 92b, and 92c. Information relating to work object 94 is inserted into load balancing work object 98 the next time it arrives at load balancing work performer 92a. Subsequently, when load balancing work object 98 arrives at load balancing work performer 92b, the pointer for work object 94 is removed from the load balancing work object and enqueued for work performer 96b. At the same time, load balancing work performer 92b may insert information for other objects (not shown) bound for partitions 90a and 90c into load balancing work object 98 to be carried back to their respective load balancing work performers.
Applicants expect that it may also be desirable to allow load balancing work performers to override the partition assigned to a work object based on the partition expressions. For example, if work object 94 of FIG. 9 has been enqueued for work performer 96a for an excessive length of time, load balancing work performer 92a may delete it from work performer queue 95 and insert a pointer for work object 94 into load balancing work object 98. Then either of load balancing work performers 92b or 92c, in response to the excessive age of work object 94, may transfer the information for work object 94 from load balancing work object 98 and enqueue a pointer for work object 94 to work performer queues 96b or 96c, respectively. As one skilled in the art will appreciate, criteria other than length of time enqueued may be used to determine when the partition assigned to a work object should be overridden, such as, the length of the destination queue, the time of day, or a heuristic based on average service times at work performers 96a, 96b, and 96c. Thus, using a load balancing technique to virtualize a global queue may provide additional benefits such as dynamically balancing queue work loads, providing work performer redundancy, and ensuring all work objects are serviced in a reasonable length of time.
In addition to load balancing issues, transactional security concerns must also be addressed when distributing work objects across work flow systems at geographically separated sites. For example, if an network link fails during transmission of a work object, a work flow system must be able to recover without losing or corrupting the work object.
Known techniques for remote data base transaction processing typically involve the use of file or record locking to provide transactional integrity. For example, a source site may first obtain a record lock for the record being transmitted and, using a remote procedure call, may obtain a corresponding record lock at the destination site. The record is then transmitted. After the destination site acknowledges receipt of the transferred record, the source site lock is removed, and then the destination site lock is removed. However, if the source site were to shut down or fail between obtaining the source lock and releasing the destination lock recovery of the transaction may be difficult or impossible.
In accordance with principles of the present invention, transactional integrity is ensured by a multi-phase work object transmission sequence as illustrated in FIG. 10. While FIG. 10 illustrates transmission of work objects, it is to be understood that selected subsets of data extracted from a work object may be handled in the same manner. First, at step 102, the source marks the work object to be transmitted as completed, i.e., processing at the current site is finished. Then, at step 104, using remote procedure calls, the source site enqueues the work object and creates a marker record at the destination site (step 106) . This operation is completed as an atomic step. The source then deletes, backs up, or otherwise archives its copy of the work object, at step 108. And lastly, at step 110, the source site work flow engine uses a remote procedure call to delete the marker record at the destination site. Note that once the work object is enqueued and the marker record created, the destination site may begin processing the work object. In contrast, when using a record or file lock, the destination site may not begin processing until the lock is removed.
Thus, the transmission system of the present invention avoids the use of file or record locking. Furthermore, the status of a work object can easily be determined at any time. For example, when a work flow system is restarted after a shutdown or failure, the presence of a marker record indicates that a work object was received, but that the source site may still need to delete its copy of the completed work object. Analogously, the presence of a work object marked complete indicates that it was in the process of being transmitted. By checking for the presence of a marker record at the destination, the source site may quickly ascertain whether the work object was successfully transmitted and may retransmit or delete the work object and marker record as appropriate.
Disclosed hereinabove are various principles of the present invention which may be employed singly or in combination to construct an enterprise-wide work flow system. One skilled in the art will also appreciate that the present invention can be practiced by other than the described embodiments, which are presented for purposes of illustration and not of limitation, the present invention being limited only by the claims which follow.

Claims

What Is Claimed Is:
1. A method of providing services in a work flow system, comprising steps of: providing a computer network; defining a first work flow class, the first work flow class definition including definitions of classes of objects available within the work flow system and an identification of a service to be provided by an instance of the first work flow class; and defining a second work flow class, the second work flow class including the first work flow class definition, the first work flow class definition providing a defined interface for accessing services provided by the instance of the first work flow class to an instance of the second work flow class.
2. The method of claim 1 further comprising steps of: instantiating an agent work object in the instance of the first work flow class responsive to receipt of a request from a work object in the instance of the second work flow class for a publicly available service; transferring data from the work object to the agent work object; processing the agent work object to provide the requested service; and returning data to the work object.
3. a method of distributing objects within a work flow system, the method comprising steps of: associating a partition expression with each potential destination of a work object within the work flow system; evaluating each partition expression; and distributing the work object to a destination responsive to the evaluated value of the corresponding partition expression.
4. The method of claim 3 wherein the step of evaluating comprises evaluating a logical expression as a function of data fields within the work object.
5. The method of claim 3 wherein the step of evaluating comprises evaluating a logical expression as a function of globally available data.
6. The method of claim 3 wherein the step of distributing the work object further comprises a step of determining if more than one partition expression evaluates to a predetermined value.
7. The method of claim 6 further comprising steps of: selecting a destination responsive to the determination that more than one partition expression evaluates to the predetermined value; and distributing the work object to the selected destination.
8. The method of claim 7 further comprising a step of assigning each potential destination a weight.
9. The method of claim 8 wherein the step of selecting a destination comprises randomly selecting a destination wherein the probability of selecting a destination is proportional to the weight of the destination relative to the weights of all destinations having a partition expression that evaluates to the predetermined value.
10. The method of claim 8 wherein the step of selecting a destination comprises selecting the destination having the lowest cost responsive to the destinations having partition expressions that evaluate to the predetermined value also having equal weights.
11. A method of distributing objects within a work flow system, the method comprising steps of: providing a predetermined partition table, the partition table comprising a plurality of entries, each one of the plurality of entries including a partition expression and a destination associated with an evaluated value of the partition expression; evaluating the partition expression for each of the plurality of entries; selecting one of the plurality of entries responsive to the value of the partition expression for that entry evaluating to a predetermined value; and distributing the object to the destination associated with the selected one of the plurality of entries .
12. The method of claim 11 wherein the step of selecting one of the plurality of entries comprises randomly selecting one of the plurality of entries.
13. The method of claim 12 further comprising a step of providing each one of the plurality of entries in the partition table a weighting factor; and wherein the step of randomly selecting one of the plurality of entries comprises randomly selecting one of the plurality of entries in proportion to the weighting factors.
14. A method of distributing an object in a computer environment comprising a plurality of interconnected computer networks, wherein the object is to be sent from a local network to a remote network, the method comprising the steps of: providing each computer network with a queue for storing objects destined for the remote network; circulating a load balancing work object through the interconnected computer networks such that the load balancing work object visits each of the networks at least once; selectively embarking information relating to work objects stored in the queue into the load balancing work object when the load balancing work object visits the local network; and selectively debarking information relating to work objects from the load balancing work object when the load balancing work object visits the remote network.
15. The method of claim 14 wherein the steps of circulating, selectively embarking, and selectively debarking are repeated at least once.
16. Apparatus for processing data comprising: a plurality of computers interconnected to form a computer network, wherein the plurality of computers are programmed with first and second routines to process the data; and at least one computer programmed to associate with each one of the first and second routines a logical expression, to select one of the first and second routines responsive to the associated logical expression being true, and sending the data to the selected routine.
17. The apparatus of claim 16 wherein the at least one computer is further programmed to associate with the first and second routines first and second weights, respectively, and wherein one of the first and second routines is selected responsive to the first and second weights.
18. The apparatus of claim 17 wherein the at least one computer is programmed to randomly select one of the first and second routines in proportion to the first and second weights.
19. The apparatus of claim 17 wherein the at least one computer is programmed to select the first routine responsive to the first and second weights being equal and the first routine having a lower cost.
20. The apparatus of claim 16 wherein the first and second routines are executing on the same computer.
21. The apparatus of claim 16 wherein the first and second routines are executing on different computers .
PCT/US1998/015565 1997-07-23 1998-07-23 System for enterprise-wide work flow automation WO1999005632A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CA002297065A CA2297065A1 (en) 1997-07-23 1998-07-23 System for enterprise-wide work flow automation
AU86649/98A AU8664998A (en) 1997-07-23 1998-07-23 System for enterprise-wide work flow automation
EP98938035A EP0996916A1 (en) 1997-07-23 1998-07-23 System for enterprise-wide work flow automation

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US08/899,408 1997-07-23
US08/899,408 US6338074B1 (en) 1997-07-23 1997-07-23 System for enterprise-wide work flow automation

Publications (1)

Publication Number Publication Date
WO1999005632A1 true WO1999005632A1 (en) 1999-02-04

Family

ID=25410910

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1998/015565 WO1999005632A1 (en) 1997-07-23 1998-07-23 System for enterprise-wide work flow automation

Country Status (5)

Country Link
US (3) US6338074B1 (en)
EP (1) EP0996916A1 (en)
AU (1) AU8664998A (en)
CA (1) CA2297065A1 (en)
WO (1) WO1999005632A1 (en)

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2353118A (en) * 1999-06-30 2001-02-14 Phoenix Technology Patent Dev Work flow management system
US6308163B1 (en) 1999-03-16 2001-10-23 Hewlett-Packard Company System and method for enterprise workflow resource management
EP1540543A2 (en) * 2002-06-27 2005-06-15 Siebel Systems, Inc. Method and apparatus to facilitate development of a customer-specific business process model
GB2460730A (en) * 2008-06-12 2009-12-16 Ibm Distributed workflow management system
EP2137584A1 (en) * 2007-04-12 2009-12-30 Thomson Licensing Centralized work flow monitoring
US7809597B2 (en) 2005-05-05 2010-10-05 Siebel Systems, Inc. Progressive refinement model for business processes
US7831453B2 (en) 2005-05-05 2010-11-09 Siebel Systems, Inc. Modeling of business process data
US7856436B2 (en) 2005-12-23 2010-12-21 International Business Machines Corporation Dynamic holds of record dispositions during record management
US7895070B2 (en) 2005-05-05 2011-02-22 Siebel Systems, Inc. Providing multiple views of a business process definition to different users
US8037029B2 (en) 2006-10-10 2011-10-11 International Business Machines Corporation Automated records management with hold notification and automatic receipts
US8880487B1 (en) 2011-02-18 2014-11-04 Pegasystems Inc. Systems and methods for distributed rules processing
US8924335B1 (en) 2006-03-30 2014-12-30 Pegasystems Inc. Rule-based user interface conformance methods
US8959480B2 (en) 2004-05-26 2015-02-17 Pegasystems Inc. Methods and apparatus for integration of declarative rule-based processing with procedural programming in a digital data-processing environment
US9189361B2 (en) 2007-03-02 2015-11-17 Pegasystems Inc. Proactive performance management for multi-user enterprise software systems
US9195936B1 (en) 2011-12-30 2015-11-24 Pegasystems Inc. System and method for updating or modifying an application without manual coding
US9678719B1 (en) 2009-03-30 2017-06-13 Pegasystems Inc. System and software for creation and modification of software
US10402756B2 (en) 2005-10-19 2019-09-03 International Business Machines Corporation Capturing the result of an approval process/workflow and declaring it a record
US10467200B1 (en) 2009-03-12 2019-11-05 Pegasystems, Inc. Techniques for dynamic data processing
US10469396B2 (en) 2014-10-10 2019-11-05 Pegasystems, Inc. Event processing with enhanced throughput
US10698599B2 (en) 2016-06-03 2020-06-30 Pegasystems, Inc. Connecting graphical shapes using gestures
US10698647B2 (en) 2016-07-11 2020-06-30 Pegasystems Inc. Selective sharing for collaborative application usage
US11048488B2 (en) 2018-08-14 2021-06-29 Pegasystems, Inc. Software code optimizer and method
US11567945B1 (en) 2020-08-27 2023-01-31 Pegasystems Inc. Customized digital content generation systems and methods

Families Citing this family (85)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040030639A1 (en) * 2002-02-22 2004-02-12 Lendingtree, Inc. Method and computer network for co-ordinating a loan over the internet
US20050273406A1 (en) * 2003-04-08 2005-12-08 Lending Tree, Inc. Method and computer network for co-ordinating a loan over the internet
US6385594B1 (en) * 1998-05-08 2002-05-07 Lendingtree, Inc. Method and computer network for co-ordinating a loan over the internet
US6606740B1 (en) * 1998-10-05 2003-08-12 American Management Systems, Inc. Development framework for case and workflow systems
US6898574B1 (en) * 1998-11-09 2005-05-24 John Francis Regan Lender and insurer transaction processing system and method
EP1003117A3 (en) * 1998-11-17 2003-07-23 Citibank, N.A. Method and system for strategic services enterprise workload management
DE19954268B4 (en) * 1998-12-01 2005-07-21 Ibm Corp. Method and system for improving workflow in workflow management systems
JP3437933B2 (en) * 1999-01-21 2003-08-18 インターナショナル・ビジネス・マシーンズ・コーポレーション Browser sharing method and system
US6854085B1 (en) * 1999-07-15 2005-02-08 Netscape Communicatons Corporation System and method for automatically pre-setting form field values
US7392210B1 (en) * 2000-04-07 2008-06-24 Jpmorgan Chase Bank, N.A. Workflow management system and method
US7051106B2 (en) * 2000-08-15 2006-05-23 Lucent Technologies Inc. Scheduling of calls with known holding times
US7412399B1 (en) 2001-03-30 2008-08-12 Bea Systems Inc. Designing business processes using distributed process flows
US7249195B2 (en) 2001-03-30 2007-07-24 Minor Ventures, Llc Apparatus and methods for correlating messages sent between services
US20060197763A1 (en) * 2002-02-11 2006-09-07 Landnet Corporation Document geospatial shape tagging, searching, archiving, and retrieval software
US7454751B2 (en) * 2001-06-07 2008-11-18 Intel Corporation Fault-tolerant system and methods with trusted message acknowledgement
US7412520B2 (en) * 2001-06-07 2008-08-12 Intel Corporation Systems and methods for recoverable workflow
US7100147B2 (en) * 2001-06-28 2006-08-29 International Business Machines Corporation Method, system, and program for generating a workflow
US7069536B2 (en) * 2001-06-28 2006-06-27 International Business Machines Corporation Method, system, and program for executing a workflow
US7043714B2 (en) * 2001-06-28 2006-05-09 International Business Machines Corporation Method, system, and program for using objects in data stores during execution of a workflow
US7228547B2 (en) 2001-07-30 2007-06-05 International Business Machines Corporation Method, system, and program for enabling access to a plurality of services
US7296056B2 (en) * 2001-07-30 2007-11-13 International Business Machines Corporation Method, system, and program for selecting one user to assign a work item in a workflow
US7047535B2 (en) 2001-07-30 2006-05-16 International Business Machines Corporation Method, system, and program for performing workflow related operations using an application programming interface
US7698427B2 (en) 2001-07-30 2010-04-13 International Business Machines Corporation Method, system, and program for transferring data from an application engine
US8244837B2 (en) * 2001-11-05 2012-08-14 Accenture Global Services Limited Central administration of one or more resources
US7543303B2 (en) * 2002-03-29 2009-06-02 International Business Machines Corporation Use of multiple queues in coupling facility to provide zoning capability in a parallel sysplex environment
US20040003353A1 (en) * 2002-05-14 2004-01-01 Joey Rivera Workflow integration system for automatic real time data management
US20040024771A1 (en) * 2002-08-01 2004-02-05 Oracle International Corporation Buffered message queue architecture for database management systems with transactional enqueue support
US7181482B2 (en) * 2002-08-01 2007-02-20 Oracle International Corporation Buffered message queue architecture for database management systems
US7185034B2 (en) * 2002-08-01 2007-02-27 Oracle International Corporation Buffered message queue architecture for database management systems with guaranteed at least once delivery
US7203706B2 (en) * 2002-08-01 2007-04-10 Oracle International Corporation Buffered message queue architecture for database management systems with memory optimizations and “zero copy” buffered message queue
US7185033B2 (en) * 2002-08-01 2007-02-27 Oracle International Corporation Buffered message queue architecture for database management systems with unlimited buffered message queue with limited shared memory
US8548902B2 (en) * 2003-08-15 2013-10-01 Larry Porter System for online lending services via an application service provider network
US7406511B2 (en) * 2002-08-26 2008-07-29 International Business Machines Corporation System and method for processing transactions in a multisystem database environment
JP2004163997A (en) * 2002-09-20 2004-06-10 Fuji Xerox Co Ltd Service processing device, service processing method, and program
US7668763B1 (en) 2002-11-25 2010-02-23 Xcm Development, Llc Tax return outsourcing and systems for protecting data
US7159206B1 (en) 2002-11-26 2007-01-02 Unisys Corporation Automated process execution for project management
US7155700B1 (en) 2002-11-26 2006-12-26 Unisys Corporation Computer program having an object module and a software project definition module which customize tasks in phases of a project represented by a linked object structure
US7174348B1 (en) 2002-11-26 2007-02-06 Unisys Corporation Computer program having an object module and a software development tool integration module which automatically interlink artifacts generated in different phases of a software project
AU2003299668A1 (en) * 2002-12-18 2004-07-22 Emc Corporation Resource allocation aware queuing of requests for media resources
US20040205019A1 (en) * 2003-04-08 2004-10-14 Lendingtree, Inc. Method and system for selecting qualification forms for financial services and financial products
US6976144B1 (en) * 2003-05-06 2005-12-13 Pegasystems, Inc. Methods and apparatus for digital data processing with mutable inheritance
US8239233B1 (en) 2003-07-17 2012-08-07 Xcm Development, Llc Work flow systems and processes for outsourced financial services
US8365193B2 (en) * 2003-08-14 2013-01-29 Oracle International Corporation Recoverable asynchronous message driven processing in a multi-node system
US7593933B2 (en) * 2004-02-20 2009-09-22 Mixed Grill, L.L.C. Method and apparatus for a collaborative interaction network
US7657889B2 (en) * 2004-04-06 2010-02-02 International Business Machines Corporation Method, system, and storage medium for searching multiple queues for prioritized work elements
US7392524B2 (en) * 2004-04-06 2008-06-24 International Business Machines Corporation Method, system, and storage medium for managing computer processing functions
US7213022B2 (en) * 2004-04-29 2007-05-01 Filenet Corporation Enterprise content management network-attached system
US7802007B2 (en) 2004-05-19 2010-09-21 Salesforce.Com, Inc. Techniques for providing connections to services in a network environment
US7702517B2 (en) * 2004-07-12 2010-04-20 International Business Machines Corporation Active and contextual risk management using risk software objects
US20070011234A1 (en) * 2004-07-29 2007-01-11 Xcm Development, Llc Computer conferencing system and features
US9645712B2 (en) 2004-10-01 2017-05-09 Grand Central Communications, Inc. Multiple stakeholders for a single business process
US20060085374A1 (en) * 2004-10-15 2006-04-20 Filenet Corporation Automatic records management based on business process management
US20060085245A1 (en) * 2004-10-19 2006-04-20 Filenet Corporation Team collaboration system with business process management and records management
US20060100944A1 (en) * 2004-11-10 2006-05-11 Lendingtree, Llc Method and computer network for co-ordinating financial services/products
US7779418B2 (en) * 2004-12-30 2010-08-17 Oracle International Corporation Publisher flow control and bounded guaranteed delivery for message queues
US7818386B2 (en) * 2004-12-30 2010-10-19 Oracle International Corporation Repeatable message streams for message queues in distributed systems
US8335704B2 (en) 2005-01-28 2012-12-18 Pegasystems Inc. Methods and apparatus for work management and routing
US7680793B2 (en) * 2005-10-07 2010-03-16 Oracle International Corporation Commit-time ordered message queue supporting arbitrary read and dequeue patterns from multiple subscribers
US8196150B2 (en) * 2005-10-07 2012-06-05 Oracle International Corporation Event locality using queue services
US20070088736A1 (en) * 2005-10-19 2007-04-19 Filenet Corporation Record authentication and approval transcript
US7640222B2 (en) * 2006-03-03 2009-12-29 Pegasystems Inc. Rules base systems and methods with circumstance translation
US20070233902A1 (en) * 2006-03-30 2007-10-04 Alan Trefler User interface methods and apparatus for rules processing
US20090132232A1 (en) * 2006-03-30 2009-05-21 Pegasystems Inc. Methods and apparatus for implementing multilingual software applications
US20070239715A1 (en) * 2006-04-11 2007-10-11 Filenet Corporation Managing content objects having multiple applicable retention periods
US8046749B1 (en) * 2006-06-27 2011-10-25 The Mathworks, Inc. Analysis of a sequence of data in object-oriented environments
US8904299B1 (en) 2006-07-17 2014-12-02 The Mathworks, Inc. Graphical user interface for analysis of a sequence of data in object-oriented environment
US10176540B2 (en) 2006-10-17 2019-01-08 First American Title Insurance Company Apparatus and method for generating title products
US20080222112A1 (en) * 2007-03-07 2008-09-11 Altep, Inc. Method and System for Document Searching and Generating to do List
US9027025B2 (en) * 2007-04-17 2015-05-05 Oracle International Corporation Real-time database exception monitoring tool using instance eviction data
US9128895B2 (en) * 2009-02-19 2015-09-08 Oracle International Corporation Intelligent flood control management
US20100223267A1 (en) 2009-02-27 2010-09-02 Accenture Global Services Gmbh Matching tools for use in attribute-based performance systems
JP5414448B2 (en) * 2009-10-16 2014-02-12 キヤノン株式会社 Image processing apparatus, control method therefor, and program
FR2951840B1 (en) * 2009-10-22 2011-12-23 Sagem Defense Securite ELECTRONIC CONTROL UNIT WITH REAL-TIME CORE MANAGING PARTITIONING
US9165086B2 (en) 2010-01-20 2015-10-20 Oracle International Corporation Hybrid binary XML storage model for efficient XML processing
US8458530B2 (en) 2010-09-21 2013-06-04 Oracle International Corporation Continuous system health indicator for managing computer system alerts
US20120095585A1 (en) * 2010-10-15 2012-04-19 Invensys Systems Inc. System and Method for Workflow Integration
US8745634B2 (en) 2010-10-15 2014-06-03 Invensys Systems, Inc. System and method for integrated workflow scaling
US20120095925A1 (en) * 2010-10-15 2012-04-19 Invensys Systems Inc. System and Method of Federated Workflow Data Storage
US8800055B2 (en) * 2010-10-20 2014-08-05 International Business Machines Corporation Node controller for an endpoint in a cloud computing environment
US9383939B2 (en) 2013-03-15 2016-07-05 International Business Machines Corporation Migrating and retrieving queued data in byte-addressable storage
US9444885B2 (en) * 2014-03-18 2016-09-13 Vmware, Inc. Workflow processing in a distributed computing environment
US10540217B2 (en) 2016-09-16 2020-01-21 Oracle International Corporation Message cache sizing
US10972358B2 (en) 2017-08-30 2021-04-06 Citrix Systems, Inc. Inferring congestion and signal quality
US10904102B2 (en) 2018-07-05 2021-01-26 Open Text Sa Ulc Systems and methods for communication flow modeling
US11888793B2 (en) 2022-02-22 2024-01-30 Open Text Holdings, Inc. Systems and methods for intelligent delivery of communications

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0457684A2 (en) * 1990-05-18 1991-11-21 Digital Equipment Corporation Routing objects on action paths in a distributed computing system
EP0793184A2 (en) * 1995-06-07 1997-09-03 Dun & Bradstreet Software Services, Inc. Method and apparatus for distributing work flow processes among a plurality of users

Family Cites Families (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4503499A (en) 1982-09-14 1985-03-05 Eaton Corporation Controlled work flow system
US4648061A (en) 1982-11-09 1987-03-03 Machines Corporation, A Corporation Of New York Electronic document distribution network with dynamic document interchange protocol generation
US4532588A (en) 1982-11-09 1985-07-30 International Business Machines Corporation Electronic document distribution network with uniform data stream
US4754428A (en) 1985-04-15 1988-06-28 Express Communications, Inc. Apparatus and method of distributing documents to remote terminals with different formats
US4713780A (en) 1985-04-15 1987-12-15 Express Communications, Inc. Electronic mail
US5216603A (en) 1985-11-18 1993-06-01 Action Technologies, Inc. Method and apparatus for structuring and managing human communications by explicitly defining the types of communications permitted between participants
US4932026A (en) 1986-12-19 1990-06-05 Wang Laboratories, Inc. Apparatus for distributing data processing across a plurality of loci of control
US5113393A (en) 1986-12-19 1992-05-12 Wang Laboratories, Inc. Apparatus for distributing data processing
EP0306781B1 (en) 1987-09-08 1994-04-20 Wang Laboratories Inc. Electronic mail circulation method and apparatus
US5155858A (en) * 1988-10-27 1992-10-13 At&T Bell Laboratories Twin-threshold load-sharing system with each processor in a multiprocessor ring adjusting its own assigned task list based on workload threshold
US5187790A (en) * 1989-06-29 1993-02-16 Digital Equipment Corporation Server impersonation of client processes in an object based computer operating system
US5191525A (en) * 1990-01-16 1993-03-02 Digital Image Systems, Corporation System and method for extraction of data from documents for subsequent processing
JPH09501517A (en) * 1993-02-08 1997-02-10 アクション・テクノロジーズ・インコーポレーテッド Method and apparatus for managing business processes
US5864679A (en) * 1993-09-06 1999-01-26 Kabushiki Kaisha Toshiba Transaction routing in a multiple processor system using an extracted transaction feature parameter and transaction historical data
JP3649345B2 (en) * 1994-05-26 2005-05-18 富士ゼロックス株式会社 Information processing system
US5745901A (en) * 1994-11-08 1998-04-28 Kodak Limited Workflow initiated by graphical symbols
DE69511425T2 (en) * 1994-11-08 2000-03-09 Vermeer Tech Inc PRODUCTION AID FOR ONLINE SERVICES WITH CHARGING CHARGES
US5634127A (en) * 1994-11-30 1997-05-27 International Business Machines Corporation Methods and apparatus for implementing a message driven processor in a client-server environment
US5680548A (en) * 1994-12-02 1997-10-21 Xcellenet, Inc. Systems and methods for work assignment and distribution from a server to remote/mobile nodes
JPH08287162A (en) * 1995-02-14 1996-11-01 Toshiba Corp Work flow system
US5603029A (en) * 1995-06-07 1997-02-11 International Business Machines Corporation System of assigning work requests based on classifying into an eligible class where the criteria is goal oriented and capacity information is available
US5940612A (en) * 1995-09-27 1999-08-17 International Business Machines Corporation System and method for queuing of tasks in a multiprocessing system
US5991733A (en) * 1996-03-22 1999-11-23 Hartford Fire Insurance Company Method and computerized system for managing insurance receivable accounts
US5931917A (en) * 1996-09-26 1999-08-03 Verifone, Inc. System, method and article of manufacture for a gateway system architecture with system administration information accessible from a browser
US6014673A (en) * 1996-12-05 2000-01-11 Hewlett-Packard Company Simultaneous use of database and durable store in work flow and process flow systems
US6041306A (en) * 1996-12-05 2000-03-21 Hewlett-Packard Company System and method for performing flexible workflow process execution in a distributed workflow management system
US5870545A (en) * 1996-12-05 1999-02-09 Hewlett-Packard Company System and method for performing flexible workflow process compensation in a distributed workflow management system
US5940804A (en) * 1996-12-18 1999-08-17 Turley; William N. Computer executable workflow resource management system
US6006193A (en) * 1996-12-18 1999-12-21 Gibson; Kenneth U. Computer executable workflow control system
US20020023175A1 (en) * 1997-06-04 2002-02-21 Brian R. Karlak Method and apparatus for efficient, orderly distributed processing
US5920725A (en) * 1997-07-02 1999-07-06 Adaptivity Inc. Run-time object-synthesis and transparent client/server updating of distributed objects using a meta server of all object descriptors
US5960404A (en) * 1997-08-28 1999-09-28 International Business Machines Corp. Mechanism for heterogeneous, peer-to-peer, and disconnected workflow operation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0457684A2 (en) * 1990-05-18 1991-11-21 Digital Equipment Corporation Routing objects on action paths in a distributed computing system
EP0793184A2 (en) * 1995-06-07 1997-09-03 Dun & Bradstreet Software Services, Inc. Method and apparatus for distributing work flow processes among a plurality of users

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
D. HERSCHMANN: "The Effects of Practical Business Constraints on User Interface Design", HUMAN FACTORS IN COMPUTING SYSTEMS, CHI '95 CONFERENCE PROCEEDINGS, 7 May 1995 (1995-05-07) - 11 May 1995 (1995-05-11), Denver, Colorado, US, pages 531 - 537, XP000538484 *
M. M. COMPTON: "Intelligent Validation and Routing of Electronic Forms in a Distributed Workflow Environment", PROCEEDINGS OF THE TENTH CONFERENCE ON ARTIFICIAL INTELLIGENCE FOR APPLICATIONS, 1 March 1994 (1994-03-01) - 4 March 1994 (1994-03-04), San Antonio, TX, US, pages 125 - 131, XP000479439 *

Cited By (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6308163B1 (en) 1999-03-16 2001-10-23 Hewlett-Packard Company System and method for enterprise workflow resource management
AU766298B2 (en) * 1999-06-30 2003-10-16 Phoenix Technology Patent Development Limited A work flow management system
US6725224B1 (en) 1999-06-30 2004-04-20 Phoenix Technology Patent Development Limited Controller system for interfacing with a work flow management system
GB2353118A (en) * 1999-06-30 2001-02-14 Phoenix Technology Patent Dev Work flow management system
US8639542B2 (en) 2002-06-27 2014-01-28 Siebel Systems, Inc. Method and apparatus to facilitate development of a customer-specific business process model
EP1540543A2 (en) * 2002-06-27 2005-06-15 Siebel Systems, Inc. Method and apparatus to facilitate development of a customer-specific business process model
EP1540543A4 (en) * 2002-06-27 2007-11-28 Siebel Systems Inc Method and apparatus to facilitate development of a customer-specific business process model
US8959480B2 (en) 2004-05-26 2015-02-17 Pegasystems Inc. Methods and apparatus for integration of declarative rule-based processing with procedural programming in a digital data-processing environment
US7831453B2 (en) 2005-05-05 2010-11-09 Siebel Systems, Inc. Modeling of business process data
US7809597B2 (en) 2005-05-05 2010-10-05 Siebel Systems, Inc. Progressive refinement model for business processes
US7895070B2 (en) 2005-05-05 2011-02-22 Siebel Systems, Inc. Providing multiple views of a business process definition to different users
US8423477B2 (en) 2005-05-05 2013-04-16 Siebel Systems, Inc. Providing multiple views of a business process definition to different classes of users
US10402756B2 (en) 2005-10-19 2019-09-03 International Business Machines Corporation Capturing the result of an approval process/workflow and declaring it a record
US7856436B2 (en) 2005-12-23 2010-12-21 International Business Machines Corporation Dynamic holds of record dispositions during record management
US8924335B1 (en) 2006-03-30 2014-12-30 Pegasystems Inc. Rule-based user interface conformance methods
US10838569B2 (en) 2006-03-30 2020-11-17 Pegasystems Inc. Method and apparatus for user interface non-conformance detection and correction
US9658735B2 (en) 2006-03-30 2017-05-23 Pegasystems Inc. Methods and apparatus for user interface optimization
US8037029B2 (en) 2006-10-10 2011-10-11 International Business Machines Corporation Automated records management with hold notification and automatic receipts
US9189361B2 (en) 2007-03-02 2015-11-17 Pegasystems Inc. Proactive performance management for multi-user enterprise software systems
EP2137584A1 (en) * 2007-04-12 2009-12-30 Thomson Licensing Centralized work flow monitoring
GB2460730A (en) * 2008-06-12 2009-12-16 Ibm Distributed workflow management system
US10467200B1 (en) 2009-03-12 2019-11-05 Pegasystems, Inc. Techniques for dynamic data processing
US9678719B1 (en) 2009-03-30 2017-06-13 Pegasystems Inc. System and software for creation and modification of software
US8880487B1 (en) 2011-02-18 2014-11-04 Pegasystems Inc. Systems and methods for distributed rules processing
US9270743B2 (en) 2011-02-18 2016-02-23 Pegasystems Inc. Systems and methods for distributed rules processing
US9195936B1 (en) 2011-12-30 2015-11-24 Pegasystems Inc. System and method for updating or modifying an application without manual coding
US10572236B2 (en) 2011-12-30 2020-02-25 Pegasystems, Inc. System and method for updating or modifying an application without manual coding
US10469396B2 (en) 2014-10-10 2019-11-05 Pegasystems, Inc. Event processing with enhanced throughput
US11057313B2 (en) 2014-10-10 2021-07-06 Pegasystems Inc. Event processing with enhanced throughput
US10698599B2 (en) 2016-06-03 2020-06-30 Pegasystems, Inc. Connecting graphical shapes using gestures
US10698647B2 (en) 2016-07-11 2020-06-30 Pegasystems Inc. Selective sharing for collaborative application usage
US11048488B2 (en) 2018-08-14 2021-06-29 Pegasystems, Inc. Software code optimizer and method
US11567945B1 (en) 2020-08-27 2023-01-31 Pegasystems Inc. Customized digital content generation systems and methods

Also Published As

Publication number Publication date
US7603649B2 (en) 2009-10-13
CA2297065A1 (en) 1999-02-04
AU8664998A (en) 1999-02-16
EP0996916A1 (en) 2000-05-03
US7010602B2 (en) 2006-03-07
US20020059466A1 (en) 2002-05-16
US6338074B1 (en) 2002-01-08
US20030093458A1 (en) 2003-05-15

Similar Documents

Publication Publication Date Title
US6338074B1 (en) System for enterprise-wide work flow automation
JP5113119B2 (en) Computer-executable workflow control system
US7430514B1 (en) System and method for processing insurance claims using a table of contents
US7401112B1 (en) Methods and apparatus for executing a transaction task within a transaction processing system employing symmetric multiprocessors
US6725445B1 (en) System for minimizing notifications in workflow management system
AU2003204420B2 (en) Systems and methods for work list prediction
US20020049573A1 (en) Automated system and method for designing model based architectures of information systems
US20030171947A1 (en) system and method for enterprise-wide business process management
JP2006501569A (en) Deployment of a multi-enterprise planning model to a cluster of application servers
US7024669B1 (en) Managing workload within workflow-management-systems
US20010056422A1 (en) Database access system
CA2400296A1 (en) A system and method for automating the assembly, processing and delivery of documents
JP2006501577A (en) Node level modification during enterprise planning model execution
US20080306806A1 (en) Methods and apparatus for dynamically allocating tasks
CA2371445A1 (en) Customer lead management system
CN102317907A (en) Working environment generation system, working environment generation method, and storage medium
US20020120716A1 (en) Server frame work for a database server
US20040093226A1 (en) Distributed method and system for managing volunteers and other individuals engaged in fundraising campaign
CN114021056A (en) Dynamic partitioning method and partitioning system for ICT (information communication technology) resource link virtual operation platform
US20020178037A1 (en) Load balancing system and method
US5432901A (en) Method of dynamically generating a local format for use by a logical unit in a VTAM-type communications session
CN108809715A (en) A kind of method and device of deployment management platform
O'Brien et al. Agents of change in business process management
Howell et al. Architecture for Payload Planning System (PPS) Software Distribution
Mead The multiuser calculator: an operating system project

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AU CA JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 86649/98

Country of ref document: AU

ENP Entry into the national phase

Ref document number: 2297065

Country of ref document: CA

Ref country code: CA

Ref document number: 2297065

Kind code of ref document: A

Format of ref document f/p: F

WWE Wipo information: entry into national phase

Ref document number: 1998938035

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1998938035

Country of ref document: EP

WWW Wipo information: withdrawn in national office

Ref document number: 1998938035

Country of ref document: EP