Abstract Yes: Workflow technology is one of the core technologies in office automation systems. Based on the features of the workflow in document flow, this paper analyzes the document flow data model established in the relational database, and provides the implementation method of the Document Flow System Based on this data model.
Office automation is a new office method that combines modern office and computer network functions, and is the product of the information society. The core of the OA system is the workflow. due to historical reasons, the document-type database Domino. It once dominated the Official Document System market with its complete workflow design. Because Domino is a non-structured database, it is difficult to integrate with other business systems because it forms a heterogeneous environment with mainstream relational databases of business systems. Building an OA workflow model in a relational database not only simplifies the integration of the OA system with other business systems, but also avoids the disadvantages of using existing workflow products, this makes the OA system more flexible to meet the needs of different enterprises.
2.OAWorkflow in document flow
2.1Basic concept of Workflow
According to the definition of the Workflow management Alliance (WfMC), a Workflow is a computer or automated whole or part of the business process. In Process Automation, files, information, or tasks are transmitted between participants according to predefined rule sets to achieve the overall business goal or assist in achieving the business goal. Simply put, a workflow is a series of interconnected and automatic business activities or tasks. The main feature of workflow technology is process automation, which includes the combination of human and machine-based activities, especially those that interact with IT applications and tools.
2.2 Features of workflows in document flow
Unlike workflows in the general sense, workflows in the document circulation system have their own characteristics and special requirements. Document circulation refers to the process of transferring office documents from preparation to archiving and approval by each department.
The document circulation mode can be summarized into the following four modes: DC mode, shunting mode, confluence mode and reflux mode.
The DC mode is the simplest transfer mode, indicating one-to-one transmission of documents among personnel. It is usually used for approval process application. For example, a document information sheet is directly handed over to the next operator after being approved by the department manager. The shunting mode indicates one-to-multiple transfer between employees in the enterprise, such as some notifications. The confluence mode indicates the delivery of multiple to one personnel among enterprises. For example, the information sheet sent by the General Manager to the managers of multiple lower-level departments is summarized to the general manager after being processed by different managers. The backflow mode indicates the possible return and redo situations during the document circulation process, so as to form a loop in the process, which is also very common.
These basic streams can be combined to form more complex streams. In addition, through the study of the above document circulation mode, you can integrate and establish the relationship between each operation node.
3. Link-based structureOADocument Flow Data Model
3.1 OADocument Flow Relationship Model
The OA document flow relationship is the flow relationship between each operation node (that is, the operation step). An operation node represents an operation task that cannot be divided, such as viewing, instructions, and execution. Multiple operation nodes use different link links to form the basic process of the workflow relationship of document flow. Therefore, analyzing the relationship between Operation nodes is essential for the implementation of document flow.
(1)And type: For a node in the document flow, the process can proceed to the next step only when all its previous work nodes are completed. If a document needs to be reviewed by the General Manager and multiple deputy general managers, the workflow can enter the next node. The operator of the current node can perform operations.
(2) Or type: For a node in the document flow, one of all its forward nodes completes. The process can enter the next node and execute the current operation node. If a document is handed over to multiple persons for execution, as long as one person completes the operation, the process can flow down, regardless of whether other persons have completed the operation.
(3)Rejection Type: this is a special link type. When a document needs to be rolled back or redone during the document flow process, it must be rolled back to the corresponding work node.
The above is an abstract analysis of the OA document circulation relationship model based on the document circulation mode. in actual application, the document circulation mode is ever-changing, these relationship types cannot abstract all and various types of workflow relationships, but they can basically summarize the relationships between most work nodes used in the OA document circulation system, lay the foundation for further implementation of the system.
3.2 Database Model in document flow
The document workflow model data and operation data are stored in the relational database table, which facilitates system development and design and utilization of document data. In the OA database, the data tables used for document flow mainly include the document table, document operation type table, document process table, process operation information table, sub-process operation table, and document operation type table.
The document table records all aspects of the document, the document operation category table records the category information of the document, and the document Process Table records the basic information of the process, the process operation information table records the detailed operation steps of the process. The sub-process operation table records the sub-process information involved in each process, the document operation type table records information about the document operation form. Shows the two-dimensional relationship diagram of the database.
Figure 1 Database Relationship Model
4.OAImplementation of document circulation
In the OA system developed for an enterprise in Shanghai, the author adopted the relationship-based model described in this article, with the document flow as the core office automation system. The system is mainly developed to achieve paperless office. The process of transferring internal and external documents of an enterprise is converted into a computer, which can improve work efficiency and supervise the handling process, it can also greatly reduce unnecessary paper waste. Automation of document circulation is a key issue in system implementation.
4.1Document Flow Process Definition
The system provides a flexible custom environment for process templates. You can not only customize common process templates, but also correspond specific process templates to specific enterprise document information in a single phase to achieve independent customization of different information sheet processes.
In the actual document circulation process, a process template will be instantiated based on the actual situation before the document officially enters the circulation. After the instantiation is completed, not only the document flow route and operation nodes are defined, but also the participants of each operation node are specified, the related information is stored in the Document Table, document process table, and process operation information table.
4.2Document circulation process and implementation based on database model
As the core of the office automation system, document circulation is the process from drafting to archiving of documents involved by enterprises, and the process of transferring documents between various departments and different office staff. In the process of circulation, documents may be stored in different branch flows and nested flows at the same time. Figure 6 provides a simple description of the conventional document circulation process.
Figure 2 Document Flow Chart
In the process of document circulation, it is critical to successfully implement the document circulation relationship model, that is, to realize the process relationship between each operation node. Therefore, in the process operation information table of the database, the Operation Node object in the process must contain an important attribute to identify the relationship between them. We name this attribute logical, the value can be 1 or 0. From the perspective of actual application of the enterprise, the logical value of the operation node should be set in different situations during the document circulation process. The node's logical value is set to 1, indicating the relationship between "and". The node's logical value is set to 0, indicating the relationship between "or. The relationship between the "draft" and "Review" nodes in section 6 is "and", and between the "handling A" and "handling B" nodes is "or ".
(1) "and" relationship route transfer policy and implementation
A route composed of "and" link nodes in the document flow process. Each node has a clear sequence. Only after a task node is completed can the operation be performed on the next node. When a document is added to a process, a task processing record is automatically inserted in the process operation information table. The operation sequence number of this record is the first task of the route, and the user number is the user personnel number that processes the task, the operation status is marked as "to-do", indicating that the operation has not been processed. The user queries the process operation information table to find the task to be processed, and then processes the task. If the processing is successful, the operation status is marked as "complete", and then the next task processing record is automatically inserted in the flow information table. This record also shows who handles the record, the operation status is marked as "to-do ".
In such a link route, if the user does not agree and needs to be rolled back, the operation that automatically inserts the next task processing record will be aborted. The operator can choose to roll back to a previous task stage. Delete all task processing records after the task stage record, and reset the operation status of the record to "to-do ", in this way, the document is re-processed from the rollback phase.
(2) "or" type link route transfer policy and implementation
A route composed of "or" type relational nodes in the document flow process. Each node needs to organize its own Control Branch process to complete the task, after the branch controller completes the branch task, the document is directly transferred to the main process for the next step, regardless of whether the branches of other nodes are completed. There is no clear sequence between nodes, which ensures independent process of each branch and does not interfere with each other, and avoids delays in the document flow process due to waiting for the completion of other branches. When a document is added to a process, the number of task processing records automatically added in the process operation information table is the number of branch processes for this task, and the operation sequence numbers of each branch process are the same (7 ). The branch controller finds the task to be processed through the query statement and then processes it. If the branch process is successfully processed, the operation status is marked as "completed ", the next task processing record is automatically added to the process operation information table. In this way, when all the branch processes are completed, the number of records in a task in the database flow operation table is equivalent to the number of all branch flows of the task.
Figure 3 The number of process branches is the same as the number of database records.
There are two scenarios for the rollback of such link routes. One scenario is that one person does not agree with the execution phase of each concurrent branch process and can roll back to a single task user or a user of the branch task, delete all task processing records after the task stage record, and reset the operation status of the record to "to-do ", in this way, the document is re-processed from the rollback phase. In another case, after a branch is completed, the handler of the next task needs to roll back a branch process, and only performs the same database operation on the branch process to be rolled back, because different branches in the database process operation information table correspond to different next task records, the rollback operation has no impact on other branch processes.
The above two types of routes are the most basic document flow routes. in actual application, these two types of routes can form various complex document flow routes, therefore, the implementation of these two types of routes is the basis for the circulation of documents.
4.3 Document Flow tracking
In the OA document circulation system, the document circulation process tracking is mainly to supervise the document circulation, and collect statistics on the information of the current unfinished process, it also provides multiple query methods, such as by time and department, to supervise official documents. After a document is sent by the document initiator, you can track the approval location of the current document (that is, the process node where the document arrives) and view the current status of the document.
In the system, the document operation status bit and whether the document has been read in the process operation information table identify the current state of the document, and track the document by changing the values of the two fields in the database.
Due to the variability and complexity of workflows, workflows have always been a challenge in the computer software industry, and it is difficult to find workflow products that fully satisfy different enterprises in the market. In the office automation system, workflow is the difficulty and core of the system. The document involved by an enterprise, whether it is an internal document or an external document, it must be transferred between different office staff of the company. In an office automation system developed for an enterprise in Shanghai, we used a mainstream database consistent with other business systems to store workflow and document data in an open relational database, this solves many problems such as system integration and information sharing in enterprises.