A method for digital granular business customization and automated execution

Abstract

The application discloses a kind of based on digital granule enterprise business customization and automation execution method, belong to enterprise business management technical field.Based on digital granule enterprise business customization and automation execution method, one enterprise business is abstracted as a model, a model is divided into metadata model, meta behavior model, meta relationship model according to data, behavior, inter-model relationship, each meta model contains corresponding attribute element, behavior element, relationship element, and the combination of attribute element, behavior element, relationship element, meta model realizes the customization of business between each other.Through state machine, the behavior element configured by user is read, the corresponding execution logic is extracted, and then the execution process of business is constructed, so as to realize the automation construction and execution of business process;User only needs to execute the behavior and order to be executed of specified business, without other configuration, reduce the difficulty of user's on hand, save secondary development cost.

Description

Technical Field

[0001] This invention relates to the field of enterprise business management technology, and in particular to a method for customizing and automating enterprise business processes based on digital granularity. Background Technology

[0002] The demand for digital transformation among enterprises is growing, but a common problem exists: low customizability of systems. Current enterprise digitalization efforts mostly focus on a few specific business processes, developing related work based on these processes. Expanding these systems requires developers to redesign and redevelop the system. The main reason for this problem is that these studies or projects lack a high degree of abstraction of the enterprise's business at the system design level, failing to identify the commonalities between business processes. This results in each business process having to be designed and developed independently. This not only leads to complex processes and lengthy cycles but also generates high human resource and system reconstruction costs.

[0003] A deep analysis of the technical roots reveals that the core problem with existing solutions lies in the insufficient abstraction of enterprise business at the system design level. Specifically, this manifests in several ways: the lack of inherent consistency between different business processes leads to each business being developed as an independent entity, resulting in non-reusable business models; the absence of standardized business decomposition and reorganization mechanisms makes it difficult to achieve business customization through configuration, necessitating code-level modifications; and poor system scalability when business logic changes, hindering rapid response to dynamic enterprise needs and severely restricting the efficiency and flexibility of digital transformation. Therefore, constructing a highly abstract, reusable, and easily scalable business customization and automated execution system has become an urgent technical challenge. Summary of the Invention

[0004] The purpose of this invention is to provide a method for enterprise business customization and automated execution based on digital granularity. This method abstracts each management business into a meta-model of the same type, further divides the meta-model, and transforms the business customization process into a meta-model combination process. By reading user-configured behavioral elements through a state machine, the corresponding execution logic is extracted, and then the business execution flow is constructed, thereby achieving automated construction and execution of the business process. Users only need to execute the specified behaviors and sequences of the business, without needing to perform any other configurations, reducing the learning curve for users and saving secondary development costs.

[0005] To achieve the above objectives, this invention provides a method for enterprise business customization and automated execution based on digital granularity, comprising the following steps:

[0006] S1. A business abstraction method based on digital granules, which abstracts enterprise business into a model, wherein the model consists of a set of meta-models composed of attribute elements, behavior elements, and relation elements;

[0007] S2, Business Customization Method: Forming a target business model by combining different meta-models;

[0008] S3. Based on a microservice architecture, implement the creation and combination of attribute elements, behavior elements, relation elements, and meta-models; based on behavior-driven finite state machines, realize the automated execution of business processes.

[0009] Preferably, in step S1, the business abstraction method based on digital granules specifically includes:

[0010] Each business corresponds to a model, which is abstracted as a collection of several types of digital particles, and the digital particles are the elements of the model;

[0011] The elements include attribute elements, behavior elements, and relation elements, which respectively correspond to the data unit, behavior unit, and inter-business call relationship of the business;

[0012] The meta-model is used as a container for meta-elements. The meta-model includes a metadata model, a meta-behavioral model, and a meta-relational model, which manage attribute meta-elements, behavioral meta-elements, and relational meta-elements, respectively.

[0013] Preferably, the attribute element is the smallest data unit of the business model, including six types: text, file, date, drop-down menu, radio button, and checkbox, which serve as a data container to store business form data;

[0014] The behavioral element is the smallest behavioral unit of the business model, including thirteen types: review, event result judgment, branch, loop, database table creation, database table deletion, database table modification, database table reading, database table writing, start, rollback, end, and termination, corresponding to the node behaviors of the business process diagram.

[0015] The relational element is used to connect different business models, including two types: fixed trigger and conditional trigger, to enable calls between business models.

[0016] Preferably, the metadata model serves as a container for attribute elements, generating corresponding data forms;

[0017] The meta-behavioral model serves as a container for behavioral elements, generating corresponding execution flows.

[0018] The meta-relation model acts as a container for relational elements, triggering other business behaviors.

[0019] Metamodels enable the reuse of metadata and hierarchical management of models.

[0020] Preferably, in step S2, the business customization method specifically includes:

[0021] Different meta-models are created by combining different meta-models, and then different meta-models are combined to form the target business model, thereby achieving business customization.

[0022] Preferably, in S3, the behavior-driven finite state machine is a quintuple (A, a0, ∑, F, δ), where A is a finite set of behaviors, a0 is the initial behavior, a0∈A, ∑ is the set of behavior transition conditions, F is the set of terminating behaviors, and δ is the state transition function.

[0023] The finite set of behaviors A contains the specific execution logic of the behavior element, the transfer conditions ∑ include unconditional transfer and branch judgment conditions, the set of termination behaviors F contains two states: end and termination; the state transition function δ supports nondeterministic transfer, where a behavior can be transferred to multiple behaviors based on conditions.

[0024] Preferably, the implementation of the behavior-driven finite state machine is based on state patterns, including:

[0025] Define an abstract state interface. Each state of the finite state machine corresponds to a concrete state class that inherits from the abstract state interface. Nodes of the behavior-driven finite state machine are constructed from behavior elements. Each node contains the specific execution logic of the behavior element and the state identifier.

[0026] Define a context class to enable transitions between different states through external input;

[0027] For a behavior-driven finite state machine (A, a0, ∑, F, δ) represented by a quintuple, A corresponds to the specific state class constructed by the behavior element in the state pattern. During execution, the node first executes its own specific execution logic, and then completes the transition between behaviors based on the current conditions in ∑. The behavior-driven finite state machine instance serves as the context class to control the transition between behaviors.

[0028] Preferably, in step S3, the data transmission method for business process execution includes:

[0029] The behavioral meta parameters are divided into initial data and form data;

[0030] The initial data is the data specified when creating the behavior element. The initial data is configured when the user creates the behavior element, and the data configured by the user is read directly during execution.

[0031] Form data is user data read when the behavior metadata is executed, after the user fills out the form generated by the corresponding metadata model;

[0032] Initial data and form data are stored in a hash table, and parameters are passed when the state machine node is executed.

[0033] Preferably, the initial data is stored as key-value pairs; the form data is associated with the attribute element name and the attribute element ID in the corresponding data table, and the data in the corresponding database table is read through the attribute element ID during execution.

[0034] Preferably, in S3, the microservice architecture includes:

[0035] The metadata model module manages attribute metadata and data form generation;

[0036] The meta-behavioral model module manages behavioral elements and the construction of execution processes.

[0037] The meta-relation model module manages relational elements and inter-business calls;

[0038] The state machine execution module uses a behavior-driven finite state machine to automate process execution.

[0039] Microservice architecture supports the distributed creation and composition of meta-models, enables inter-module communication through service registration and discovery mechanisms, and supports the dynamic expansion and reuse of business models.

[0040] The advantages and positive effects of the digital particle-based enterprise business customization and automated execution method described in this invention are:

[0041] 1. This invention abstracts business into attribute elements, behavior elements, relation elements, and meta-models through "digital granules," forming standardized and reusable units. Business expansion does not require secondary development, and enterprise managers can independently complete customization through "meta-model combination," thereby reducing costs.

[0042] 2. This invention is based on a behavior-driven finite state machine (ADFSM) with a quintuple (A, a0, ∑, F, δ). It encapsulates behavioral logic and transition conditions through state patterns. The process execution does not require manual intervention, has a high degree of automation, a high success rate, and a short response time. The nondeterministic transition function δ supports complex branch logic and has high adaptability.

[0043] 3. This invention breaks down business into microservice modules such as metadata models and metabehavioral models. Distributed creation and composition are achieved through service registration and discovery. Modules can be deployed and upgraded independently, resulting in high operational efficiency and improved throughput.

[0044] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0045] Figure 1 This is a flowchart of the present invention;

[0046] Figure 2 This is a schematic diagram illustrating the business abstraction of the present invention;

[0047] Figure 3 A schematic diagram illustrating the management operations of this invention;

[0048] Figure 4The present invention provides a leave application process and its abstracted behavioral elements; (a) is a leave application flowchart; (b) is an abstract behavioral element diagram.

[0049] Figure 5 This is a schematic diagram of the state-pattern-based ADFSM of the present invention;

[0050] Figure 6 This is the UML class diagram of the behavior nodes of this invention;

[0051] Figure 7 This is a UML diagram of an ADFSM instance of the present invention;

[0052] Figure 8 This invention adds a UML class diagram of the behavior nodes of the hash table;

[0053] Figure 9 A UML diagram of the ADFSM instance of the hash table has been added to this invention. Detailed Implementation

[0054] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. In case of any inconsistency, the meaning set forth in this specification or derived from the content described herein shall prevail. Furthermore, the terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0055] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0056] like Figure 1 As shown, a method for enterprise business customization and automated execution based on digital granularity includes the following steps:

[0057] S1. A business abstraction method based on digital granules abstracts enterprise business into a model, which consists of a meta-model set composed of attribute elements, behavior elements, and relation elements.

[0058] The business abstraction method based on "digital granules" abstracts an enterprise's business into a collection organically composed of several types of "digital granules." First, each business corresponds to a model. Then, each model is further abstracted into a collection of several types of "digital granules," where "digital granules" are the "meta-elements" contained within the model. A business model primarily consists of three types of meta-elements: "attribute meta-elements," "behavioral meta-elements," and "relational meta-elements." These meta-elements are not scattered throughout the model but are uniformly managed by a container called a "meta-model." Each type of meta-element has a corresponding meta-model, and finally, several meta-models organically combine to form a single model. A schematic diagram of business abstraction is shown below. Figure 2 As shown.

[0059] The three types of elements contained in a business model are as follows:

[0060] Attribute element: An attribute element is the smallest data unit contained in a business model. For a business operation, it is often accompanied by several data forms, and each data item in the form is an attribute element of that business model. An attribute element is a "container" of data, not the data itself. This invention designs six types of attribute elements, covering common data types contained in a data form. See Table 1 for details of attribute element types.

[0061] Table 1 Attribute Metatype Table

[0062]

[0063]

[0064] Behavioral Elements: Once the data is available, the business model can be executed. Each enterprise business typically has an execution flowchart, and the behavior corresponding to each process node in the flowchart is the behavioral element of that business model. A behavioral element represents the smallest behavioral unit of a business model. This invention designs thirteen behavioral elements, covering the behaviors that a common management business might execute. See Table 2 for a detailed list of behavioral element types.

[0065] Table 2 Behavioral Meta-Type Table

[0066]

[0067]

[0068] Relational Elements: Each business model is not isolated; they may call each other. For example, a leave application process, in addition to executing the leave application workflow, may also require executing corresponding payroll management procedures if the applicant's leave is too long and their monthly bonus is deducted. Triggering calls between different business processes are "connected" through relational elements. This invention designs relational elements with two different triggering methods, detailed in Table 3.

[0069] Table 3. Two triggering methods for relational elements

[0070]

[0071]

[0072] A business function may contain hundreds or even thousands of different types of elements. Simply abstracting the model as a collection of elements would make the model difficult to manage. Furthermore, different elements may form fixed combinations for reuse in different models; simply abstracting the model as a collection of elements would also make these fixed combinations difficult to reuse. To address these issues, this invention introduces the concept of a meta-model.

[0073] A metamodel can be understood as a container for a certain type of meta-data. A metamodel contains one or more meta-data of the corresponding type. There are three types of metamodels in the system: metadata model, metabehavioral model, and metarelational model, which serve as containers for attribute meta-data, behavioral meta-data, and relational meta-data, respectively. A business model is ultimately composed of several metamodels. The introduction of metamodels not only facilitates the management of business models but also enables the reuse of meta-data. A detailed description of each type of metamodel is shown in Table 4.

[0074] Table 4 Three Types of Metamodels

[0075]

[0076]

[0077] S2, Business Customization Method: Forming a target business model by combining different meta-models.

[0078] After abstracting the enterprise's business into a business model using the above methods, the process of customizing enterprise management business is transformed into a process of combining different meta-models to form a model. That is, different elements are combined to create different meta-models, and then these different meta-models are combined to form the final business model, thus achieving business customization. A diagram illustrating business customization is shown below. Figure 3 As shown.

[0079] The enterprise's customized management system is designed and implemented based on the "digital granular" business abstraction method. It utilizes a microservice architecture to design and implement the three types of meta-models and their creation and combination processes. Table 5 compares common enterprise management systems with the customized business system based on "digital granular" abstraction proposed in this invention.

[0080] Table 5 System Comparison

[0081]

[0082]

[0083] S3. Based on a microservice architecture, implement the creation and combination of attribute elements, behavior elements, relation elements, and meta-models; based on behavior-driven finite state machines, realize the automated execution of business processes.

[0084] Design and Implementation of Behavior-Driven Finite State Machine

[0085] For the execution flow of a business process, this invention designs its execution process as a state machine transition process. For such a state machine, it mainly transitions between different states through different inputs to execute transition functions. The execution flow of a business process is a process of executing one action and then transitioning to the next action to continue execution; these transitions are often unconditional. Therefore, this invention designs an Action-Driven Finite State Machine (ADFSM), which can be formally defined as a quintuple (a, a0, ∑, F, δ). A is a finite set of actions, where an action is the specific execution logic corresponding to an action element; a0 is the initial action of the state machine, a0∈A. ∑ is a set of input action transition conditions; for ordinary actions, this means unconditionally executing the next action; for actions like branches and result judgments, these conditions are the judgment conditions configured for each action element at creation. F is the set of terminating actions. S represents the set of termination behaviors in a traditional state machine. A management process has two termination behaviors: "End" and "Terminate." "End" indicates that the process has been completed normally, while "Terminate" indicates that the process has ended prematurely for some reason. δ is the state transition function, similar to an NFA, δ: A×∑→2 A That is, an action may have no or multiple actions that can be transferred in response to a condition.

[0086] For example, for Figure 4 The diagram shows a simple leave application process and its abstracted behavioral element. The corresponding finite set of behaviors is A = {start, review, result judgment, end, termination}; the initial behavior is a0 = {start}; the transfer condition is ∑ = {leave application approved, leave application not approved, default}, where default means directly transferring to the next behavior; δ is the behavior transfer table shown in Table 6.

[0087] Table 6 Behavior Transfer Table

[0088]

[0089] For an ADFSM(A, a0, ∑, F, δ), the behavior a∈A in the ADFSM can be regarded as the state s∈S in a traditional state machine; the initial behavior a0 of the ADFSM can be regarded as the initial state s0 of a traditional state machine; the transition function of the ADFSM is similar to that of the NFA, both being nondeterministic, meaning they can transition from one state to multiple states. The behavior in the ADFSM can contain output logic, and the behavior itself can be regarded as the execution of output logic, similar to the output function G and output set O of Moore machines and Mealy machines. The terminating behavior in the ADFSM corresponds to the accepting state in a traditional state machine. Based on the above correspondence, the ADFSM possesses characteristics of both Moore machines and Mealy machines, as well as characteristics of DFA and NFA.

[0090] Implementation of behavior-driven finite state machines

[0091] The implementation of ADFSM is based on the State pattern. The State pattern defines an abstract state interface, and each state of the state machine is a concrete state class that inherits from this interface, implementing the logic of the transition function. Simultaneously, the State pattern defines a context class that uses external input to control transitions between different states. For the quintuple representation of ADFSM(A, a0, ∑, F, δ), A is similar to each concrete state class in the State pattern; it is a concrete behavior class constructed from a behavior element, i.e., a node in the ADFSM. A node contains not only the specific execution logic (action) of the behavior element but also a state as an identifier. During execution, each node first executes its own action, and then performs a transition based on the current condition (condition) in the conditional input ∑. Finally, the ADFSM instance acts as the context class in the State pattern, controlling the transitions between behaviors. A schematic diagram of the ADFSM implementation based on the State pattern is shown below. Figure 5 As shown.

[0092] A UML class diagram of a behavior node is as follows Figure 6As shown, the most important aspects of an action node are its state and current action. The `State` type is an enumeration of states, and the `AbstractAction` type is the abstract parent class for all concrete actions. Each type of action node has a corresponding action class that inherits from `AbstractAction` and executes specific logic, while also having a corresponding state enumeration within the `State` enumeration. Secondly, each action node also contains a guard condition for action transition. The `Guard` type is the abstract parent class for all condition classes. For ordinary actions, a null value for `guard` results in a direct action transition, while actions like branching and looping have an instance that inherits from `Guard` and executes specific logic. Finally, each node also needs to maintain a pointer to the next action node (`nextNodePointer`). For special actions like branching, it also needs to maintain pointers to other action nodes that the branch does not satisfy (`otherNodePointer`).

[0093] Each behavior node calls the `DoTransition()` method to transition between behaviors, returning a pointer to the next behavior. This method first calls the specific behavior class of the current node to execute the corresponding logic, then performs a conditional judgment based on the configured conditions, and chooses to return either `nextNodePointer` or `otherNodePointer` based on the judgment result.

[0094] A UML diagram of an ADFSM instance is as follows Figure 7 As shown, firstly, an ADFSM instance generates a list of behavior nodes (nodeList) for the corresponding state machine (StateMachine) based on the list of behavior meta-elements configured in the business model. It also maintains a pointer to the current behavior (currentNodePointer) and the corresponding state (currentState). The state machine operates by retrieving the current behavior node from the list based on the current behavior node pointer, calling the node's DoTransition() method to execute the behavior, updating currentNodePointer based on the return value, and then retrieving the next behavior node and repeating the process until a termination or end node is encountered, at which point the process ends.

[0095] Data transfer methods during business process execution

[0096] A behavior element may require different parameters and involve different data transmissions when executing specific behavioral logic. How to implement this data transmission is also a problem to be considered in realizing the automated execution of business processes. This invention divides the parameters required by a behavior element into two types:

[0097] Initial data: This refers to the data that needs to be specified when creating an action element. For example, for the action "Approve," the corresponding approver needs to be specified; for the action "Branch," the corresponding branching conditions need to be set. This type of data is configured when the user creates the element, and the data configured by the user is directly read during execution.

[0098] Form data: This refers to the user's form data that needs to be read when a behavior element is executed. This type of data cannot be specified when creating the behavior element; it can only be read after the user has filled out the form generated by the corresponding metadata model. Because each piece of form data corresponds to an attribute element, the corresponding attribute element needs to be specified when creating the behavior element.

[0099] For initial data, special strings similar to key-value pairs can be used for storage: "key1:value1; key2:value2". Form data also needs to be stored in the same format, with its corresponding attribute names "key1:name1; key2:name2", and the attribute ID in the corresponding data table for each attribute name also needs to be stored: "name1:id1; name2:id2". Then, during execution, the data from the corresponding database table is retrieved based on the ID.

[0100] When constructing an ADFSM node based on behavioral elements, the system parses the string of initial data contained in the behavioral elements, obtaining several key-value pairs. The ADFSM node needs to add a hash table to store these key-value pairs. The UML class diagram of a node with an added hash table (HashMap) is as follows: Figure 8 As shown.

[0101] For each behavior element containing form data, the system parses the relevant string and organizes it into key-value pairs in the format "attribute element name: attribute element ID". Then, based on these key-value pairs, the system queries the database for the corresponding attribute element data entered by the current user and stores it in the ADFSM instance. An ADFSM instance also needs a hash table to store this data, which is passed as a parameter when each node executes. The UML class diagram for an ADFSM instance with an added hash table is shown below. Figure 9 As shown.

[0102] Microservice architecture includes:

[0103] The metadata model module manages attribute metadata and data form generation;

[0104] The meta-behavioral model module manages behavioral elements and the construction of execution processes.

[0105] The meta-relation model module manages relational elements and inter-business calls;

[0106] The state machine execution module uses a behavior-driven finite state machine to automate process execution.

[0107] Microservice architecture supports the distributed creation and composition of meta-models, enables inter-module communication through service registration and discovery mechanisms, and supports the dynamic expansion and reuse of business models.

[0108] To demonstrate the effectiveness of the method described in this invention, stress tests were conducted on the task scheduling interface constructed using the method described in this invention, focusing on aspects such as request success rate, service throughput, and request response time. Simultaneously, a control group using the native Activiti scheduling method was used to highlight the performance advantages of this system in service orchestration. Performance test results using the native Activiti method are shown in Table 7, and performance test results using the method described in this invention are shown in Table 8.

[0109] Table 7 Performance test results using the native Activiti method

[0110]

[0111] Table 8 Performance test results using the method described in this invention

[0112]

[0113] As can be seen from the table, compared with Activiti's native synchronous call method, the performance indicators of the enterprise business customization machine automation execution method and system based on digital granules proposed in this invention are significantly improved.

[0114] Regarding request success rate, the Activiti native method shows a gradual decrease in success rate as the number of concurrent users increases, dropping to 70.82% when the number of concurrent users reaches 500. In contrast, the enterprise business customization automated execution method and system based on digital granularity proposed in this invention exhibits higher stability and reliability under high concurrency conditions, maintaining a 100% success rate even with 500 concurrent users.

[0115] Regarding average response time, the native Activiti method shows a significant increase in average response time with the increase in the number of concurrent users, rising from 1246.45 milliseconds with 50 concurrent users to 6774.23 milliseconds with 500 concurrent users. However, the enterprise business customization automated execution method and system based on digital granularity proposed in this invention exhibits significantly better response time than the native Activiti method, demonstrating outstanding performance under high concurrency. The average response time shows a relatively gradual increase throughout the testing process, rising from 142.63 milliseconds with 50 concurrent users to 681.27 milliseconds with 500 concurrent users.

[0116] In terms of throughput, the throughput of the native Activiti method gradually increases with the number of concurrent users, from 24.25 TPS with 50 concurrent users to 40.73 TPS with 500 concurrent users. However, the throughput of the enterprise business customization machine automation execution method and system based on digital granularity proposed in this invention is significantly higher than that of the native Activiti method, increasing from 52.07 TPS with 50 concurrent users to 286.67 TPS with 500 concurrent users. The throughput of the enterprise business customization machine automation execution method and system based on digital granularity proposed in this invention is far superior to that of the native Activiti method at all concurrent user counts, demonstrating higher processing capabilities.

[0117] Therefore, the digital granular enterprise business customization and automated execution method described in this invention abstracts each management business into a meta-model of the same type, further divides the meta-model, and transforms the business customization process into a meta-model combination process; by reading the user-configured behavioral elements through a state machine, extracting the corresponding execution logic, and then constructing the business execution flow, the automated construction and execution of the business process is realized; the user's learning curve is reduced, and secondary development costs are saved; it has the advantages of high success rate, high adaptability, short response time, and high throughput.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for enterprise business customization and automated execution based on digital granularity, characterized in that, Includes the following steps: S1. A business abstraction method based on digital granules, which abstracts enterprise business into a model, wherein the model consists of a set of meta-models composed of attribute elements, behavior elements, and relation elements; In S1, the business abstraction method based on digital granules is specifically as follows: Each business corresponds to a model, which is abstracted as a collection of several types of digital particles, and the digital particles are the elements of the model; The elements include attribute elements, behavior elements, and relation elements, which respectively correspond to the data unit, behavior unit, and inter-business call relationship of the business; The metamodel is used as a container for meta-elements. The metamodel includes a metadata model, a metabehavioral model, and a metarelational model, which manage attribute meta-elements, behavioral meta-elements, and relational meta-elements, respectively. S2, Business Customization Method: Forming a target business model by combining different meta-models; S3. Based on a microservice architecture, implement the creation and combination of attribute elements, behavior elements, relation elements, and meta-models; based on behavior-driven finite state machines, realize the automated execution of business processes; In S3, the behavior-driven finite state machine is a quintuple. ,in, A For a finite set of behaviors, a finite set of behaviors A It contains the specific execution logic of a behavior element; a behavior is the specific execution logic corresponding to a behavior element. a 0 This represents the initial behavior of the state machine. ; It is a set of input behavior transition conditions. The transition conditions include unconditional transition and branch judgment conditions. For ordinary behaviors, the next behavior is executed unconditionally; for behaviors such as branching and result judgment, the conditions are the judgment conditions configured for each behavior element when it is created. F To terminate the set of behaviors, , The set of terminating behaviors in a traditional state machine. F It includes two states: End and Termination. End indicates that the business process has been executed normally, while Termination indicates that the business process has ended prematurely. δ The state transition function. δ It supports nondeterministic transitions, where an action can transition to multiple actions based on conditions, and an action can transition to actions where a condition is either absent or present.

2. The method for enterprise business customization and automated execution based on digital granularity as described in claim 1, characterized in that: The attribute element is the smallest data unit of the business model, including six types: text, file, date, drop-down menu, radio button, and checkbox, which serve as a data container to store business form data; The behavioral element is the smallest behavioral unit of the business model, including thirteen types: review, event result judgment, branch, loop, database table creation, database table deletion, database table modification, database table reading, database table writing, start, rollback, end, and termination, corresponding to the node behaviors of the business process diagram. The relational element is used to connect different business models, including two types: fixed trigger and conditional trigger, to enable calls between business models.

3. The method for enterprise business customization and automated execution based on digital granularity as described in claim 2, characterized in that: The metadata model acts as a container for attribute elements, generating corresponding data forms; The meta-behavioral model serves as a container for behavioral elements, generating corresponding execution flows. The meta-relation model acts as a container for relational elements, triggering other business behaviors. Metamodels enable the reuse of metadata and hierarchical management of models.

4. The method for enterprise business customization and automated execution based on digital granularity as described in claim 3, characterized in that, In S2, the business customization method is specifically as follows: Different meta-models are created by combining different meta-models, and then different meta-models are combined to form the target business model, thereby achieving business customization.

5. The method for enterprise business customization and automated execution based on digital granularity as described in claim 4, characterized in that, The implementation of the behavior-driven finite state machine is based on state patterns and includes: Define an abstract state interface. Each state of the finite state machine corresponds to a concrete state class that inherits from the abstract state interface. Nodes of the behavior-driven finite state machine are constructed from behavior elements. Each node contains the specific execution logic of the behavior element and the state identifier. Define a context class to enable transitions between different states through external input; For behavior-driven finite state machines represented by quintuples , A The corresponding state pattern is a specific state class constructed by the behavior element; during execution, the node first executes its own specific execution logic, and then completes the transition between behaviors based on the current conditions in ∑. The behavior-driven finite state machine instance serves as the context class to control the transition between behaviors.

6. The method for enterprise business customization and automated execution based on digital granularity as described in claim 5, characterized in that, In S3, the data transmission method for business process execution includes: The behavioral meta parameters are divided into initial data and form data; The initial data is the data specified when creating the behavior element. The initial data is configured when the user creates the behavior element, and the data configured by the user is read directly during execution. Form data is user data read when the behavior metadata is executed, after the user fills out the form generated by the corresponding metadata model; Initial data and form data are stored in a hash table, and parameters are passed when the state machine node is executed.

7. The method for enterprise business customization and automated execution based on digital granularity as described in claim 6, characterized in that: The initial data is stored as key-value pairs; the form data is associated with the attribute element name and the attribute element ID in the corresponding data table, and the data in the corresponding database table is read through the attribute element ID during execution.

8. The method for enterprise business customization and automated execution based on digital granularity as described in claim 7, characterized in that, In S3, the microservice architecture includes: The metadata model module manages attribute metadata and data form generation; The meta-behavioral model module manages behavioral elements and the construction of execution processes. The meta-relation model module manages relational elements and inter-business calls; The state machine execution module uses a behavior-driven finite state machine to automate process execution. Microservice architecture supports the distributed creation and composition of meta-models, enables inter-module communication through service registration and discovery mechanisms, and supports the dynamic expansion and reuse of business models.

Images