Method for adjusting process engine

Through visual editing tools and custom attributes, non-professionals can deploy and adjust the process engine, solving the complexity and redundant node problems of process engine deployment, achieving efficient optimization and cost reduction of the process engine.

CN119005663APending Publication Date: 2024-11-22GUIYANG JINYANG CONSTR DATA SERVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411210571.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The prior art is difficult to help non-professional personnel deploy and adjust process engines, especially in complex business processes, which require professional knowledge and skills, are costly, and have problems with redundant and invalid node arrangements.

Method used

By obtaining the process engine application scenarios to be edited, using visual editing tools to create process nodes, user-defined attributes and connection relationships, automatically recommend process nodes, simulate running optimization flowcharts, and help non-professionals complete process engine deployment and adjustment.

Benefits of technology

It lowers the threshold for process engine deployment and adjustment, automatically recommends process nodes and connection relationships, identifys and removes redundant and invalid nodes, optimizes flow charts, improves the efficiency and accuracy of process engines, and reduces costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005663A_ABST
    Figure CN119005663A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of process management, in particular to a process engine adjusting method. Comprising the following steps: S1, obtaining a to-be-edited process engine application scene, and identifying an application demand according to a reference scene; s2, creating a visual canvas through a process engine editing tool, acquiring a plurality of process nodes from a pre-configured process node library according to application requirements, and adding the process nodes into the visual canvas; s3, flow nodes selected by a user and self-defined attributes given to the flow nodes are obtained, initialization attributes of the flow nodes are determined according to the self-defined attributes, and connection relations of the flow nodes are determined according to the initialization attributes, the self-defined attributes and the application scenes of the flow nodes; the initialization attributes comprise node input, node flow directions and flow direction quantity, and deploying the process engine according to the connection relationship of the process nodes. And non-professionals can be helped to complete deployment of the process engine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of process management, and in particular to a process engine adjustment method. Background Art

[0002] A process is a combination of a series of activities. For example, in the OA system of an enterprise office, there are a large number of application approval processes. In the manufacturing industry, there are a large number of production and sales processes from sales orders to manufacturing and then to payment collection.

[0003] With the development of automated information technology, it has become an inevitable development trend to manage business processes using automated business process systems. In terms of process implementation, there are two main implementation methods. One is to use code to implement. For example, to implement a business process for overtime application with code, you need to connect to SSO single sign-on, obtain the information of the initiator and approver through the interface, and save the form data at the same time. However, with the changes in business processes, this management method requires developers to recompile the code when the business process changes, and its management complexity is very high. Especially when it comes to business processes of complex businesses, the current hard-coded editing method is costly. Another way is to use process induction to implement it. The process engine connects to the data required by the application scenario, such as overtime application, and the process engine connects to SSO, OU, approver configuration, permissions, etc. To implement such a process, you only need to care about the process configuration, process nodes and process forms. The process flow and process data processing are all completed by the process engine. At present, common process flow engine designers can be implemented through visual programming to edit the configuration files of each process node. Although process engine deployment can be achieved through visual programming, the deployment of the process still requires certain professional knowledge and skills. For non-technical personnel, there is a certain usage threshold. It is impossible to clearly determine the flow direction of each node, and it is easy to arrange some redundant nodes or invalid nodes. Summary of the invention

[0004] The technical problem solved by the present invention is to provide a process engine adjustment method, which can help non-professionals complete the deployment of the process engine.

[0005] The basic solution provided by the present invention is a method for adjusting a process engine, comprising the following steps: S1. Obtain the process engine application scenario to be edited and identify the application requirements based on the referenced scenario; S2. Create a visualization canvas through the process engine editing tool, obtain several process nodes from the pre-configured process node library according to application requirements, and add them to the visualization canvas; S3. Obtain the process node selected by the user and the custom attributes assigned to the process node, determine the initialization attributes of the process node according to the custom attributes, determine the connection relationship of each process node according to the initialization attributes, custom attributes and application scenarios of each process node, the initialization attributes include node input, node flow direction and flow direction quantity, and deploy the process engine according to the connection relationship of the process nodes; S4. Run the process engine, input different initial data, obtain output results, generate an output report based on the output results of the process engine, and optimize the process nodes based on the output report; S5. Generate a flowchart according to the deployment of the process engine, associate the flowchart with the process engine, and publish the flowchart to the departments related to the process engine reference.

[0006] The principle and advantage of the present invention are: first, the application scenarios caused by the process are obtained, such as leave approval, payment approval, etc. After that, the recommended process nodes are retrieved according to the application scenarios, added to the visual canvas, and the process nodes selected by the user are obtained. After the user selects the process node, custom attributes are assigned to each process node, such as approver, approval standard, approval waiting time, etc. After that, according to the custom attributes of the process node, the initialization attributes are determined, and the output attributes of the process node are determined. The initialization attributes include node input, node flow direction, and flow quantity. The node input is the process node from which the data input of the process node comes, the node flow direction is the node to which the output result of the node is to flow, and the flow quantity refers to whether the process node will flow to multiple nodes according to different results. For example, in the payment approval, according to the different payment amounts, more than 50,000 yuan requires approval by the director, 10,000-50,000 yuan requires approval by the manager, and less than 10,000 yuan requires approval by the minister. According to the connection relationship of the process nodes, after the process engine is deployed, the process engine is run to obtain the running results, and a report is generated. The process engine is optimized according to the report and then released. At the same time, according to the part of the process engine, it is sent to the relevant departments and personnel. Compared with the existing technology, it can automatically recommend the required process nodes, allowing users to enter the custom attributes of each node and automatically generate the connection relationship of the process nodes. After that, the process nodes are also operated and optimized to help non-professionals complete the layout of the process engine.

[0007] Further, the S3 comprises the following steps: S31, in addition to the start node and the end node, the process nodes are divided into judgment nodes, approval nodes and approval analysis nodes according to energy-saving functions; S32, the custom attributes of the judgment node include a judgment strategy, and the node input, node flow direction and flow quantity of the judgment node are determined according to the judgment strategy; S33, the custom attributes of the approval node include a node level, and the node input and node flow direction of the approval node are determined according to the node level; S34, the custom attributes of the approval analysis node include an audit result analysis strategy, and the node flow direction of the approval analysis node is determined according to the audit result analysis strategy; S35. Generate a connection relationship between each process node according to the node input, node flow direction and flow quantity of each process node.

[0008] In addition to the start node and the end node, the process nodes are divided into judgment nodes, approval nodes, and approval analysis nodes according to their functions. Judgment nodes refer to the judgment of inputs, and the flow is directed back to different nodes according to the input content. For example, in the payment approval, a process node for payment amount judgment is set, and a judgment node for "whether the amount exceeds 50,000" and "whether the amount is less than 10,000" is set. At the same time, 3 approval nodes are set, including "manager approval", "director approval", and "minister approval". The user sets the node level of the 3 approval nodes, "director approval", which is higher than "manager approval" and "higher than minister approval". At this time, the node input, node flow, and flow output of each process node are determined according to the custom attributes of the approval node and the judgment node, "whether the amount exceeds 50,000" The node input is the start node, and the number of flows is two, one flows to the judgment node of "Is the amount less than 10,000", and one flows to "Director's approval". The node flow directions of "Is the amount less than 10,000" are two, one flows to "Manager's approval" and one flows to "Minister's approval". After that, you can also set the "Approval Analysis Node". The input nodes of the "Approval Analysis Node" are "Director's Approval", "Manager's Approval" and "Minister's Approval". The Approval Analysis Node is used for subsequent operations after analyzing and identifying the approval results. For example, after identifying that the approval is passed, the approval will be copied to the finance department, or after identifying that the approval is not passed, it is directly ended or returned to the start node to resubmit the approval.

[0009] Further, the S4 comprises the following steps: S41, obtaining the original input data of the historical process in the application scenario, and obtaining the output result of the end node; S42, generating an output report, wherein the output report includes the running times of each process node and the node output; S43. According to the number of operation times of the process nodes and the node outputs, invalid nodes and redundant nodes are identified, invalid nodes are removed, and redundant nodes are integrated.

[0010] Further, S43 includes the following steps: S431. Generate a judgment number threshold according to the total number of runs of the process engine, and mark the process nodes whose run times are lower than the judgment number threshold as invalid nodes; S432. According to the output results of the process engine, identify the process nodes with different outputs in the same output results. When there are multiple process nodes output by nodes with the same node function, determine them as suspected redundant nodes. Merge the suspected redundant nodes and run them again to identify whether the output results are consistent. If they are consistent, retain the integrated process nodes. If they are inconsistent, maintain the original process nodes.

[0011] By running the process engine, invalid nodes and redundant nodes are identified according to the number of runs of each process node in the process engine and the node output. The input data during the simulation run is the historical original data of the application scenario. If the relevant process application was originally used in the application scenario, it can be directly retrieved through the API interface. If the original sampled paper process is obtained by scanning and identifying the paper document. Only by reflecting the actual process needs of the enterprise through the original data can invalid nodes and redundant nodes be identified more accurately. For example, when the number of runs of the process engine is 100 times and the number of runs of a process node is 1, the process node is an invalid node and can be removed. Or if the running results of a process node are different and have no effect on the output results, it may be a redundant node and can be integrated. For example, the above-mentioned "Is the amount more than 50,000" and "Is the amount less than 10,000" two judgment nodes can be integrated into a "Is the amount higher than 50,000 and less than 10,000" judgment node, which directly flows to the three "approval nodes".

[0012] Further, the method further comprises the following steps: S61, receiving process adjustment information about a process engine from a user, wherein the process adjustment information is an adjusted flow chart of the process engine; S62, calling the flowchart associated with the process, identifying the difference between the flowchart and the adjusted flowchart, and redeploying the connection relationship and custom attributes of the process nodes according to the difference; S63. Execute S4 on the redeployed process node and publish it again.

[0013] If the process needs to be adjusted later, the user only needs to modify the adjusted flowchart. After identifying the changed process nodes according to the flowchart, readjust the connection relationship and custom properties of each process node. For example, in the flowchart, change the judgment node of "Is it less than 10,000" to "Is it less than 20,000", and then modify the judgment strategy of the corresponding judgment node. When users modify process nodes, they only need to adjust the flowchart, which reduces the difficulty of adjustment.

[0014] Furthermore, the S4 further comprises the following steps: S441, after the original input data of the historical process is run through the process engine, the running times X of the approval nodes of different node levels are identified; S442, according to the running number X of the approval nodes of each node level, the running number of the approval nodes of each node level is counted and ranked from the most to the least; S443. According to the ranking of the number of runs, the approval node of the node level with the most runs is marked as the main approval node, and the difference N between the number of runs of the main approval node and the number of runs of the other approval nodes, as well as the node level difference D are calculated respectively: ; D= ; is the number of times the main approval node is run, is the number of times other approval nodes are run. The node level of the main approval node, The node level of other approval nodes; S444. According to the node level of the main approval node, the difference N, and the relationship between each other approval node and the main approval node, determine whether the connection relationship between the main approval node and other approval nodes needs to be adjusted; According to the connection relationship between the nodes, identifying the node relationship between the main approval node and other approval nodes, wherein the node relationship includes a progressive relationship and a parallel relationship; When the node level difference D between the main review node and other approval nodes is less than 0, the running times of other approval nodes in each progressive relationship are identified according to other approval nodes that have progressive association with the main approval node. ,like , then no adjustment is required, otherwise adjust the approval analysis strategy after the main approval node until ; According to other approval nodes that are progressively associated with the main approval node, identify the running times of other approval nodes in each parallel relationship ,like ,Right now Greater than the number of times any other approval node is run , adjust the custom attributes of the judgment node before the approval node with a parallel relationship, so that Approaching ; When there is a node level difference D between the main approval node and other approval nodes that is greater than 0, the node relationship of other approval nodes with a level difference D greater than 0 is identified respectively. If it is a progressive relationship, according to the difference in the number of runs N, the larger N is, the larger the adjustment range of the approval analysis node of the flow direction of the other approval node is. If it is a parallel relationship, the larger N is, the larger the adjustment range of the judgment node connected to the other approval node and the main approval node is.

[0015] Since this application allows users to design their own process engines, there may be unreasonable settings in custom properties for more complex process engines when users design them. Therefore, this application simulates the operation of the process engine and determines whether there are unreasonable places based on the number of times each process node is run. It mainly analyzes the number of times the approval node is run. First, the approval node with the most runs is identified. If the level difference D between the approval node with the most runs and other approval nodes is less than 0, it means that the approval node with the most runs is the approval node with the lowest node level. According to the logic of daily work, if it is a progressive approval, the approval is reported layer by layer, and there may be failures in the middle. Therefore, if That is, the number of times each approval node runs is greater than or equal to the next approval node. The number indicates that the approval is normal, otherwise adjustments will be made.

[0016] If it is a parallel relationship, there is a judgment node in front. According to the judgment strategy, those below a certain standard will be sent to the approval node with a low node level, and those above a certain standard will be released to the approval node with a high node level. At this time, the custom properties of the judgment node in front are adjusted so that when the judgment node is allocating approvals, the lower the node level, the more approvals it will be allocated.

[0017] When D is greater than 0, that is, the main approval node is not the node with the lowest node level, then identify other approval nodes with lower node levels than the main approval node, and make adjustments based on the difference in the number of runs N. When it is a submission relationship, the larger N is, the greater the adjustment range of the approval analysis node of the flow direction of the other approval node. If it is a parallel relationship, the larger N is, the greater the adjustment range of the judgment node connected to the other approval node and the main approval node. When the number of runs of the approval node with a high node level is large, the custom attributes of the approval analysis node and the judgment node related to the approval node with a lower node level are adjusted to avoid the situation where a certain approver has too much approval work.

[0018] Through the technical solution of the present application, after the process engine is arranged, the running times of the approval nodes of each node level are adjusted to make the running times of the approval nodes of each node level more reasonable, thereby avoiding approval errors or excessive workload of a certain approver after actual use. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following is further described in detail through specific implementation methods: The embodiment is basically as shown in the attached Figure 1 As shown: A method for adjusting a process engine comprises the following steps: S1. Obtain the process engine application scenario to be edited and identify the application requirements based on the referenced scenario; S2. Create a visualization canvas through the process engine editing tool, obtain several process nodes from the pre-configured process node library according to application requirements, and add them to the visualization canvas; S3. Obtain the process node selected by the user and the custom attributes assigned to the process node, determine the initialization attributes of the process node according to the custom attributes, determine the connection relationship of each process node according to the initialization attributes, custom attributes and application scenarios of each process node, the initialization attributes include node input, node flow direction and flow direction quantity, and deploy the process engine according to the connection relationship of the process nodes; S4. Run the process engine, input different initial data, obtain output results, generate an output report based on the output results of the process engine, and optimize the process nodes based on the output report; S5. Generate a flowchart according to the deployment of the process engine, associate the flowchart with the process engine, and publish the flowchart to the departments related to the process engine reference.

[0021] First, obtain the application scenarios caused by the process, such as leave approval, payment approval, etc. Then, retrieve the recommended process nodes according to the application scenarios, add them to the visualization canvas, and obtain the process nodes selected by the user. After the user selects the process node, assign custom attributes to each process node, such as approver, approval criteria, approval waiting time, etc. Then, determine the initialization attributes and the output attributes of the process node according to the custom attributes of the process node. The initialization attributes include node input, node flow direction, and flow quantity. The node input is the process node from which the data input of the process node comes, the node flow direction is the node to which the output result of the node flows, and the flow quantity refers to whether the process node will flow to multiple nodes depending on the results. For example, in payment approval, depending on the amount of payment, more than 50,000 yuan requires approval from the director, 10,000-50,000 yuan requires approval from the manager, and less than 10,000 yuan requires approval from the department head. According to the connection relationship of the process nodes, after the process engine is deployed, run the process engine, get the running results, and generate a report. According to the report, optimize the process engine and publish it. At the same time, according to the part of the process engine, it is sent to the relevant departments and personnel. Compared with the existing technology, it can automatically recommend the required process nodes, allowing users to enter the custom attributes of each node and automatically generate the connection relationship of the process nodes. After that, the process nodes are also operated and optimized to help non-professionals complete the layout of the process engine.

[0022] The S3 comprises the following steps: S31, in addition to the start node and the end node, the process nodes are divided into judgment nodes, approval nodes and approval analysis nodes according to energy-saving functions; S32, the custom attributes of the judgment node include a judgment strategy, and the node input, node flow direction and flow quantity of the judgment node are determined according to the judgment strategy; S33, the custom attributes of the approval node include a node level, and the node input and node flow direction of the approval node are determined according to the node level; S34, the custom attributes of the approval analysis node include an audit result analysis strategy, and the node flow direction of the approval analysis node is determined according to the audit result analysis strategy; S35. Generate a connection relationship between each process node according to the node input, node flow direction and flow quantity of each process node.

[0023] In addition to the start node and the end node, the process nodes are divided into judgment nodes, approval nodes, and approval analysis nodes according to their functions. Judgment nodes refer to the judgment of inputs, and the flow is directed back to different nodes according to the input content. For example, in the payment approval, a process node for payment amount judgment is set, and a judgment node for "whether the amount exceeds 50,000" and "whether the amount is less than 10,000" is set. At the same time, 3 approval nodes are set, including "manager approval", "director approval", and "minister approval". The user sets the node level of the 3 approval nodes, "director approval", which is higher than "manager approval" and "higher than minister approval". At this time, the node input, node flow, and flow output of each process node are determined according to the custom attributes of the approval node and the judgment node, "whether the amount exceeds 50,000" The node input is the start node, and the number of flows is two, one flows to the judgment node of "Is the amount less than 10,000", and one flows to "Director's approval". The node flow directions of "Is the amount less than 10,000" are two, one flows to "Manager's approval" and one flows to "Minister's approval". After that, you can also set the "Approval Analysis Node". The input nodes of the "Approval Analysis Node" are "Director's Approval", "Manager's Approval" and "Minister's Approval". The Approval Analysis Node is used for subsequent operations after analyzing and identifying the approval results. For example, after identifying that the approval is passed, the approval will be copied to the finance department, or after identifying that the approval is not passed, it is directly ended or returned to the start node to resubmit the approval.

[0024] The S4 comprises the following steps: S41, obtaining the original input data of the historical process in the application scenario, and obtaining the output result of the end node; S42, generating an output report, wherein the output report includes the running times of each process node and the node output; S43. According to the number of operation times of the process nodes and the node outputs, invalid nodes and redundant nodes are identified, invalid nodes are removed, and redundant nodes are integrated.

[0025] The S43 comprises the following steps: S431. Generate a judgment number threshold according to the total number of runs of the process engine, and mark the process nodes whose run times are lower than the judgment number threshold as invalid nodes; S432. According to the output results of the process engine, identify the process nodes with different outputs in the same output results. When there are multiple process nodes output by nodes with the same node function, determine them as suspected redundant nodes. Merge the suspected redundant nodes and run them again to identify whether the output results are consistent. If they are consistent, retain the integrated process nodes. If they are inconsistent, maintain the original process nodes.

[0026] The S4 further comprises the following steps: S441, after the original input data of the historical process is run through the process engine, the running times X of the approval nodes of different node levels are identified; S442, according to the running number X of the approval nodes of each node level, the running number of the approval nodes of each node level is counted and ranked from the most to the least; S443. According to the ranking of the number of runs, the approval node of the node level with the most runs is marked as the main approval node, and the difference N between the number of runs of the main approval node and the number of runs of the other approval nodes, as well as the node level difference D are calculated respectively: ; D= ; is the number of times the main approval node is run, is the number of times other approval nodes are run. The node level of the main approval node, The node level of other approval nodes; S444. According to the node level of the main approval node, the difference N, and the relationship between each other approval node and the main approval node, determine whether the connection relationship between the main approval node and other approval nodes needs to be adjusted; According to the connection relationship between the nodes, identifying the node relationship between the main approval node and other approval nodes, wherein the node relationship includes a progressive relationship and a parallel relationship; When the node level difference D between the main review node and other approval nodes is less than 0, the running times of other approval nodes in each progressive relationship are identified according to other approval nodes that have progressive association with the main approval node. ,like , then no adjustment is required, otherwise adjust the approval analysis strategy after the main approval node until ; According to other approval nodes that are progressively associated with the main approval node, identify the running times of other approval nodes in each parallel relationship ,like ,Right now Greater than the number of times any other approval node is run , adjust the custom attributes of the judgment node before the approval node with a parallel relationship, so that Approaching ; When there is a node level difference D between the main approval node and other approval nodes that is greater than 0, the node relationship of other approval nodes with a level difference D greater than 0 is identified respectively. If it is a progressive relationship, according to the difference in the number of runs N, the larger N is, the larger the adjustment range of the approval analysis node of the flow direction of the other approval node is. If it is a parallel relationship, the larger N is, the larger the adjustment range of the judgment node connected to the other approval node and the main approval node is.

[0027] By running the process engine, invalid nodes and redundant nodes are identified according to the number of runs of each process node in the process engine and the node output. The input data during the simulation run is the historical original data of the application scenario. If the relevant process application was originally used in the application scenario, it can be directly retrieved through the API interface. If the original sampled paper process is obtained by scanning and identifying the paper document. Only by reflecting the actual process needs of the enterprise through the original data can invalid nodes and redundant nodes be identified more accurately. For example, when the number of runs of the process engine is 100 times and the number of runs of a process node is 1, the process node is an invalid node and can be removed. Or if the running results of a process node are different and have no effect on the output results, it may be a redundant node and can be integrated. For example, the above-mentioned "Is the amount more than 50,000" and "Is the amount less than 10,000" two judgment nodes can be integrated into a "Is the amount higher than 50,000 and less than 10,000" judgment node, which directly flows to the three "approval nodes".

[0028] The following steps are also included: S61, receiving process adjustment information about a process engine from a user, wherein the process adjustment information is an adjusted flow chart of the process engine; S62, calling the flowchart associated with the process, identifying the difference between the flowchart and the adjusted flowchart, and redeploying the connection relationship and custom attributes of the process nodes according to the difference; S63. Execute S4 on the redeployed process node and publish it again.

[0029] If the process needs to be adjusted later, the user only needs to modify the adjusted flowchart. After identifying the changed process nodes according to the flowchart, readjust the connection relationship and custom properties of each process node. For example, in the flowchart, change the judgment node of "Is it less than 10,000" to "Is it less than 20,000", and then modify the judgment strategy of the corresponding judgment node. When users modify process nodes, they only need to adjust the flowchart, which reduces the difficulty of adjustment.

[0030] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for adjusting a process engine, characterized in that: The following steps are involved: S1. Obtain the process engine application scenario to be edited and identify the application requirements based on the referenced scenario; S2. Create a visualization canvas through the process engine editing tool, obtain several process nodes from the pre-configured process node library according to application requirements, and add them to the visualization canvas; S3. Obtain the process node selected by the user and the custom attributes assigned to the process node, determine the initialization attributes of the process node according to the custom attributes, determine the connection relationship of each process node according to the initialization attributes, custom attributes and application scenarios of each process node, the initialization attributes include node input, node flow direction and flow direction quantity, and deploy the process engine according to the connection relationship of the process nodes; The S3 comprises the following steps: S31, in addition to the start node and the end node, the process nodes are divided into judgment nodes, approval nodes and approval analysis nodes according to energy-saving functions; S32, the custom attributes of the judgment node include a judgment strategy, and the node input, node flow direction and flow quantity of the judgment node are determined according to the judgment strategy; S33, the custom attributes of the approval node include a node level, and the node input and node flow direction of the approval node are determined according to the node level; S34, the custom attributes of the approval analysis node include an audit result analysis strategy, and the node flow direction of the approval analysis node is determined according to the audit result analysis strategy; S35, generating a connection relationship between each process node according to the node input, node flow direction and flow quantity of each process node; S4. Run the process engine, input different initial data, obtain output results, generate an output report based on the output results of the process engine, and optimize the process nodes based on the output report; The S4 comprises the following steps: S41, obtaining the original input data of the historical process in the application scenario, and obtaining the output result of the end node; S42, generating an output report, wherein the output report includes the running times of each process node and the node output; S43, identifying invalid nodes and redundant nodes according to the number of operation times of the process nodes and the node outputs, removing the invalid nodes, and integrating the redundant nodes; The S4 further comprises the following steps: S441, after the original input data of the historical process is run through the process engine, the running times X of the approval nodes of different node levels are identified; S442, according to the running number X of the approval nodes of each node level, the running number of the approval nodes of each node level is counted and ranked from the most to the least; S443. According to the ranking of the number of runs, the approval node of the node level with the most runs is marked as the main approval node, and the difference N between the number of runs of the main approval node and the number of runs of the other approval nodes, as well as the node level difference D are calculated respectively: ; D= ; is the number of times the main approval node is run, is the number of times other approval nodes are run. The node level of the main approval node, The node level of other approval nodes; S444. According to the node level of the main approval node, the difference N, and the relationship between each other approval node and the main approval node, determine whether the connection relationship between the main approval node and other approval nodes needs to be adjusted; According to the connection relationship between the nodes, identifying the node relationship between the main approval node and other approval nodes, wherein the node relationship includes a progressive relationship and a parallel relationship; When the node level difference D between the main review node and other approval nodes is less than 0, the running times of other approval nodes in each progressive relationship are identified according to other approval nodes that have progressive association with the main approval node. ,like , then no adjustment is required, otherwise adjust the approval analysis strategy after the main approval node until ; According to other approval nodes that are progressively associated with the main approval node, identify the running times of other approval nodes in each parallel relationship , adjust the custom attributes of the judgment node before the approval node with a parallel relationship, so that Approaching ; When there is a node level difference D between the main approval node and other approval nodes that is greater than 0, the node relationship of other approval nodes with a level difference D greater than 0 is identified respectively. If it is a progressive relationship, according to the difference in the number of runs N, the larger N is, the larger the adjustment range of the approval analysis node of the flow direction of the other approval node is. If it is a parallel relationship, the larger N is, the larger the adjustment range of the judgment node connected to the other approval node and the main approval node is. S5. Generate a flowchart based on the deployment of the process nodes, associate the flowchart with the process engine, and publish the flowchart to the department related to the process engine application.

2. The process engine adjustment method according to claim 1, characterized in that: The S43 comprises the following steps: S431. Generate a judgment number threshold according to the total number of runs of the process engine, and mark the process nodes whose run times are lower than the judgment number threshold as invalid nodes; S432. According to the output results of the process engine, identify the process nodes with different outputs in the same output results. When there are multiple process nodes output by nodes with the same node function, determine them as suspected redundant nodes. Merge the suspected redundant nodes and run them again to identify whether the output results are consistent. If they are consistent, retain the integrated process nodes. If they are inconsistent, maintain the original process nodes.

3. The process engine adjustment method according to claim 2, characterized in that: The following steps are also included: S61, receiving process adjustment information about a process engine from a user, wherein the process adjustment information is an adjusted flow chart of the process engine; S62, calling the flowchart associated with the process, identifying the difference between the flowchart and the adjusted flowchart, and redeploying the connection relationship and custom attributes of the process nodes according to the difference; S63. Execute S4 on the redeployed process node and publish it again.