Workflow automation rule library construction method and device, and medium

Abstract

The present application relates to the field of workflow management. Disclosed are a workflow automation rule library construction method and device, and a medium. The method comprises: acquiring historical business data respectively corresponding to multiple fields and, on the basis of the historical business data, acquiring corresponding ERP modeling data, ERP operational data and ERP business specifications; on the basis of a knowledge graph technology, generating corresponding check items; determining docking interfaces for a rule library and each check item in an ERP knowledge graph; classifying business workflow automation rules; and, on the basis of an operation of a user in the rule library, defining in the rule library a business workflow automation rule by means of the check items. Compared with a mode in which an expression is created by a human, the present application increases the degree of automation and reduces situations of rule omission. Compared with a mode in which a checking and verifying component is used, the present application can also define rules of complex business scenarios by means of relationships in knowledge graphs, so as to ensure the degree of automation and rule definition efficiency.

Description

A process automation rule base construction method, device and medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on March 29, 2024, with application number 202410371855.9 and invention name “A method, device and medium for constructing a process automation rule base”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of workflow management, and specifically to a method, device, and medium for constructing a process automation rule library. Background Art

[0003] Process automation refers to the use of computer programs, technologies, and tools to automatically execute a series of tasks, steps, or processes instead of human operations and intervention. These processes can cover a variety of fields and industries, such as manufacturing, marketing, procurement, service, human resources, finance, etc.

[0004] The process automation rule base in traditional enterprise management software is generally divided into two types:

[0005] 1. By defining expressions, sales staff or process managers can define simple rules using models and expressions. For example, key factors that affect the approval conclusion include whether the amount field on the form is approximately 10,000 yuan, whether the reimbursement department and the apportionment department are the same department, whether accommodation and travel expenses exceed the standard, and whether the document has uploaded images and attachments.

[0006] However, under this type, sales staff or process managers generally need to manually define rules, which not only has a low degree of automation, but also makes it easy for rules to be missed or updated in a timely manner.

[0007] Second, register the business-customized logic as a check and verification component through the check and verification component. For example, if you need to check in real time whether the supplier's credit information meets the remittance conditions, this credit information needs to be obtained through a third-party interface. Alternatively, you need to check whether there are sensitive words in the document approval process. These sensitive words often come from a dedicated and irregularly updated vocabulary library. In this scenario, you need to write a custom check and verification component to perform rule verification.

[0008] However, under this type, when faced with complex business scenarios, it is often necessary to develop customized inspection and verification components, and the components are also program-defined. They can only replace high-frequency repetitive work, such as identifying whether the document contains sensitive words in the vocabulary library. For more complex business scenarios, such as complex issues involving cross-departmental, multi-party participation and resource allocation, supervision of complex systems, processes or financial transactions, especially those involving risk management and compliance, manual review and monitoring are still required to detect potential problems and abnormal situations, which makes its degree of automation still limited.

[0009] Summary of the Invention

[0010] To solve the above problems, this application proposes a process automation rule base construction method, including:

[0011] Based on the associated ERP system, obtain historical business data corresponding to multiple fields, and obtain corresponding ERP modeling data, ERP operation data and ERP business specifications based on the historical business data;

[0012] Generate corresponding inspection items based on the ERP modeling data, the ERP operation data, and the ERP business specifications based on knowledge graph technology, where the inspection items correspond to at least a plurality of entities and relationships connecting the entities;

[0013] Determine the interface between the rule base and each inspection item in the ERP knowledge graph;

[0014] Classifying business process automation rules to obtain multiple rule types, wherein the rule types include at least one of flow rules, execution rules, evolution rules, exception rules, approval rules, and compliance rules;

[0015] Based on the user's operation in the rule base, a business process automation rule is defined in the rule base through the check items.

[0016] On the other hand, the present application also proposes a process automation rule base construction device, including:

[0017] at least one processor; and,

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to execute the process automation rule base construction method as described in the above example.

[0020] On the other hand, the present application also proposes a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as: the process automation rule base construction method described in the above example.

[0021] The process automation rule base construction method proposed in this application can bring the following beneficial effects:

[0022] By building corresponding check items through the knowledge graph and defining business process automation rules based on these check items, compared to manually building expressions, the level of human involvement is reduced, which improves the degree of automation. Furthermore, because the relationship structure between entities in the knowledge graph is more clearly visible, the occurrence of rule omissions is reduced. Furthermore, compared to checking components, the relationships in the knowledge graph can also be used to define rules in complex business scenarios, ensuring the degree of automation and efficiency of rule definition. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] FIG1 is a flow chart of a method for constructing a process automation rule base in an embodiment of the present application;

[0025] FIG2 is a schematic diagram of an automatic approval rule architecture in one scenario according to an embodiment of the present application;

[0026] FIG3 is a schematic diagram of the knowledge graph technology processing process in one scenario in an embodiment of the present application;

[0027] FIG4 is a schematic diagram of the structure of the inspection items in the first scenario in an embodiment of the present application;

[0028] FIG5 is a schematic diagram of the structure of the inspection items in the second scenario in an embodiment of the present application;

[0029] FIG6 is a schematic diagram of the structure of the inspection items in the third scenario in an embodiment of the present application;

[0030] FIG7 is a schematic diagram of an automation rule base definition function interface in one scenario according to an embodiment of the present application;

[0031] FIG8 is a schematic diagram of an interface for adding rules to an automation rule library in one scenario according to an embodiment of the present application;

[0032] FIG9 is a schematic diagram of the interaction between the rule base and the automation process in one scenario in an embodiment of the present application;

[0033] FIG10 is a schematic diagram of a process automation rule base construction device in an embodiment of the present application. DETAILED DESCRIPTION

[0034] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0035] The following describes in detail the technical solutions provided by various embodiments of the present application in conjunction with the accompanying drawings.

[0036] As shown in FIG1 , an embodiment of the present application provides a method for constructing a process automation rule base, including:

[0037] S101: Based on the associated ERP system, historical business data corresponding to multiple fields are obtained, and corresponding ERP modeling data, ERP operation data and ERP business specifications are obtained according to the historical business data.

[0038] The process automation rule base construction method in this application can be applied to the automatic approval rule architecture shown in Figure 2. As shown in Figure 2, it contains multiple parts, namely basic components, rule services, and rule applications. This application mainly focuses on the use of knowledge graphs in basic components, as well as the application of rule definition, rule management, and rule classification in rule services, so as to achieve the construction of a multi-domain ERP process automation rule base (hereinafter referred to as the rule base) in the rule application part. The other parts of Figure 2 are not involved in this application, or are only briefly involved, so they will not be repeated here.

[0039] Pre-link with the Enterprise Resource Planning (ERP) system and obtain the corresponding historical business data through the corresponding interface in the ERP system, so as to further obtain ERP modeling data, ERP operation data and ERP business specifications.

[0040] Specifically, based on the associated ERP system, for each field, multiple execution businesses included in the field and multiple execution operations included in each execution business are determined.

[0041] Taking the compliance of transactions in procurement contract review as an example, the corresponding field is the procurement field, the corresponding execution business is the contract review business, and the corresponding execution operation is the transaction compliance operation.

[0042] Since the purpose of this review is to audit events in process automation, historical business data that failed the review in this area is obtained as analysis samples.

[0043] At this time, according to the business modeling Schema corresponding to the historical business data, the ERP modeling data corresponding to the historical business data is obtained.

[0044] Taking the purchase contract review as an example, the purchase requisition will be business modeled in the ERP system. Therefore, a schema structure that conforms to the knowledge graph specification can be constructed based on the schema of the purchase requisition in the ERP system.

[0045] At this point, the business modeling schema of the purchase contract in the ERP system is taken as an example. Table 1 lists some attributes of the purchase contract.

[0046] Table 1

[0047] Continue to obtain the ERP operation data corresponding to the historical business data based on the execution operation, which refers to the data executed by the user during the process execution. And according to the execution business, obtain the ERP business specifications corresponding to the historical business data, which refers to the corresponding specifications made in advance for process review.

[0048] S102: Generate corresponding inspection items based on the ERP modeling data, the ERP operation data and the ERP business specifications based on the knowledge graph technology, where the inspection items correspond to at least multiple entities and the relationships connecting the entities.

[0049] As shown in FIG3 , for each domain, entities and attributes within the domain are extracted based on its corresponding ERP modeling data, and relationships within the domain are extracted based on its corresponding ERP operation data and ERP business specifications.

[0050] Entities include: domain, business execution, and operation execution, and attributes are used to describe entities.

[0051] At this time, knowledge fusion is performed based on knowledge graph technology, and triples corresponding to each field are generated according to entities and relationships. The form of the triples is: entity, relationship, entity.

[0052] As shown in FIG3 , during the knowledge fusion process, the identified entities can be linked, which can be done automatically or with the assistance of manual annotation to ensure the quality of the data.

[0053] Then, the knowledge in various fields (including production and manufacturing, procurement, human resources, financial sharing, supply chain and other fields in the ERP system, as well as fields provided by third-party libraries) is merged to obtain a public knowledge base.

[0054] The knowledge in the public knowledge base is then subjected to ontology construction, knowledge reasoning, and quality assessment, thereby realizing knowledge processing based on knowledge graph technology, so that it can be subsequently precipitated into rules and applied in rule fields, rule businesses, and rule operations.

[0055] Here, the triples corresponding to each field in this application are fused, and the generated knowledge is called the inspection items corresponding to each field.

[0056] Inspection items refer to various types of inspections required for various fields, business execution, and operations. For example, as shown in Figure 4, the inspection items in the procurement application field include multiple inspection items corresponding to the business execution, including supplier qualification review, contract review, amount review, and application reason review. Furthermore, as shown in Figure 5, the inspection items of the contract review (referring to the procurement contract review) further include multiple inspection items corresponding to the execution operations, including contract subject audit, contract clause audit, contract execution status audit, contract change status audit and other review items.

[0057] Based on this, the obtained triples may include (purchase application inspection items, relationships, supplier qualification review), (purchase application inspection items, relationships, contract review), etc. as shown in Figure 4, and (contract review, relationships, contract subject audit), (contract review, relationships, contract clause audit), etc. as shown in Figure 5.

[0058] In each triple, entities with the same parts are all entities corresponding to the domain or the execution business. Therefore, for each triple, the designated entity of the type of the domain or the execution business is determined.

[0059] Traverse each triple in turn, find the specified triples with the same specified entity, and fuse the specified triples based on the specified entity to generate the corresponding inspection items for each field.

[0060] Among them, the direction of the relationship in the inspection item is from high dimension to low dimension, and the order of the dimensions of each entity from high to low is: field, execution business, and execution operation.

[0061] During fusion, the same dimension can be fused first. For example, the type of entity at the pointing end in the relationship can be determined first, the entity at the field can be selected, and then the same entities at the pointing end can be fused to form the state shown in Figures 4 and 5.

[0062] Continue to fuse the inspection items in Figures 4 and 5, selecting entities with the same pointing and pointed ends for fusion, to form the inspection items shown in Figure 6. At this point, the inspection items are inspection items in the procurement application domain (also known as the procurement domain), and after being extracted in the form of a knowledge graph, they are persisted in the relational database according to the three dimensions of domain, execution business, and execution operation.

[0063] S103: Determine the docking interface between the rule base and each inspection item in the ERP knowledge graph.

[0064] Determine the corresponding application scenario in the rule library. The application scenario refers to the actual scenario corresponding to each check item during the process automation processing. In this actual scenario, the corresponding check item needs to be checked.

[0065] The application scenario is split to generate a scenario split table; the fields in the scenario split table include: scenario elements, scenario name, detailed description; the field values ​​corresponding to the scenario name include: domain, execution business, execution operation, business modeling identifier, and operation data identifier, as shown in Table 2.

[0066] Table 2

[0067] At the same time, determine the interface input parameter table of the docking interface of the ERP knowledge graph; among them, the fields in the interface input parameter table include: input parameter definition, parameter number, parameter name, auxiliary description; the field value of the field corresponding to the parameter name includes: domain identifier, business identifier, operation identifier, business modeling identifier, and operation data identifier, as shown in Table 3.

[0068] Table 3

[0069] At this point, the application scenarios can be described through the scenario splitting table, thereby obtaining the description of each application scenario, and then based on the interface input table, a docking interface is established between each application scenario in the rule base and each inspection item in the ERP knowledge graph.

[0070] After calling the knowledge graph interface, the corresponding event is triggered in the application scenario. At this time, the docking interface is called. The docking interface needs to return the unique identifier of this event and the analysis results. The analysis results mainly include whether the transaction is compliant and the result description.

[0071] Among them, the docking interface can return an event return table, and the analysis results are reflected in the event return table. The fields in the event return table include: parameter number, parameter name, auxiliary description, and the field value corresponding to the parameter name includes: event instance identifier, conclusion, and cause, which can be shown in Table 4.

[0072] Table 4

[0073] S104: Classify the business process automation rules to obtain multiple rule types, where the rule types include at least one of flow rules, execution rules, evolution rules, exception rules, approval rules, and compliance rules.

[0074] Specifically, each rule type may be as shown in Table 5.

[0075] Table 5

[0076] S105: Based on the user's operation in the rule base, define a business process automation rule in the rule base using the check items.

[0077] Specifically, as shown in FIG7 , the automation rule base definition function interface is displayed, and the automation rule base definition function interface includes a domain navigation area, a rule display area, and a rule editing area.

[0078] The domain navigation area includes various areas, where users can select and display the business process automation rules corresponding to each area. The rule display area displays existing business process automation rules for that area, along with their serial number, name, rule type, creator, and last modification time. The rule editing area includes add, modify, and delete functions for operating business process automation rules.

[0079] Based on the user's operation in the rule editing area, when you click Add, the business process automation rules are defined in the rule library through the entities corresponding to each check item.

[0080] As shown in FIG8 , the settings include: domain, execution business, execution operation, rule name, rule type, and rule source, and the rule source at least corresponds to the knowledge graph.

[0081] Among them, the rule name supports custom names. The rule type is an enumeration option through one of the defined business process automation rule categories. The rule source is an enumeration option, and there are three types: expression, custom component, and knowledge graph. For the solution in this application, select the knowledge graph. Field, click the rule field to pop up the field help interface, whose data comes from the field table deposited by the constructed knowledge graph, and you can select the field that has been modeled. Business (also known as execution business), click the business to pop up the business help page, and you can select the business that has been modeled. Its data comes from the business table deposited by the knowledge graph. Operation (also known as execution operation), click the operation to pop up the operation help page, whose data comes from the business table deposited by the knowledge graph, and you can select the business scenario operation that has been modeled.

[0082] By building corresponding check items through the knowledge graph and defining business process automation rules based on these check items, compared to manually building expressions, this approach requires less human involvement and improves automation. Furthermore, because the relationship structure between entities in the knowledge graph is more clearly visible, it reduces the risk of rule omissions. Furthermore, compared to using check components, the relationships in the knowledge graph can also be used to define rules in complex business scenarios, ensuring automation and efficiency.

[0083] The technical solution in this application is mainly aimed at the construction process of the rule base. After the rule base is built, as shown in Figure 9, in the process of process automation, the process runs from start-approval-service-end. After the start, the process is created and the process engine event is triggered. Based on the corresponding application scenarios, different process events are triggered, including flow events, execution events, approval events, exception events, etc.

[0084] At this time, the rule library provides a query interface, sends event instances to the rule library through the process automation rule configuration, and then reads the corresponding business process automation rules through the query interface. The rule library calls through the customized interface, and performs rule reasoning and event prediction through the knowledge graph intelligent decision analysis model (that is, the ERP knowledge graph). The reasoning results are sent to the process engine, which affects the process automation flow process and realizes automated processing.

[0085] During the rule creation process, rule creation is achieved based on rule definition, rule query, rule execution, and classification of process automation rules (including flow rules, execution rules, evolution rules, exception rules, etc.).

[0086] In one embodiment, the rule library provides a query interface to enable users to query the corresponding business process automation rules based on entities. When querying, the rules set in the rule library can be filtered according to the domain, execution business, and execution operation, and the corresponding rules can be configured during process modeling.

[0087] In addition, when the automated process runs, it reads the configured rules. The rule library provides a parsing result interface. When the automated process runs, the configured rules are passed to the rule library parsing result interface, and the rule library then returns the parsing results to the process engine.

[0088] As shown in FIG10 , the embodiment of the present application further provides a process automation rule base construction device, including:

[0089] at least one processor; and,

[0090] a memory communicatively connected to the at least one processor; wherein,

[0091] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the process automation rule base construction method as described in any of the above embodiments.

[0092] An embodiment of the present application further provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as: the process automation rule base construction method described in any of the above embodiments.

[0093] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for constructing a process automation rule base, wherein: include: Based on the associated ERP system, obtain historical business data corresponding to multiple fields, and obtain corresponding ERP modeling data, ERP operation data and ERP business specifications based on the historical business data; Generate corresponding inspection items based on the ERP modeling data, the ERP operation data, and the ERP business specifications based on knowledge graph technology, where the inspection items correspond to at least a plurality of entities and relationships connecting the entities; Determine the interface between the rule base and each inspection item in the ERP knowledge graph; Classifying business process automation rules to obtain multiple rule types, wherein the rule types include at least one of flow rules, execution rules, evolution rules, exception rules, approval rules, and compliance rules; Based on the user's operation in the rule base, defining a business process automation rule in the rule base through the check item; According to the ERP modeling data, the ERP operation data and the ERP business specifications, corresponding inspection items are generated based on the knowledge graph technology, specifically including: For each domain, entities and attributes within the domain are extracted based on the corresponding ERP modeling data, and relationships within the domain are extracted based on the corresponding ERP operation data and ERP business specifications. The entities include: domain, executed business, and executed operation, and the attributes are used to describe the entities. Based on the knowledge graph technology, triples corresponding to each field are generated according to the entities and the relationships. The triples are in the form of: entity, relationship, entity; The triples corresponding to each field are fused to generate the inspection items corresponding to each field; Determine the interface between the rule base and each inspection item in the ERP knowledge graph, including: Determine the corresponding application scenario in the rule base, split the application scenario, and generate a scenario split table; wherein the fields in the scenario split table include: scenario element, scenario name, detailed description; the field value corresponding to the scenario name includes: domain, execution business, execution operation, business modeling identifier, and operation data identifier; Determine the interface input parameter table of the docking interface of the ERP knowledge graph; wherein the fields in the interface input parameter table include: input parameter definition, parameter number, parameter name, auxiliary description; the field corresponding to the parameter name The field values ​​include: domain identifier, business identifier, operation identifier, business modeling identifier, and operation data identifier; The application scenario is described by the scenario splitting table, and based on the interface input parameter table, a docking interface is established between each application scenario in the rule base and each inspection item in the ERP knowledge graph; The method further comprises: Based on the currently triggered event, calling the docking interface, and receiving the event return table corresponding to the event returned by the docking interface; The fields in the event return table include: parameter number, parameter name, auxiliary description, and the field value of the field corresponding to the parameter name includes: event instance identifier, conclusion, and cause.

2. The method according to claim 1, wherein Based on the associated ERP system, obtain historical business data corresponding to multiple fields, and obtain corresponding ERP modeling data, ERP operation data, and ERP business specifications based on the historical business data, specifically including: Based on the associated ERP system, for each field, determine the multiple execution businesses included in the field and the multiple execution operations included in each execution business; And obtain historical business data that failed the review in this field; According to the business modeling schema corresponding to the historical business data, the ERP modeling data corresponding to the historical business data is obtained, and according to the execution operation, the ERP operation data corresponding to the historical business data is obtained, and according to the execution business, the ERP business specifications corresponding to the historical business data are obtained.

3. The method according to claim 1, wherein The triples corresponding to each field are fused to generate the corresponding inspection items for each field, including: For each triple, determine the designated entity whose type is domain and performs business; Traverse each triple in turn, find the designated triples with the same designated entity, and fuse the designated triples based on the designated entity to generate inspection items corresponding to each field; The relationship in the inspection items is directed from high dimension to low dimension, and the order of the dimensions of each entity from high to low is: field, execution business, execution operation.

4. The method according to claim 1, wherein Based on the user's operation in the rule base, the business process automation rules are defined in the rule base through the check items, specifically including: Displaying the automation rule base definition function interface, which includes a domain navigation area, a rule display area, and a rule editing area; Based on the user's operation in the rule editing area, in the rule base, each check item is checked. Define business process automation rules for the corresponding entity; Among them, the settings include: field, execution business, execution operation, rule name, rule type, and rule source, and the rule source at least corresponds to the knowledge graph.

5. The method according to claim 1, wherein The method further comprises: A query interface is provided for the rule library to enable users to query corresponding business process automation rules based on the entity.

6. A process automation rule base construction device, wherein: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the process automation rule base construction method as described in any one of claims 1 to 5.

7. A non-volatile computer storage medium storing computer-executable instructions, wherein: The computer executable instructions are configured as: the process automation rule base construction method according to any one of claims 1 to 5.

Images