Process change risk prediction method and device, server and storage medium

Abstract

The embodiment of the invention provides a process change risk prediction method and device, a server and a storage medium. The method comprises the following steps: receiving to-be-changed process information and process change information sent by a service end; constructing a flow directed graph according to the to-be-changed flow information; inputting the process change information and the process directed graph into a process change risk prediction model, so that the process change risk prediction model outputs risk probability distribution values of a plurality of preset indexes according to the process change information and the process directed graph; according to the risk probability distribution values of the plurality of preset indexes and the process change information, obtaining a risk value after the to-be-changed process is changed; the risk value is used for quantifying the risk influence after the process change. According to the method, the risk prediction accuracy is improved.

Description

Technical Field

[0001] This application relates to the field of process management technology, and in particular to a method, apparatus, server and storage medium for predicting the risks of process changes. Background Technology

[0002] In order to respond to market changes and improve operational efficiency, enterprises need to frequently adjust and change the business processes in the system to meet the needs of rapid process iteration, but the risks of process changes are uncontrollable.

[0003] Currently, existing technologies typically rely on manual checks to assess the risks associated with process changes. However, manual checks depend on experience and cannot accurately quantify the risks following process changes, thus reducing the accuracy of risk prediction. Summary of the Invention

[0004] This application provides a method, apparatus, server, and storage medium for predicting the risks of process changes, in order to improve the accuracy of risk prediction.

[0005] In a first aspect, embodiments of this application provide a method for predicting the risk of process changes, comprising: receiving process information to be changed and process change information sent by a business terminal; constructing a directed process graph based on the process information to be changed; inputting the process change information and the directed process graph into a process change risk prediction model, such that the process change risk prediction model outputs risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph; obtaining the risk value after the process to be changed is modified based on the risk probability distribution values ​​of the multiple preset indicators and the process change information; the risk value is used to quantify the risk impact after the process change.

[0006] In one possible implementation, a directed graph of the process is constructed based on the process information to be changed, including: extracting multiple process nodes and multiple process edges from the process information to be changed; wherein each process edge represents the flow relationship between process nodes; extracting the node attributes of each process node and the edge attributes of each process edge from the process information to be changed; and constructing a directed graph of the process based on the node attributes of each process node and the edge attributes of each process edge.

[0007] In one possible implementation, the process change risk prediction model outputs risk probability distribution values ​​for multiple preset indicators based on process change information and a directed process graph. This includes: obtaining multiple causal variables and the causal relationships between them from the node attributes of multiple process nodes and the edge attributes of multiple process edges; extracting the variable values ​​of each causal variable from the process change information; and performing conditional probability calculations for each preset indicator based on the causal relationships between them and the variable values ​​of each causal variable to obtain the risk probability distribution values ​​for multiple preset indicators.

[0008] In one possible implementation, the risk value of the process to be changed is obtained based on the risk probability distribution values ​​of multiple preset indicators and process change information. This includes: obtaining the risk probability distribution values ​​of compliance indicators and failure indicators from the risk probability distribution values ​​of multiple preset indicators; obtaining the risk value of resource indicators from the process change information; and obtaining the risk value of the process to be changed after the process is changed based on the risk probability distribution values ​​of compliance indicators, failure indicators, and resource indicators.

[0009] In one possible implementation, before inputting the process change information and the directed process graph into the process change risk prediction model, the method further includes: defining multiple causal variables and the causal relationships between them; constructing a process causal topology graph based on the multiple causal variables and the causal relationships between them; obtaining risk data from historical processes; and performing conditional probability training based on the process causal topology graph using the risk data from historical processes to obtain the process change risk prediction model.

[0010] In one possible implementation, after obtaining the risk value after modifying the process to be changed, the method further includes: extracting multiple process nodes and node attributes of each process node from the process information to be changed; wherein the node attributes include resource attributes; obtaining the resource quantity corresponding to the resource attributes of each process node; using a resource balancing algorithm, with the risk value minimization as the condition, calculating the resource adjustment amount of each process node; and optimizing the process change information based on the resource adjustment amount of each process node.

[0011] In one possible implementation, the method further includes: obtaining process information of a process currently running in the system; performing a compliance check on the process information based on a pre-built compliance rule base to determine whether there is any non-compliant process information in the process information; if it is determined that there is non-compliant process information in the process information, sending the process information and the non-compliant process information to the business end, so that the business end identifies the process information as process information to be changed, and generates process change information based on the non-compliant process information.

[0012] In one possible implementation, before obtaining the process information of the processes running in the system, the method further includes: obtaining multiple specification documents; extracting the process rules of each process in the system from the multiple specification documents; and saving the process rules to a compliance rule base to obtain a pre-built compliance rule base.

[0013] Secondly, embodiments of this application provide a risk prediction device for process changes, comprising: a receiving module for receiving process information to be changed and process change information sent by a business terminal; a construction module for constructing a directed process graph based on the process information to be changed; a prediction module for inputting the process change information and the directed process graph into a process change risk prediction model, such that the process change risk prediction model outputs risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph; and an acquisition module for acquiring the risk value of the process to be changed after modification based on the risk probability distribution values ​​of multiple preset indicators and the process change information; the risk value is used to quantify the risk impact of the process change.

[0014] Thirdly, embodiments of this application provide a server, including: a memory and a processor;

[0015] The memory stores the instructions that the computer executes;

[0016] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0018] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0019] The process change risk prediction method, apparatus, server, and storage medium provided in this application construct a directed graph of the process information to be changed. Based on the process change information and the directed graph, the process change risk prediction model outputs risk probability distribution values ​​for multiple preset indicators. This establishes a data-driven risk prediction method, avoiding reliance on manual experience to quantify the risks of process changes. By quantifying the risk value after the process change from multiple dimensions based on the risk probability distribution values ​​of multiple preset indicators and the process change information, the accuracy of risk prediction is improved. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0021] Figure 1 A schematic diagram illustrating a scenario for the risk prediction method for process changes provided in this application embodiment;

[0022] Figure 2 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 1 ;

[0023] Figure 3 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 2 ;

[0024] Figure 4 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 3 ;

[0025] Figure 5 A schematic diagram of the structure of the risk prediction device for process changes provided in the embodiments of this application;

[0026] Figure 6 This is a schematic diagram of the server structure provided in an embodiment of this application.

[0027] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] Figure 1 A schematic diagram illustrating a scenario for the risk prediction method for process changes provided in this application embodiment, such as... Figure 1 As shown, it includes: business terminal 101, server 102 and display terminal 103.

[0030] In this embodiment, server 102 receives the process information to be changed and the process change information sent by business terminal 101; server 102 constructs a directed process graph based on the process information to be changed; inputs the process change information and the directed process graph into the process change risk prediction model, so that the process change risk prediction model outputs the risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph; obtains the risk value after the process to be changed is changed based on the risk probability distribution values ​​of multiple preset indicators and the process change information; server 102 sends the risk value to display terminal 103, and display terminal 103 displays the risk value.

[0031] To respond to market changes and improve operational efficiency, businesses often need to frequently adjust and modify their business processes to meet the demands of rapid process iteration. However, the risks associated with these changes are often uncontrollable. Currently, existing technologies typically rely on manual checks to assess the risks of process changes. However, manual checks depend on experience and cannot accurately quantify the risks following changes, thus reducing the accuracy of risk prediction.

[0032] To address the technical problems in existing technologies, this application proposes the following technical concept: Considering that manual inspection relies on experience and cannot accurately quantify the risks after process changes, the inventors devised a data-driven risk prediction method to quantify the risks of process changes from multiple dimensions. Specifically, a directed graph of the process information to be changed is constructed. Based on the process change information and the directed graph, the process change risk prediction model outputs risk probability distribution values ​​for multiple preset indicators. This data-driven risk prediction method avoids relying on manual experience to quantify the risks of process changes. Based on the risk probability distribution values ​​of multiple preset indicators and the process change information, the risk value after the process change is quantified from multiple dimensions, improving the accuracy of risk prediction.

[0033] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0034] Figure 2 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 1 ,like Figure 2 As shown, the method includes:

[0035] S201: Receive the process information to be changed and the process change information sent by the business terminal.

[0036] Optionally, the process can be changed proactively by the business side based on business needs, or it can be changed due to a conflict between the current process and the process rules in the compliance rule base. The compliance rule base stores the process rules for each process in the current system, and this will be described in subsequent embodiments.

[0037] In this embodiment, the process information to be modified includes multiple process nodes and multiple process edges. Each process node includes node attributes, and each process edge includes edge attributes. The node attributes include process node identifier, process node type, resource attributes, and constraint attributes, etc.; the edge attributes include the starting node of the process edge, the ending node of the process edge, and the flow conditions of the process edge, etc.

[0038] The process node types include task types and gateway types; resource attributes include the amount of resources required by the current process node and the systems it depends on; constraint attributes include the process rules and maximum processing time of the current process node.

[0039] The process change information includes the change operations required for the process to be changed. Optionally, the change operations can be: adding, deleting, and modifying process nodes; adding, deleting, and modifying process edges; adding, deleting, and modifying node attributes; adding, deleting, and modifying edge attributes; adding, deleting, and modifying the content entering process nodes, such as data and forms; adding, deleting, and modifying the content completing process nodes, such as results and reports; and adding, deleting, and modifying the responsible roles for handling process nodes, such as employees.

[0040] S202: Construct a directed process graph based on the process information to be changed.

[0041] Specifically, multiple process nodes and multiple process edges are extracted from the process information to be changed; each process edge represents the flow relationship between process nodes; the node attributes of each process node and the edge attributes of each process edge are extracted from the process information to be changed; and a directed process graph is constructed based on the node attributes of each process node and the edge attributes of each process edge.

[0042] In this embodiment, a directed process graph is constructed by defining the interfaces of ProcessNode and ProcessEdge, based on the directed relationships between process nodes and process edges.

[0043] In this embodiment, a set of process node elements and a set of process edge elements are extracted from the process information to be modified. The set of process node elements includes multiple process nodes, as well as the process node identifier, process node type, resource attributes, and constraint attributes of each process node; the set of process edge elements includes multiple process edges, as well as the edge attributes of each process edge, such as the start node, end node, and flow conditions of the process edge.

[0044] In this embodiment, a graph algorithm tool is invoked to construct a directed graph of the process based on the set of process node elements and the set of process edge elements. The directed graph of the process is then validated, including validation for no circular dependencies and validation for no isolated nodes, to ensure that a valid directed graph of the process is obtained.

[0045] S203: Input the process change information and the directed process graph into the process change risk prediction model, so that the process change risk prediction model outputs the risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph.

[0046] Specifically, step S203 includes Sa~Sc:

[0047] Sa: Obtain multiple causal variables and the causal relationships between them from the node attributes of multiple process nodes and the edge attributes of multiple process edges.

[0048] In this embodiment, the causal variables and the causal relationships between them are defined in the process change risk prediction model. The causal variables are the core variables that affect the effectiveness of the process, such as the resource allocation of process nodes, the processing time of process nodes, the data throughput across process nodes, the global compliance risk coefficient, and the failure probability of dependent systems.

[0049] For example, the causal relationship between resource allocation of process nodes, processing time of process nodes, data throughput across process nodes, global compliance risk coefficient and failure probability of dependent systems is as follows: resource allocation of process nodes affects processing time of process nodes, processing time of process nodes affects data throughput across process nodes, data throughput across process nodes affects global compliance risk coefficient, and failure probability of dependent systems affects global compliance risk coefficient.

[0050] Sb: Extract the variable values ​​of each causal variable from the process change information.

[0051] In this embodiment, process change information needs to be converted into variable values ​​that can be recognized by the process change risk prediction model.

[0052] For example, the process change information is as follows:

[0053] Change node: Process node B,

[0054] Resource adjustments: {CPU: +30%, Memory: -10%}

[0055] Process rule change: Added ISO20022 data verification.

[0056] Expected efficiency change: Processing time: +25%.

[0057] In this embodiment, the upstream process node of process node B is process node A, and process node A adjusts the resources for process node B. The causal variables extracted from the process change information include the resource allocation of the process node and the processing time of the process node. The variable value for the resource allocation of the process node is a 30% increase in CPU and a 10% decrease in memory, and the variable value for the processing time of the process node is a 25% increase in processing time.

[0058] Sc: Based on the causal relationships between various causal variables and the variable values ​​of each causal variable, conditional probability calculations are performed for each preset indicator to obtain the risk probability distribution values ​​of multiple preset indicators.

[0059] In this embodiment, the conditional probabilities are obtained by training a Bayesian network structure learning algorithm based on defined causal variables and the causal relationships between them, combined with historical process risk data. The historical process risk data records actual process cases of risk associations. Based on the variable values ​​of each causal variable extracted from the process change information, and according to the trained conditional probabilities, the risk probability distribution values ​​of multiple preset indicators are output.

[0060] In this embodiment, the preset indicators include compliance indicators and failure indicators. The compliance indicator refers to the global compliance risk coefficient, and the failure indicator refers to the failure probability of the dependent system.

[0061] For example, the risk probability distribution of compliance indicators is: 70% probability of risk decreasing by 5%, 25% probability of risk remaining unchanged, and 5% probability of risk increasing by 5%; the risk probability distribution of failure indicators is: 95% probability of risk decreasing to 8% and 5% probability of risk remaining at 10%.

[0062] S204: Based on the risk probability distribution values ​​of multiple preset indicators and process change information, obtain the risk value after the process to be changed is modified; the risk value is used to quantify the risk impact after the process change.

[0063] Specifically, from the risk probability distribution values ​​of multiple preset indicators, the risk probability distribution values ​​of compliance indicators and failure indicators are obtained; from the process change information, the risk values ​​of resource indicators are obtained; and based on the risk probability distribution values ​​of compliance indicators, failure indicators, and resource indicators, the risk values ​​after the process to be changed are obtained.

[0064] In this embodiment, the risk value after the process to be changed is obtained by weighting the compliance indicators, failure indicators and resource indicators according to their preset weights, and summing the risk probability distribution values ​​of the compliance indicators, failure indicators and resource indicators.

[0065] Optionally, the risk value assessment criterion for resource metrics is the percentage by which CPU adjustments exceed a threshold or the percentage by which memory adjustments exceed a threshold. For example, if CPU resource adjustments increase by 30% and the threshold is 20%, then the risk value for the resource metrics is 0.5.

[0066] In this embodiment, the risk value of the compliance indicator is calculated based on the risk probability distribution value of the compliance indicator, and the risk value of the failure indicator is calculated based on the risk probability distribution value of the failure indicator.

[0067] For example, the risk probability distribution of compliance indicators is as follows: 70% probability of risk decreasing by 5%, 25% probability of risk remaining unchanged, and 5% probability of risk increasing by 5%. A decrease in risk represents a reduction in risk, with a risk value of 0; no change in risk represents stable risk, with a risk value of 0.5; and an increase in risk represents an increase in risk, with a risk value of 1. The risk value is calculated using a weighted summation of the risk value multiplied by the risk probability. The risk value of the compliance indicator is 5%×1 + 25%×0.5 + 70%×0 = 0.175.

[0068] For example, the risk probability distribution of the failure indicator is as follows: there is a 70% probability that the risk will decrease to 8%, and a 30% probability that the risk will remain at 10%. Maintaining risk indicates stable risk, and the risk value is set to 0.5; decreasing risk indicates reduced risk, and the risk value is set to 0. The risk value is calculated using a weighted summation of the risk value and the risk probability. The risk value of the failure indicator is 30% × 0.5 + 70% × 0 = 0.15.

[0069] For example, with the default weights of compliance indicators (0.3), fault indicators (0.3), and resource indicators (0.4), the risk value after changing the process to be changed is 0.175×0.3+0.15×0.3+0.5×0.4=0.2975.

[0070] In this embodiment, the preset weights of compliance indicators, fault indicators, and resource indicators can be adjusted according to the actual situation, but it is necessary to ensure that the sum of the three weights is one.

[0071] In this embodiment, it is possible to display the process information to be changed, the process change information, and the corresponding risk value.

[0072] Optionally, the risk value can be reduced by adjusting the resource quantity of each process node. Specifically, multiple process nodes and their node attributes are extracted from the process information to be changed; the node attributes include resource attributes; the resource quantity corresponding to the resource attributes of each process node is obtained; a resource balancing algorithm is used to calculate the resource adjustment amount of each process node with the risk value minimized; and the process change information is optimized based on the resource adjustment amount of each process node.

[0073] In this embodiment, the resource balancing algorithm is based on linear programming. An optimization model is constructed and solved using a GUROBI solver. The core objective is to minimize the risk value by adjusting the resource amounts at each process node under resource budget constraints. This process requires defining decision variables, setting an objective function, and setting constraints. The decision variables are the resource adjustments at each process node, the objective function is to minimize the risk value, and the constraint is that the total resource amount does not exceed the budget.

[0074] In this embodiment, if there is a resource shortage at any process node, the process change information can be optimized based on the resource adjustment amount of each process node by initiating elastic scaling, dynamic routing strategies, and degradation mechanisms. Initiating elastic scaling refers to allocating resources from the allocated resource pool; dynamic routing strategies refer to redirecting 30% of traffic to other process nodes; and degradation mechanisms refer to temporarily disabling non-core verification rules.

[0075] In summary, by constructing a directed graph of the process information to be changed, the process change risk prediction model outputs risk probability distribution values ​​for multiple preset indicators based on the process change information and the directed graph. This establishes a data-driven risk prediction method, avoiding reliance on manual experience to quantify the risks of process changes. By quantifying the risk value after process change from multiple dimensions based on the risk probability distribution values ​​of multiple preset indicators and process change information, the accuracy of risk prediction is improved. Furthermore, by adjusting the resource allocation of each process node and optimizing the process change information, the risk value is minimized.

[0076] refer to Figure 3 , Figure 3 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 2 Based on the above embodiments, this embodiment describes the process of constructing a process change risk prediction model, as detailed below:

[0077] S301: Define multiple causal variables and the causal relationships between them.

[0078] In this embodiment, multiple causal variables include: resource allocation of process nodes, processing time of process nodes, data throughput across process nodes, global compliance risk coefficient, and failure probability of dependent systems.

[0079] In this embodiment, the causal relationships among the causal variables are as follows: the resource allocation of process nodes affects the processing time of process nodes; the processing time of process nodes affects the data throughput across process nodes; the data throughput across process nodes affects the global compliance risk coefficient; and the failure probability of the dependent system affects the global compliance risk coefficient.

[0080] S302: Construct a process causal topology diagram based on multiple causal variables and the causal relationships between them.

[0081] In this embodiment, a Bayesian network modeling tool is used, with each causal variable treated as a node in the Bayesian network. Node attributes carry basic information about the variables, such as resource allocation. The causal relationships between the causal variables are used as directed edges in the Bayesian network, and causal relationships, such as resource-efficiency relationships, are labeled to obtain a process causal topology graph.

[0082] S303: Obtain risk data from historical processes.

[0083] In this embodiment, the risk data of historical processes records actual process cases of risk association.

[0084] S304: Based on the risk data of historical processes, conditional probability training is performed on the process causal topology graph to obtain a process change risk prediction model.

[0085] In this embodiment, the process causal topology graph and historical process risk data are loaded into the training environment. Based on the Bayesian network structure learning algorithm, the causal relationship of the process causal topology graph is fine-tuned. Using a Bayesian estimator, the conditional probability of each node under different combinations of parent node values ​​is calculated based on the historical process risk data to train the process change risk prediction model.

[0086] In this embodiment, during the conditional probability training process, risk dimensions such as efficiency decay and rule deviation can be considered. Efficiency decay is the percentage decrease in processing volume per unit time, obtained from the processing time of process nodes before and after the process change; rule deviation is the number of times process rules are violated, obtained through compliance checks using a pre-built compliance rule base.

[0087] In summary, by defining multiple causal variables and the causal relationships between them, a process causal topology graph is constructed. Combining historical process risk data with conditional probability training based on this graph, a process change risk prediction model is obtained. This establishes a data-driven and causal relationship-driven risk prediction method, avoiding reliance on manual experience to quantify process change risks and improving the accuracy of risk prediction.

[0088] refer to Figure 4 , Figure 4 A flowchart illustrating the risk prediction method for process changes provided in this application embodiment. Figure 3 Based on the above embodiments, this embodiment describes the process of driving process changes through a compliance rule base, as detailed below:

[0089] S401: Obtain process information of the process currently running in the system.

[0090] In this embodiment, the process information includes: start process node, intermediate process node, end process node, process edges between process nodes, input and output descriptions of each process node, and the role of the person handling each process node.

[0091] Optionally, the process information can be in XML format.

[0092] For example, a reimbursement process includes: employees submitting reimbursement forms, department managers approving them, checking whether the reimbursement amount exceeds the limit, and filing expenses less than 5,000 yuan directly, while expenses greater than 5,000 yuan are approved and filed by the finance manager.

[0093] S402: Based on the pre-built compliance rule base, perform compliance checks on the process information to determine whether there is any non-compliant process information.

[0094] In this embodiment, the process rules for the current process are extracted from the pre-built compliance rule base; the process information, process rules, and prompt words are input into the trained large model, and the large language model is checked for non-compliant process information.

[0095] In this embodiment, the process of building the compliance rule base is as follows: obtain multiple specification documents; extract the process rules of each process in the system from the multiple specification documents; save the process rules to the compliance rule base to obtain the pre-built compliance rule base.

[0096] In this embodiment, the specification document refers to the specification document that constrains the process, including but not limited to process documents and requirements documents. The content format of the specification document includes, but is not limited to, tables and text. The content of plain text areas is extracted using optical character recognition (OCR) technology. For table areas, cell content is extracted using OCR technology and table line detection, and converted into structured data. Semantic analysis is then used to extract process rules from the extracted text and structured table content. These extracted process rules are saved to obtain a pre-built compliance rule library.

[0097] Alternatively, a compliance rule base can be implemented using a vector database.

[0098] Optionally, when a new specification document appears, or a new specification appears in any specification document, the large language model extracts the process rules that need to be followed in the new specification document or the new specification, and then saves them in the compliance rule base. At the same time, the new process rules need to be used to perform compliance judgment on the process.

[0099] For example, the new specification document and the following content are input into the large language model: The document sent is the new specification document. Please read the document and extract the process rules that need to be followed. These process rules are related to the processes of the IT system, and the processes of the IT system cannot contradict these process rules.

[0100] For example, a process rule extracted from a data security specification document states that when approving data storage, the location of the data storage must be clearly defined.

[0101] S403: If it is determined that there is non-compliant process information in the process information, the process information and the non-compliant process information are sent to the business end so that the business end can identify the process information as the process information to be changed, and generate process change information based on the non-compliant process information.

[0102] In this embodiment, it is supported to display non-compliant process information.

[0103] In this embodiment, if non-compliant process information is determined to exist in the process information, the non-compliant process information is output and sent to the business end so that the business end can issue an early warning. At the same time, the process information is identified as process information to be changed, and process change information is generated based on the non-compliant process information, triggering a risk prediction operation for process change.

[0104] In summary, by using a pre-built compliance rule base to conduct compliance checks on processes and obtain information on non-compliant processes, the pre-built compliance rule base extracts process rules from multiple specification documents to ensure the accuracy of compliance checks. The business side can generate process change information based on non-compliant process information, thereby triggering risk prediction operations for process changes, forming a closed loop to ensure the timely resolution of non-compliant processes.

[0105] Figure 5 This is a schematic diagram of the structure of the risk prediction device for process changes provided in the embodiments of this application, as shown below. Figure 4 As shown, the process change risk prediction device provided in this embodiment includes: a receiving module 501, a construction module 502, a prediction module 503, and an acquisition module 504.

[0106] The receiving module 501 is used to receive the process information to be changed and the process change information sent by the business terminal;

[0107] Module 502 is used to construct a directed graph of the process based on the process information to be changed;

[0108] The prediction module 503 is used to input process change information and process directed graph into the process change risk prediction model, so that the process change risk prediction model outputs the risk probability distribution values ​​of multiple preset indicators based on the process change information and process directed graph.

[0109] The acquisition module 504 is used to obtain the risk value after the process to be changed based on the risk probability distribution values ​​of multiple preset indicators and process change information; the risk value is used to quantify the risk impact after the process change.

[0110] In one possible implementation, the construction module 502 is specifically used to: extract multiple process nodes and multiple process edges from the process information to be changed; wherein each process edge represents the flow relationship between process nodes; extract the node attributes of each process node and the edge attributes of each process edge from the process information to be changed; and construct a directed process graph based on the node attributes of each process node and the edge attributes of each process edge.

[0111] In one possible implementation, the prediction module 503 is specifically used to: obtain multiple causal variables and the causal relationships between each causal variable from the node attributes of multiple process nodes and the edge attributes of multiple process edges; extract the variable values ​​of each causal variable from the process change information; and perform conditional probability calculations for each preset indicator based on the causal relationships between each causal variable and the variable values ​​of each causal variable to obtain the risk probability distribution values ​​of multiple preset indicators.

[0112] In one possible implementation, the acquisition module 504 is specifically used to: acquire the risk probability distribution values ​​of compliance indicators and failure indicators from the risk probability distribution values ​​of multiple preset indicators; acquire the risk values ​​of resource indicators from process change information; and acquire the risk values ​​of the process to be changed after modification based on the risk probability distribution values ​​of compliance indicators, failure indicators, and resource indicators.

[0113] In one possible implementation, the risk prediction device for process changes further includes a first construction module, which is specifically used to: define multiple causal variables and the causal relationships between them; construct a process causal topology graph based on the multiple causal variables and the causal relationships between them; acquire risk data of historical processes; and perform conditional probability training based on the process causal topology graph using the risk data of historical processes to obtain a process change risk prediction model.

[0114] In one possible implementation, the risk prediction device for process changes further includes an optimization module, which is specifically used to: extract multiple process nodes and node attributes of each process node from the process information to be changed; wherein the node attributes include resource attributes; obtain the resource quantity corresponding to the resource attributes of each process node; use a resource balancing algorithm to calculate the resource adjustment amount of each process node under the condition of minimizing the risk value; and optimize the process change information based on the resource adjustment amount of each process node.

[0115] In one possible implementation, the risk prediction device for process changes further includes a judgment module, which is specifically used to: obtain process information of the process currently running in the system; perform compliance checks on the process information according to a pre-built compliance rule base to determine whether there is any non-compliant process information in the process information; if it is determined that there is non-compliant process information in the process information, send the process information and the non-compliant process information to the business end, so that the business end determines the process information as the process information to be changed, and generates process change information based on the non-compliant process information.

[0116] In one possible implementation, the risk prediction device for process changes further includes a second construction module, which is specifically used to: acquire multiple specification documents; extract process rules for each process in the system from the multiple specification documents; and save the process rules to a compliance rule base to obtain a pre-built compliance rule base.

[0117] The risk prediction device for process changes provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0118] Figure 6 This is a schematic diagram of the server structure provided in an embodiment of this application. As shown in the figure, the server provided in this embodiment includes at least one processor 601 and a memory 602. Optionally, the server also includes a communication component 603. The processor 601, memory 602, and communication component 603 are connected via a bus 604.

[0119] In a specific implementation, at least one processor 601 executes computer execution instructions stored in memory 602, causing at least one processor 601 to perform the above-described method.

[0120] The specific implementation process of processor 601 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0121] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0122] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0123] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0124] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0125] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0126] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0127] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0128] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0131] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0132] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0133] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for predicting the risks of process changes, characterized in that, Applied to servers, including: Receive information on processes to be modified and process changes sent by the business side; Based on the process information to be changed, construct a directed process graph; The process change information and the directed process graph are input into the process change risk prediction model, so that the process change risk prediction model outputs the risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph. Based on the risk probability distribution values ​​of the multiple preset indicators and the process change information, the risk value after the process to be changed is obtained; the risk value is used to quantify the risk impact after the process change.

2. The method according to claim 1, characterized in that, The step of constructing a directed graph of the process based on the process information to be changed includes: From the process information to be changed, extract multiple process nodes and multiple process edges; where each process edge represents the flow relationship between process nodes. Extract the node attributes of each process node and the edge attributes of each process edge from the process information to be changed. Based on the node attributes of each process node and the edge attributes of each process edge, a directed process graph is constructed.

3. The method according to claim 2, characterized in that, The process change risk prediction model outputs risk probability distribution values ​​for multiple preset indicators based on the process change information and the directed process graph, including: From the node attributes of the multiple process nodes and the edge attributes of the multiple process edges, obtain multiple causal variables and the causal relationships between each causal variable; Extract the values ​​of each causal variable from the process change information; Based on the causal relationships between the causal variables and the variable values ​​of the causal variables, conditional probability calculations are performed for each preset indicator to obtain the risk probability distribution values ​​of multiple preset indicators.

4. The method according to claim 1, characterized in that, The step of obtaining the risk value after modifying the process to be modified based on the risk probability distribution values ​​of the multiple preset indicators and the process change information includes: From the risk probability distribution values ​​of the multiple preset indicators, obtain the risk probability distribution values ​​of the compliance indicators and the risk probability distribution values ​​of the failure indicators; Obtain the risk values ​​of resource indicators from the process change information; Based on the risk probability distribution values ​​of the compliance indicators, the risk probability distribution values ​​of the fault indicators, and the risk values ​​of the resource indicators, obtain the risk value after the process to be changed is modified.

5. The method according to claim 1, characterized in that, Before inputting the process change information and the directed process graph into the process change risk prediction model, the method further includes: Define multiple causal variables and the causal relationships between them; Based on the multiple causal variables and the causal relationships between them, a process causal topology diagram is constructed. Obtain risk data from historical processes; Based on the risk data of the historical process, a conditional probability training is performed on the causal topology graph of the process to obtain a process change risk prediction model.

6. The method according to claim 1, characterized in that, After obtaining the risk value after modifying the process to be changed, the process also includes: From the process information to be changed, extract multiple process nodes and node attributes of each process node; wherein the node attributes include resource attributes. Obtain the resource quantity corresponding to the resource attributes of each process node; Using a resource balancing algorithm, with the condition of minimizing the risk value, the resource adjustment amount for each process node is calculated; The process change information is optimized based on the resource adjustment amount of each process node.

7. The method according to claim 1, characterized in that, The method further includes: Obtain process information of the processes currently running in the system; Based on a pre-built compliance rule base, the process information is subjected to compliance checks to determine whether there is any non-compliant process information in the process information; If it is determined that there is non-compliant process information in the process information, the process information and the non-compliant process information are sent to the business terminal, so that the business terminal determines the process information as process information to be changed, and generates process change information based on the non-compliant process information.

8. The method according to claim 7, characterized in that, Before obtaining the process information of the process currently running in the system, the method further includes: Obtain multiple specification documents; Extract the process rules for each process in the system from the multiple specification documents; The process rules are saved to the compliance rule base to obtain a pre-built compliance rule base.

9. A risk prediction device for process changes, characterized in that, Applied to servers, including: The receiving module is used to receive the process information to be changed and the process change information sent by the business side; The construction module is used to build a directed graph of the process based on the process information to be changed; The prediction module is used to input the process change information and the directed process graph into the process change risk prediction model, so that the process change risk prediction model outputs the risk probability distribution values ​​of multiple preset indicators based on the process change information and the directed process graph. The acquisition module is used to acquire the risk value of the process to be changed after the process is changed, based on the risk probability distribution values ​​of the multiple preset indicators and the process change information; the risk value is used to quantify the risk impact of the process change.

10. A server, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-8.

12. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-8.

Images