Process management method and device

Abstract

The embodiment of the invention provides a process management method and device, and a business processing method comprises the steps: obtaining a target process, and carrying out the feature extraction based on the target process to obtain a process feature; determining a target splitting strategy set adopted for the target process based on preset process splitting strategies; and based on the target splitting strategy set, splitting and recombining the target process to generate a replacement process corresponding to each target splitting strategy. According to the embodiment of the invention, the optimization of the business process can be automatically completed, the complex sub-processes are split and recombined, the maintenance cost of the complex process is reduced, the process readability is improved, and the business processing efficiency is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to cloud computing technology, and more particularly to a process management method and device. Background Art

[0002] A process is a series of orderly, interrelated activities or steps that can transform input into output to achieve a specific goal. Processes can be divided into many categories. According to different application scenarios and requirements, they can be roughly divided into the following categories: (1) Agentic Workflow: This type of process usually involves multiple participants or systems, emphasizes the allocation and collaboration of tasks, and is suitable for complex enterprise business processes, such as customer service and project management; (2) Data Flow: This type of process focuses on the transmission and processing of data between systems. It is often used to describe the path of data input, output, storage and processing. It is common in the fields of information system design and data analysis; (3) Workflow: Workflow is an automated method for defining and controlling business processes, including the order of tasks, the roles and rules of participants. Workflow management systems (WFMS) can be used to optimize and supervise daily business operations.

[0003] However, the above-mentioned processes generally have the following problems: when there are too many nodes in the process, the node type is difficult to correspond to the business or data, the logic is difficult to understand, and the number of repeated nodes increases, making it difficult to mine or reuse them. In addition, the existing process construction methods lack automated processing solutions and are relatively complex to operate. Therefore, how to automate complex processes and provide users with simpler operations when building processes is an urgent problem to be solved. Summary of the invention

[0004] The embodiments of the present application provide a process management method and device, which realize automatic identification and splitting of complex processes, can reduce the maintenance cost of complex processes, improve the readability of processes, and improve the automation and flexibility of the entire business processing.

[0005] In a first aspect, an embodiment of the present application provides a process management method, comprising: obtaining a target process, and performing feature extraction based on the target process to obtain process features; based on the process features, determining a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies; based on the target splitting strategy set, splitting and reorganizing the target process to generate a replacement process corresponding to the target process.

[0006] By implementing the embodiments of the present application, it is possible to automatically sort out complex processes and split complex processes into simpler and more integrated sub-processes, and optimize business processes by outputting optional alternatives to users to obtain user input. Specifically, after receiving the target process input by the user, the process characteristics of the target process (such as the structural characteristics of the target process, the historical behavior preferences of the user who initiated the business process, etc.) can be obtained through feature extraction, and the information of the target process can be fully mined to provide more accurate information support for determining the target splitting strategy set; in parallel, after receiving the target process input by the user, the process management method can automatically identify target sub-processes that are more complex or need to be split and optimized, without the need for manual selection by the user, that is, the function of sorting out the target process can be realized, saving labor costs; further, based on the process characteristics, a target splitting strategy set is determined for the target process. For example, the target sub-process that needs to be split in the target process can be first determined, and based on the process characteristics, a target sub-split strategy set can be determined for the target sub-process, and then based on the target sub-process, a target sub-split strategy set can be determined for the target sub-process. The target splitting strategy set is determined based on the target sub-splitting strategy set, wherein the target sub-splitting strategy set can be determined by calculating the confidence of one or more matching process splitting strategies and sorting the confidence calculation results. In this way, the target splitting strategy set can be determined according to the degree of adaptation to the target sub-process, so that the target splitting strategy set and the replacement process can be more suitable for the target process, and the optimized target process can be more reasonable and smoother; further, the target process is automatically split according to one or more target splitting strategy sets, and one or more corresponding replacement processes can be obtained, which provides users with more possibilities for business target process optimization and is easier to adapt to different user needs and operating environments, thereby improving practical value and decision-making quality, while improving the flexibility of the method and realizing automatic optimization of complex processes.

[0007] In a possible implementation of the first aspect, feature extraction is performed based on the target process to obtain process features, including: receiving user features input by a target user, and performing feature extraction based on the user features and the target process to obtain process features; the target process is constructed by the target user, and the user features are user portraits of the target user. By implementing the embodiments of the present application, the process features can include more information by receiving user features of the target user and integrating the user features into the process features. When determining the target splitting strategy set, two aspects of information (information of the target process itself and relevant information of the target user) are provided through the process features, so that various factors can be considered more comprehensively when determining the target splitting strategy set.

[0008] In a possible implementation of the first aspect, determining the target splitting strategy set to be adopted for the target process includes: determining N target sub-processes that need to be split in the target process, where N is an integer greater than or equal to 1; based on the process characteristics, determining the target sub-splitting strategy set to be adopted for each of the target sub-processes from a plurality of preset process splitting strategies; and determining the target splitting strategy set based on the target sub-splitting strategy set of the N target sub-processes. By implementing the embodiments of the present application, the sub-processes that need to be split in the target process can be automatically determined without manual selection or input by the user, and automatic combing of the target process can be achieved; modular processing can be achieved by determining the splitting strategy for the target process according to the target sub-process, and the analysis and processing of the parts of the target process that need to be split can be achieved in parallel, thereby improving computing efficiency.

[0009] In a possible implementation of the first aspect, based on the process characteristics, a target sub-splitting strategy set to be adopted for each of the target sub-processes is determined from a plurality of preset process splitting strategies, including: receiving an expected splitting strategy input by a user, analyzing the expected splitting strategy, generating a user intent, wherein the input form of the expected splitting strategy includes one or more of text, image, audio, and reference business processes; based on the user intent and the process characteristics, and the preset plurality of process splitting strategies, determining a target sub-splitting strategy set to be adopted for each of the target sub-processes. In the implementation of the present application, by receiving the user's expected splitting strategy, the user's splitting intent can be analyzed and generated, and relevant information can be provided when determining the target splitting strategy set, so that when determining the splitting strategy, in addition to automatically selecting a suitable splitting strategy according to the process characteristics, the user's splitting intent is also fully referred to, so that the generated replacement process block can better meet the user's needs; at the same time, the input form of the expected splitting strategy can include one or more of text, image, audio, and reference business processes, which improves the flexibility of user interaction and enhances the ease of use and robustness of the method, and can also improve the user's experience.

[0010] In a possible implementation of the first aspect, any one of the target sub-splitting strategy sets includes M preset process splitting strategies; the determination of the target sub-splitting strategy set adopted for each of the target sub-processes includes: for any one of the target sub-processes, scoring the corresponding M preset process splitting strategies based on the user intention and the process characteristics to obtain M confidences, where M is an integer greater than or equal to 1; sorting the M preset process splitting strategies from high to low according to the M confidences to obtain the sorting results of the M preset process splitting strategies; and generating the target sub-splitting strategy set corresponding to the target sub-process based on the sorting results. In implementing the embodiment of the present application, by calculating the confidence of the process splitting strategies included in the target sub-splitting strategy set, it is possible to quantify whether the process splitting strategies included in the target sub-splitting strategy set are suitable for the target sub-process and the degree of suitability; at the same time, sorting the process splitting strategies included in the target sub-splitting strategy set can provide a basis for the subsequent elimination of one or several process splitting strategies with lower confidence rankings.

[0011] In a possible implementation of the first aspect, a target splitting strategy set is determined based on the target sub-splitting strategy set of the N target sub-processes, including: based on the ranking result of the confidence, at least one process splitting strategy is determined from each target sub-splitting strategy set corresponding to the N target sub-processes to form one or more target splitting strategy sets. In implementing the embodiment of the present application, by selecting one or more process splitting strategies with high confidence ranking in the target sub-splitting strategy set corresponding to each target sub-process, one or more optimization schemes for the overall target process (i.e., one or more target splitting strategy sets) can be combined, which can provide users with complete and diversified process optimization schemes that can adapt to different users or environments; at the same time, some unreasonable process splitting strategies or those with high confidence ranking but low values ​​can be excluded, thereby improving the flexibility of the method and enhancing the user experience.

[0012] In a possible implementation of the first aspect, based on the target splitting strategy set, the target process is split and reorganized to generate a replacement process corresponding to the target process, including: based on the one or more target splitting strategy sets, the target process is split and reorganized to generate one or more replacement processes corresponding to the target splitting strategy set. In the implementation of the present application, the target process can be split based on the target splitting strategy set to obtain a replacement process, and can be provided to the user for viewing and selection, so that the user can be more clear about the optimization method of the replacement process; in addition, after determining the splitting strategy for the overall target process, the N sub-processes in the target process are split and reorganized based on the target splitting strategy (rather than splitting and reorganizing the target sub-process after determining the N target sub-processes and the corresponding target sub-split strategies), which can reduce the amount of calculation, improve the optimization efficiency, and thus improve the user experience.

[0013] In a possible implementation of the first aspect, the process management method provided in the embodiment of the present application also includes: based on the one or more replacement processes and the process characteristics, respectively generating titles and description information for the one or more replacement processes through a large language model; and saving the one or more replacement processes and the corresponding titles and description information. By implementing the embodiment of the present application, a title and description information can be generated for the replacement process based on the process characteristics through a large language model, which can realize automatic naming of the replacement process, and can provide users with relevant basic information about the replacement process through the title and description information (without the need for users to browse specific replacement processes to obtain information). For example, the title and description information can include the main purpose and advantages of each replacement process split for users to view or select, saving users browsing time.

[0014] In a possible implementation of the first aspect, the process method provided by the embodiment of the present application also includes: outputting the one or more saved replacement processes and corresponding titles and description information so that the user can select at least one replacement process; after receiving the user's selection input, saving the replacement process selected by the user. By implementing the embodiment of the present application, the replacement process and the corresponding title and description can be output to the user for the user to select, and after receiving the user's selection, the replacement process selected by the user can be saved to complete the optimization of the original target process, which can enhance the user's control, improve the decision-making quality, and enhance the flexibility of the system.

[0015] In a possible implementation of the first aspect, the multiple process splitting strategies include: one or more of a splitting strategy according to a common sequence, a splitting strategy according to a business, a splitting strategy according to a role, and a splitting strategy according to a structure. By implementing the embodiment of the present application, four preset, basic splitting strategies can basically achieve full coverage of business process situations that need to be split, that is, it has a wide range of applicability; at the same time, by roughly classifying and planning the situations of the target processes that need to be split, it is possible to quickly analyze and split and reorganize the target processes input by the user, which can save computing time and improve efficiency.

[0016] In a second aspect, an embodiment of the present application provides a process management device, comprising: a receiving unit, for receiving a target process input by a user; a splitting unit, for parsing the target process input by the user to obtain process characteristics; based on the process characteristics, determining a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies, and based on the target splitting strategy set, splitting and reorganizing the target process to generate a replacement process corresponding to the target process; and a storage unit, for storing the replacement process.

[0017] In a possible implementation of the second aspect, the receiving unit is further used to receive user features input by a target user and an expected splitting strategy input by the target user, the target process is constructed by the target user, and the user features are a user portrait of the target user.

[0018] In a possible implementation of the second aspect, the splitting unit is specifically used to determine the target sub-process that needs to be split; analyze the expected splitting strategy to generate user intent, and based on the user intent, the user characteristics, and the process characteristics, determine the target sub-splitting strategy set to be adopted for the target sub-process from multiple preset process splitting strategies; based on the target sub-splitting strategy set, determine the target splitting strategy set, and split and reorganize the target process to obtain corresponding replacement process blocks; generate a corresponding title and description for the replacement process.

[0019] In a possible implementation manner of the second aspect, the storage unit is specifically configured to obtain and store the replacement process and a title and description information corresponding to the replacement process.

[0020] In a third aspect, an embodiment of the present application provides a server, characterized in that it includes: a memory and one or more processors; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the process management method described in any one of the above-mentioned first aspects.

[0021] In a fourth aspect, an electronic device is provided, comprising: a memory and one or more processors; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the process management method described in any one of the above-mentioned first aspects.

[0022] In a fifth aspect, the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and when the computer program is executed by a processor, the process management method described in any one of the above-mentioned first aspects is implemented.

[0023] In a sixth aspect, an embodiment of the present application provides a computer program product, which includes instructions. When the computer program product is executed by a computer, the computer can execute the process management method described in any one of the above-mentioned first aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a system architecture diagram of the process management method provided in the embodiment of the present application;

[0025] Figure 2 is a schematic flow chart of a process management method 100 provided in an embodiment of the present application;

[0026] Figure 3 It is a schematic diagram of a framework of a process management method provided in an embodiment of the present application;

[0027] Figure 4 It is a schematic diagram of a process for determining an extraction strategy set provided by an embodiment of the present application;

[0028] Figure 5 A schematic diagram of a splitting and reassembling process provided in an embodiment of the present application;

[0029] Figure 6 It is a target flow diagram provided by an embodiment of the present application;

[0030] Fig. 7A This is a schematic diagram of a replacement sub-process obtained by structural splitting provided in an embodiment of the present application;

[0031] Figure 7B This is a schematic diagram of a replacement sub-process obtained by splitting according to roles provided in an embodiment of the present application;

[0032] Figure 7C This is a schematic diagram of a replacement sub-process obtained by splitting a common sequence provided in an embodiment of the present application;

[0033] Fig.7DIt is a schematic diagram of a replacement process provided by an embodiment of the present application;

[0034] Fig. 7E It is another replacement flow diagram provided in an embodiment of the present application;

[0035] Figure 7F It is another replacement flow diagram provided in an embodiment of the present application;

[0036] Figure 8 It is another target flow diagram provided by an embodiment of the present application;

[0037] Fig. 9 This is a schematic diagram of a replacement sub-process generated according to business splitting and reorganization provided in an embodiment of the present application;

[0038] Fig.10 It is another replacement flow diagram provided in an embodiment of the present application;

[0039] Fig.11 It is a schematic diagram of a process of generating a title and description for a replacement process provided by an embodiment of the present application;

[0040] Fig. 12A This is a schematic diagram of an optimization strategy output interface provided in an embodiment of the present application;

[0041] Fig. 12B This is a schematic diagram of a comparison diagram before and after optimization provided in an embodiment of the present application;

[0042] Fig.13 A schematic diagram of the structure of a business process management device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The embodiments of the present application provide a process management method and device.

[0044] The terms used in the following embodiments of the present application are only for describing specific embodiments, and are not intended to limit the present application. As used in the specification and the appended claims of the present application, the singular expressions "one", "a kind", "some", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0045] Figure 1 : is a system architecture diagram of the process management method provided in the embodiment of the present application. Figure 1As shown, the user can build or optimize the target process through an electronic device (including but not limited to a tablet computer, desktop computer, desktop computer, laptop computer, handheld computer, notebook computer, smart phone, etc.), the server receives the user request and user data, and processes them based on the logic module, and returns the results such as the replacement process block and the optimized target process to the user. At the same time, the server can help the user store data such as the target process.

[0046] Combine the following Figure 2 The main process of the process management method provided in the embodiment of the present application is described.

[0047] Figure 2 is a schematic flow chart of a process management method 100 provided in an embodiment of the present application, such as Figure 2 As shown, the method 100 includes steps S101 to S104.

[0048] S101: Obtain the target process.

[0049] Specifically, the user can input a complete target process, such as a set of logically complete processes including a start node and an end node and connecting the two. Exemplarily, the process can be various types of processes that include a series of process nodes and execution methods in an orderly manner, and can specifically include one or more of the following categories: (1) Business process: A business process is a series of activities completed by different people to achieve a specific value goal. The activities are not only strictly limited in sequence, but also have clear arrangements and definitions for the content, methods, and responsibilities of the activities, so that different activities can be transferred between different job roles; (2) Service process: refers to a detailed analysis and sorting of customer requirements to clarify the specific content and implementation methods of the requirements; (3) Process flow: refers to a series of processing, conversion and handling steps from raw materials to final products or services in industrial production; (4) Development process: Development process Programming refers to the work of programming, testing, debugging, etc. based on the design, and ultimately realizing the development and deployment of the system. The development process includes coding, testing, integration, etc.; (5) Operation and maintenance process: The operation and maintenance process refers to the daily operation and maintenance management work after the system is launched to ensure the stable operation and security of the system. The operation and maintenance process includes monitoring, troubleshooting, backup, etc.; (6) Quality management process: The quality management process refers to the management and control of project quality during the entire project development process. The quality management process includes quality planning, quality control, quality assessment, etc.; (7) Project management process: The project management process refers to the process of planning, organizing, coordinating and controlling the entire project. The project management process includes project establishment, project planning, project execution, project monitoring, and project closing, etc.

[0050] Among them, taking the business process as an example, the business process is hierarchical, which is reflected in the logical relationship from top to bottom, from the whole to the part, from the macro to the micro, and from the abstract to the concrete; in addition, the business process can be modeled, designed and optimized through various development tools and methods. For example, the business process modeling tool can be used to visually describe the various parts of the business process and the relationship between them, and the business process modeling tool can also be used to realize the automation and integration of the business process.

[0051] When building a business process using a business process modeling tool, the business process usually consists of the following:

[0052] (1) Process nodes: Process nodes are the basic components of a business process. Each node represents a specific task, operation or decision point. Common process nodes include start nodes, task nodes, decision nodes, parallel nodes and end nodes. The start node refers to the first node in a business process. The task node represents a specific task or operation that needs to be performed. The task can be a manual task (such as approval, filling out a form, etc.) or an automatic task (such as sending an email, triggering other systems, etc.). The decision node is a conditional branch node that can determine the direction of the process based on specific conditions or rules. The parallel node allows the process to be executed along multiple branch paths at the same time. The end node is the last node in the business process, indicating the completion of the business process.

[0053] (2) Sequence flow: A sequence flow is usually a solid line with an arrow at one end, which is used to link the elements within a process in a diagram or a single pool and show the order in which these elements are executed;

[0054] (3) Other key contents: process variables used to store and transfer data, process scripts used to execute specific business logic or tasks, etc.

[0055] For example, a user can enter the business process of "reimbursement", which may include process nodes such as "the person making the reimbursement collects expense receipts and reimbursement vouchers", "fills in the expense reimbursement form", "submits to the department head or superior for approval", "review by the financial director", "approval by the company's general manager", "review by the cashier", "performs payment procedures", and "saves and manages the reimbursement records".

[0056] In addition, users can also enter sub-processes that need to be split. Among them, the sub-process in the business process refers to the partial process nodes in the business process that can be further subdivided, relatively independent but closely related to the overall process. The sub-process usually completes a specific link or task in the business process and can be reused in different business processes. For example, in the above-mentioned "reimbursement" process, "submit to the department head or superior for approval", "review by the financial director", and "approval by the company's general manager" can be used as an approval-related sub-process.

[0057] S102: Extracting process features based on the target process.

[0058] Feature extraction is performed on the target process input by the user to obtain process features, where the process features may include structural features and business information of the target process.

[0059] For business processes, the structural characteristics mainly include: a. Goal-orientedness. The design and implementation of business processes are to achieve specific business goals. The goal-orientedness of business processes is reflected in the fact that each step and each link is closely centered around specific business goals, ensuring that the entire process can be efficiently executed towards the goal; b. Integrity. Integrity refers to the characteristics of a business process or workflow as a complete system, and the inherent connections or interdependencies between the various parts of the process; c. Orderliness, which refers to the characteristics of each step and each link in a business process or workflow being carried out in a certain order and logic; d. Correlation, which refers to the characteristics of close connections and mutual influences between the various parts of a business process; e. Dynamicity, which refers to the characteristics of a business process or workflow that can be flexibly adjusted and optimized in the face of changes in the external environment and adjustments in internal needs.

[0060] Specifically, when extracting features from a target (business) process (hereinafter referred to as the "target business process"), the key performance indicators (KPIs) in the target business process can be obtained to evaluate the target nature of the target business process; the integrity of the target business process can be quantified by calculating the process completeness index and process correlation to evaluate whether each task in the business process is complete and whether the logical relationship between activities is clear; the orderliness of the target business process can be quantified by comparing the actual execution order with the set order, calculating the compliance rate of the task execution order of the target business process, and calculating whether each task meets its prerequisites before execution, that is, calculating the prerequisite satisfaction rate; the relevance of the target business process can be quantified by calculating the information transmission efficiency of the information in the business process; and the dynamics of the target business process can be quantified by obtaining the adjustment frequency of the target business process.

[0061] In addition, business information refers to data and information generated in the business process and related to the business process, including but not limited to: a. Input information, which is the data required at the beginning of the business process, such as customer orders, customer information and other data in the product sales process; b. Process information, which is the new data and information generated by processing the input information in the business process; c. Output information, which refers to the final data generated at the end of the business process.

[0062] S103: Determine a target splitting strategy set based on process characteristics.

[0063] Specifically, the target splitting strategy set is a set of process splitting strategies adopted for the overall target process. Exemplarily, the target process input by the user may include one or more sub-processes that need to be split (i.e., target sub-processes). For each sub-process that needs to be split, the splitting strategies that can be adopted may include multiple (one or more of the splitting strategies according to the common sequence, the splitting strategies according to the business, the splitting strategies according to the role, and the splitting strategies according to the structure). In the embodiment of the application itself, the target sub-process (i.e., the sub-process that needs to be split) can be first determined, and then the process splitting strategies that can be adopted for each sub-process can be determined based on the target sub-process, the process characteristics and the user's intention, and then the process splitting strategy set adopted for the overall target process can be determined based on the determined process splitting strategies adopted for each target sub-process.

[0064] Specifically, the preset process splitting strategies (splitting strategy according to common sequence, splitting strategy according to business, splitting strategy according to role, and splitting strategy according to structure) are four basic splitting strategies obtained after analyzing and studying a large number of processes. They can basically achieve full coverage of the situations of processes that need to be split. For different target processes, one or more suitable process splitting strategies can be quickly determined according to the characteristics of the target sub-processes.

[0065] Among them, the target splitting strategy set refers to a strategy set formed by determining a corresponding process splitting strategy for all sub-processes that need to be split in the target process, that is, if the target process includes N target sub-processes, then the target splitting strategy set contains N corresponding process splitting strategies. It can be understood that one or more target splitting strategy sets can be determined based on one or more process splitting strategies corresponding to each sub-process that needs to be split in the target process (for example, multiple target splitting strategy sets can be determined by permutations and combinations).

[0066] Specifically, the sub-processes that need to be split (i.e., target sub-processes) mainly include the following situations: sub-processes with more complex processes, i.e., the sub-process includes multiple steps or decision points, and the logical relationship between the steps or decision points is more complex; sub-processes with universality, i.e., the sub-process can be reused, and splitting it can improve the reuse rate of the process; sub-processes involving collaboration among multiple departments or teams; sub-processes containing tasks that can be executed in parallel, and splitting it can more easily realize the parallel execution of tasks and improve the execution efficiency of business processes.

[0067] S104: Perform splitting and reorganization based on the target splitting strategy set.

[0068] Specifically, the target process is split and reorganized based on the determined one or more target splitting strategy sets.

[0069] In one possible implementation, for any target splitting strategy set: the target process includes N target sub-processes, then according to the N process splitting strategies included in the target process splitting strategy set, the target process is traversed in the logical order of the process to respectively determine the N target sub-processes, and the N sub-processes are split and reorganized according to the corresponding process splitting strategies.

[0070] The preset process splitting strategies include one or more of the splitting strategies according to common sequence, the splitting strategies according to business, the splitting strategies according to roles and the splitting strategies according to structure. Splitting according to common sequence means identifying the common parts or repeated steps in the business process, and extracting these common sequences separately as an independent process or module; splitting according to business means splitting according to different business goals or functions of the business process; splitting according to roles means splitting according to different roles or user groups involved in the business process; splitting according to structure means identifying key process nodes in the business process, and using these process nodes as the basis for splitting.

[0071] Exemplarily, after splitting and reorganizing according to one or more target splitting strategy sets, one or more corresponding replacement processes can be obtained, wherein the replacement process except the target sub-process can be the same as the original target process, or can be appropriately adjusted according to the multiple target sub-processes after splitting, so as to make the optimized target process (i.e., the replacement process) more harmonious and smooth.

[0072] The implementation of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0073] Figure 3 It is a schematic diagram of a framework of a process management method provided in an embodiment of the present application.

[0074] like Figure 3As shown, first, obtain user input. Specifically, Figure 3 The dotted box corresponding to "Get user input" includes three user input data, among which the solid box "Target process" is a mandatory option, that is, the user must at least enter the target process, which can be a complete process (for example, a logically complete set of business processes including a start node and an end node and connecting the two), or a sub-process that needs to be split. Among them, the target process can include: (1) Business process: A business process is a series of activities completed by different people to achieve a specific value goal. There is not only a strict sequence between the activities, but also the content, method, responsibility, etc. of the activities are clearly arranged and defined, so that different activities can be transferred between different job roles; (2) Service process: refers to the detailed analysis and sorting of customer requirements to clarify the specific content and implementation methods of the requirements; (3) Process flow: refers to a series of processing, conversion and processing steps from raw materials to final products or services in industrial production; (4) Development process: The development process refers to the program writing, testing, and debugging based on the design. etc., and finally realize the development and deployment of the system. The development process includes coding, testing, integration and other links; (5) Operation and maintenance process: The operation and maintenance process refers to the daily operation and maintenance management work after the system is launched to ensure the stable operation and security of the system. The operation and maintenance process includes monitoring, troubleshooting, backup and other links; (6) Quality management process: The quality management process refers to the management and control of project quality during the entire project development process. The quality management process includes quality planning, quality control, quality assessment and other links; (7) Project management process: The project management process refers to the process of planning, organizing, coordinating and controlling the entire project. The project management process includes project establishment, project planning, project execution, project monitoring and project closing and other links.

[0075] In addition, the dashed boxes “expected splitting strategy” and “user characteristics” are optional items, that is, the embodiment of the present application can also receive the expected splitting strategy and user characteristics input by the user.

[0076] The expected splitting strategy may be a descriptive statement, a decision point in the target process, or a preset splitting strategy. The preset splitting strategy here includes splitting strategies such as splitting strategies according to a common sequence, splitting strategies according to business, splitting strategies according to roles, and splitting strategies according to structures.

[0077] In addition, the input form of the desired split strategy may include text, image, audio, reference business process, etc. Exemplarily, the user may use manual input or voice input to enter a descriptive sentence, such as "ensure the smoothness of core business processes", "strengthen the coordination and integration of internal organizations", "optimize customer management processes, improve sales efficiency and customer satisfaction", etc., or upload one or more pictures, which may include text (such as the text corresponding to the descriptive sentence manually or voice-inputted above), an architecture diagram of the reference process, etc., and may also upload a reference process similar to the target process, that is, the process that the user wishes to refer to (the user may refer to its structural characteristics, title description, etc.). Similarly, the reference process may be a complete process or a sub-process.

[0078] In addition, if Figure 3 As shown, the embodiment of the present application can also receive user characteristics input by the user, and the user characteristics refer to the user portrait of the target user who creates the target business process. The user portrait mainly reflects the specific behaviors and preferences of the target user in process management, including but not limited to: process creation habits, for example, some users prefer to create concise and clear processes, and some users prefer to create detailed but slightly complex processes; process execution methods, including tools, methods or strategies used when executing processes, for example, some users prefer to use automated tools, and some users prefer manual operations; team collaboration and communication, such as the collaboration method and communication habits between teams in business processes, for example, some users like instant messaging, and some users choose to use email or meetings to communicate.

[0079] For example, the user may input the target user's role in the process, operating permissions, operating habits (such as commonly used functions, operating paths, etc.), problems and requirements encountered by the target user in the process, etc.

[0080] like Figure 3 As shown, further, an extraction strategy set is determined. Figure 4 is a flow chart of determining an extraction strategy set provided by an embodiment of the present application. Figure 4 As shown, determining the extraction strategy set may include steps S201 to S205.

[0081] S201: Extracting features of the target process and user features to obtain process features.

[0082] Specifically, since user features are optional, when no user features are received, only feature extraction is performed on the target process to obtain process features; when user features are input by the user, the user features are fused and analyzed with the target process to obtain process features.

[0083] S202: Analyze the expected splitting strategy to generate user intent.

[0084] Specifically, since the expected splitting strategy is also optional, this step is ignored when the expected splitting strategy is not received; when the expected splitting strategy is received, the expected splitting strategy input by the user can be analyzed and learned by using a model or other method to generate the user intention.

[0085] In a possible implementation, after receiving the expected splitting strategy input by the user, a Chain of Thought (CoT) can be constructed based on the expected splitting strategy to provide prompts to the Large Language Model (LLM) to generate user intent through the large language model. Among them, the Chain of Thought CoT is an improved text or instruction (Prompt) technology that provides input to the model to guide it to generate specific outputs. Its purpose is to improve the performance of the large language model LLM in complex reasoning tasks, such as arithmetic reasoning, commonsense reasoning, symbolic reasoning, etc. In the embodiment of the present application, CoT can be achieved by requiring the model to output intermediate step-by-step reasoning steps to the large language model before generating user intent.

[0086] It should be noted that the above steps S201 and S202 can be executed in parallel, and the numbers are only used to distinguish different steps.

[0087] S203: Determine the target sub-process in the target process.

[0088] A target process may include one or more target sub-processes that need to be split. The embodiment of the present application can automatically determine the target sub-processes that need to be split in the target process. In a possible implementation, several process nodes can be determined as a target sub-process based on the tasks completed by the process nodes, the roles performed, and the logical relationship between the process nodes. N target sub-processes can be determined in a target process, where N is an integer greater than or equal to 1.

[0089] S204: Determine a target sub-splitting strategy set corresponding to the target sub-process.

[0090] Specifically, N target sub-processes can be determined in the target process. Based on the generated user intention and process characteristics, and the preset multiple process splitting strategies, the target sub-splitting strategy set adopted for each target sub-process is determined. Specifically, the target sub-splitting strategy set corresponding to each target sub-process may include M preset splitting strategies, M is an integer greater than 1, and the number of preset splitting strategies included in the target sub-splitting strategies corresponding to the N target sub-processes may be different, that is, the M corresponding to the N target sub-processes may not be equal; further, the process splitting strategies corresponding to the N target processes are scored respectively, taking into account the actual situation of the business process and the preferences and splitting intentions of the target users.

[0091] In a possible implementation, after obtaining the target process input by the user, the process characteristics and the user intention are obtained, and N target sub-processes in the target process are determined. In addition, through the analysis and processing of the process characteristics and the user intention, it can be obtained that the user prefers to create business processes according to roles and perform tasks according to roles and departments in historical behaviors;

[0092] For one of the N target sub-processes, analysis shows that there are many identical task nodes in the target sub-process, and it is more suitable to adopt a strategy of splitting according to a common sequence; through the analysis of the process characteristics and the target sub-process, four preset process splitting strategies are determined, namely, splitting according to a common sequence strategy, splitting according to business strategy, splitting according to roles strategy and splitting according to structure strategy, and M is 4 at this time.

[0093] Furthermore, the four process splitting strategies are analyzed and calculated respectively to obtain the corresponding confidence levels.

[0094] Exemplarily, through analysis of the target sub-process, it can be found that there are 9 task nodes in the target sub-process, of which 5 task nodes are the same task nodes.

[0095] Exemplarily, the calculation method for the strategy confidence of splitting according to the common sequence can be:

[0096] C=V l ·W l +V p ·W p

[0097] Among them, C is the confidence level, which ranges from 0 to 1; V l is the ratio of repeated task nodes to all process nodes of the target sub-process, which is used to measure the length of repeated nodes (common sequence). l The specific calculation method can be:

[0098]

[0099] Among them, N r is the number of identical task nodes in the target subprocess, N a The total number of task nodes for the target sub-process;

[0100] W l is the weight of the proportion of repeated nodes to all task nodes of the target sub-process, that is, V l The weight is chosen as 0.6 here; V p Score the matching degree between the current strategy and the target user’s preference, which is used to measure the degree to which the current splitting strategy meets the target user’s preference. p The specific calculation method can be:

[0101]

[0102] Where P is the matching degree between the current strategy and the user's expectations and preferences, and its value is an integer between 1 and 10. The larger the value, the more the current split strategy conforms to the target user's behavioral characteristics and preferences. The matching degree between the current strategy and the user's expectations and preferences is determined based on the process characteristics and the comparison between the user's intention and the current strategy. N p is the total number of matching levels, which is 10 here;

[0103] W p The weight of the score for the matching degree between the current strategy and the target user’s preference, i.e., V p The weight is chosen as 0.4 here, where V l +V p is 1.

[0104] For example, for the above embodiment, the ratio of repeated character nodes to all task nodes of the target sub-process is After comparing and analyzing the process characteristics and user intentions with the common sequence splitting strategy, the P value is 4, and the matching score between the current strategy and the target user preference is Then the confidence of the strategy split according to the common sequence is C = V l ·W l +V p ·W p =0.55555·0.6+0.4·0.4=0.49333.

[0105] For example, for the strategy of splitting by business, the confidence calculation method may be:

[0106] C=V m ·W m +V p ·W p

[0107] Among them, similar to the calculation of confidence in splitting according to the common sequence, C is the confidence, with a value between 0 and 1; V m Score the business confusion of the target sub-process to measure the degree of business confusion or overlap between the task nodes contained in the target sub-process, that is, whether the tasks in the target sub-process are mixed, most links are in the charge of a large department or team, and the boundaries between departments are unclear. Specifically, the calculation method can be:

[0108]

[0109] Among them, M is the degree of business confusion, which is an integer between 1 and 10. The larger the value, the more complicated the target sub-process is and the less clear the functional boundaries are. m is the total number of levels of the business chaos, which is 10 here; W m The weight of the business chaos score, V m The weight of is chosen as 0.6 here;

[0110] Exemplarily, similar to the calculation of confidence in splitting according to the common sequence, V p Score the matching degree between the current strategy and the target user’s preference, which is used to measure the degree to which the current splitting strategy meets the target user’s preference. p and W p The specific calculation method of can refer to the relevant description of the strategy of splitting according to the common sequence above;

[0111] For example, for the above embodiment, after analyzing the target sub-process, M takes a value of 4, and the business chaos score of the target sub-process is After comparing and analyzing the process characteristics and user intentions with the business splitting strategy, the P value is 4, and the matching score between the current strategy and the target user preference is Then the confidence of the business split strategy is C = V l ·W l +V p ·W p =0.4·0.6+0.4·0.4=0.4.

[0112] For example, for the strategy of splitting by role, the confidence calculation method may be:

[0113] C=V c ·W c +V p ·W p

[0114] Among them, similar to the calculation of confidence in splitting according to the common sequence, C is the confidence, with a value between 0 and 1; Vc is the ratio of task nodes that can be completed by the same department or functional personnel to the total task nodes of the target sub-process, V c The specific calculation method can be:

[0115]

[0116] Among them, N c is the number of task nodes that can be completed by the same department or functional personnel in the target sub-process. Specifically, if the target sub-process mainly involves two functional personnel, then the sum of the number of task nodes completed by the two functional personnel is the number of task nodes that can be completed by the same department or functional personnel in the target sub-process, that is, N c , N a is the total number of task nodes of the target sub-process; W c is the weight of the proportion of task nodes that can be completed by the same department or functional personnel to the total task nodes of the target sub-process, that is, V c The weight of is chosen as 0.6 here;

[0117] Exemplarily, similar to the calculation of confidence in splitting according to the common sequence, V p Score the matching degree between the current strategy and the target user’s preference, which is used to measure the degree to which the current splitting strategy meets the target user’s preference. p and W p The specific calculation method of can refer to the relevant description of the strategy of splitting according to the common sequence above;

[0118] For example, for the above embodiment, after analyzing the target sub-process, the number of task nodes in the target sub-process that can be completed by the same department or functional personnel is 4, and the proportion of task nodes completed by the same department or functional personnel to the total task nodes of the target sub-process is After analyzing the process characteristics, user intentions, and role-based splitting strategies, the P value is 7, and the matching score between the current strategy and the target user preference is Then the confidence of the strategy split according to the common sequence is C = V l ·W l +V p ·W p =0.44444·0.6+0.7·0.4=0.546664.

[0119] For example, according to the strategy of structural splitting, in a possible implementation, the target sub-process includes a decision point, and the sub-process is mainly determined by the decision node for the subsequent process, then the strategy of structural splitting for the target sub-process can be specifically split according to the decision point, and the confidence calculation method of the splitting strategy can be:

[0120] C=V d ·W d +V p ·W p

[0121] Among them, similar to the calculation of confidence in splitting according to the common sequence, C is the confidence, with a value between 0 and 1; V d Score the decision point to measure the degree to which the target sub-process is suitable for splitting according to the decision point. d The specific calculation method can be:

[0122]

[0123] Where D is the decision complexity index, which is used to measure whether the target sub-process includes decision points, the complexity of the judgment conditions of the decision points, and whether the target sub-process is mainly based on conditional judgment to determine the processing steps. The value is between 0 and 10. When there is no decision point in the target sub-process, the value is 0. The larger the D is, the more complex the judgment conditions of the decision points in the target sub-process are, and the target sub-process is mainly based on conditional judgment to determine the processing steps, that is, the more suitable it is to be split according to the decision points; N d is the total number of levels of the business chaos, which is 10 here; W d The weight for scoring the decision point, i.e., V d The weight of is chosen as 0.6 here;

[0124] Exemplarily, similar to the calculation of confidence in splitting according to the common sequence, V p Score the matching degree between the current strategy and the target user’s preference, which is used to measure the degree to which the current splitting strategy meets the target user’s preference. p and W p The specific calculation method of can refer to the relevant description of the strategy of splitting according to the common sequence above;

[0125] For example, for the above embodiment, after analyzing the target sub-process, the decision complexity index D is 6, and the decision evaluation is divided into After analyzing the process characteristics, user intentions, and role-based splitting strategies, the P value is 4, and the matching score between the current strategy and the target user preference is Then the confidence of the strategy split according to the common sequence is C = V l ·W l +V p ·W p =0.6·0.6+0.4·0.4=0.52.

[0126] It should be understood that the above confidence calculation method is only an exemplary description of the embodiment of the present application and does not constitute a specific limitation on the embodiment of the present application. The confidence calculation can also be carried out by other methods, which are not limited in the embodiment of the present application.

[0127] Furthermore, the M splitting strategies included in the target splitting strategy set are sorted from high to low according to the calculated confidences, and the M process splitting strategies and the confidence sorting results between the M process strategies constitute a target sub-split strategy set.

[0128] Exemplarily, the ranking results of the four splitting strategies included in the above embodiment are: strategy of splitting according to role (confidence is 0.546664), strategy of splitting according to structure (confidence is 0.52), strategy of splitting according to common sequence (confidence is 0.49333), and strategy of splitting according to business (confidence is 0.4).

[0129] S205: Determine a target splitting strategy set.

[0130] Specifically, at least one process splitting strategy is determined from each target sub-splitting strategy set corresponding to the N target sub-processes determined in step S204 to form one or more target splitting strategy sets.

[0131] In a possible implementation, the process splitting strategy with the highest confidence in N target sub-splitting strategy sets can be selected to form a target splitting strategy set. Specifically, for each target sub-splitting strategy set corresponding to each target sub-process, a preset process splitting strategy with the highest confidence ranking is selected. In this way, N process splitting strategies can be selected, and each splitting strategy corresponds to a unique target sub-process. The selected N process splitting strategies constitute a target splitting strategy set. In addition, the process splitting strategy with the second highest confidence ranking in each target sub-splitting strategy set can also be selected accordingly, and so on;

[0132] In another possible implementation, if the number of target sub-processes and the process splitting strategies included in the target sub-splitting strategy set corresponding to all target sub-processes are small (i.e., the sum of "N" and "M" corresponding to all target sub-processes is small), the process splitting strategies included in the target sub-splitting strategy set corresponding to the N target sub-processes can be arranged and combined to obtain multiple target splitting strategy sets. Similarly, each determined target splitting strategy set includes N process splitting strategies, and each process splitting strategy corresponds to a unique target sub-process. Specifically, if a certain target process includes 3 target sub-processes, 3 corresponding target sub-splitting strategy sets can be determined for the 3 target sub-processes: target sub-splitting strategy set 1, target sub-splitting strategy set 2, and target sub-splitting strategy set 3, wherein target sub-splitting strategy set 1 includes preset process splitting strategy 11 and preset process splitting strategy 12 (and the order between the two represents the confidence ranking), target sub-splitting strategy set 2 includes preset process splitting strategy 21 and preset process splitting strategy 22 (the order between the two represents the confidence ranking), and target sub-splitting strategy set 1 includes preset The process splitting strategy is 31, and each preset process splitting strategy corresponds to a specific confidence level. Then, after permutations and combinations, four target splitting strategy sets can be obtained: (1) preset process splitting strategy 11, preset process splitting strategy 21, preset process splitting strategy 31; (2) preset process splitting strategy 11, preset process splitting strategy 22, preset process splitting strategy 31; (3) preset process splitting strategy 12, preset process splitting strategy 21, preset process splitting strategy 31; (3) preset process splitting strategy 12, preset process splitting strategy 22, preset process splitting strategy 31.

[0133] It should be understood that the above-mentioned method for determining the target splitting strategy set is only an exemplary description of the embodiment of the present application and does not constitute a specific limitation on the embodiment of the present application. The target splitting strategy set can also be determined by other methods, which is not limited by the embodiment of the present application.

[0134] like Figure 3 As shown, after the process of determining the extraction strategy set is completed, the target process will be split and reorganized according to the determined target splitting strategy set.

[0135] In order to explain the steps of splitting and reorganization more specifically, Figure 5 , Figure 6 Provide explanation.

[0136] Figure 5 A schematic diagram of a splitting and reorganization process provided in an embodiment of the present application, such as Figure 5 As shown, the splitting and reassembling includes steps S301-S304.

[0137] S301: Select a target splitting strategy set.

[0138] Specifically, one or more target splitting strategy sets can be obtained through the above step S205. When there are too many target splitting strategy sets, the confidence of each target splitting strategy set can be calculated separately according to the confidence of the process splitting strategy included in the target splitting strategy set, so as to select one or more with higher confidence ranking as the final target strategy set. This can avoid splitting and reorganization according to more target strategy sets, reduce the amount of calculation, and improve efficiency. In addition, it can also provide users with a reasonable number of high-quality optimization solutions to enhance user experience.

[0139] S302: Routing to the process splitting strategy according to the target splitting strategy set.

[0140] Specifically, splitting is performed according to one or more target splitting strategy sets. Exemplarily, for each target splitting strategy set: route to the specified splitting method according to the process splitting strategy included in the target splitting strategy set. For example, the target splitting strategy set includes 4 process splitting strategies: strategy for splitting according to structure, strategy for splitting according to role, strategy for splitting according to common sequence, and strategy for splitting according to business, and each process splitting strategy corresponds to a unique target sub-process (that is, the target process includes 4 target sub-processes), then for the strategy for splitting according to structure, it can jump to the decision node and subsequent process in the corresponding target sub-process that are split according to decision points according to the target process; similarly, for the strategy for splitting according to role, it will jump to the corresponding character node and subsequent process in the corresponding target sub-process that can be completed by the same department or functional personnel.

[0141] In addition, step S302 can be executed in parallel by routing to the corresponding N process splitting strategies according to one or more target splitting strategy sets, thereby improving the efficiency of target process processing.

[0142] S303: Split and reorganize the target sub-process according to the routing strategy.

[0143] For example, for the strategy of splitting by structure: it can be routed to the decision node represented by the "branch" and split; for the strategy of splitting by role: it can be routed to the task node marked with a certain person or title in the target sub-process and split; for the strategy of splitting by common sequence: it can be routed to repeated task nodes; for the strategy of splitting by business, it can be routed to task nodes with business characteristics (which can obviously be completed by a functional department or team).

[0144] In order to more clearly illustrate the splitting and reorganization of the target sub-process, the following is combined Figure 6 and Fig. 7A , Figure 7B , Figure 7C Provide specific instructions.

[0145] Figure 6 is a target process diagram provided in an embodiment of the present application, wherein the target process is a business process (hereinafter referred to as the target business process), such as Figure 6 As shown, the target business process includes two relatively independent processes. Figure 6 The process from the start node to the end node on the left is the campus recruitment briefing reservation process. Figure 6 The process from the start node to the end node on the right is the conference booking process, and the conference booking process includes a decision node, namely Figure 6 The node where the "branch" is located in the decision node indicates that different processes can be adopted according to the type of the booking personnel. For example, if the meeting type is "operational", a medium-sized meeting is booked, and the assistant prepares fruit slices and drinks. If the meeting type is communication, a small meeting is booked. If the meeting type is seminar, a large meeting is booked, and materials and agendas are prepared, and the assistant prepares tea breaks. A target sub-process can be determined in this target process, that is, Figure 6 The nodes included in the dashed box are sub-processes, and N is 1 at this time.

[0146] In one possible implementation, Figure 6 The target sub-process shown determines the target sub-splitting strategy set, which includes three process splitting strategies, and has completed confidence calculation and sorting. The sorting results are: splitting by structure, splitting by role, and splitting by common sequence. At this time, the M corresponding to the target sub-splitting strategy set is 3. According to the three process splitting strategies, they are routed to the corresponding task nodes for splitting and reorganization.

[0147] For example, based on the strategy of splitting according to structure, Figure 6 By splitting the target sub-process shown in the figure, we can get Fig. 7A The replacement sub-process 1 is shown. Fig. 7A This is a schematic diagram of a replacement sub-process obtained by structural splitting provided by an embodiment of the present application. Fig. 7A As shown, replace subprocess 1 with the decision node ( Figure 6 The process nodes in the target sub-process are split and reorganized. Specifically, Figure 6 The "Book a Medium-sized Meeting" node in the target sub-process shown and the subsequent "Assistant-Prepare Fruit Cuts and Drinks" process node are split and reorganized and replaced with "Sub-process 1-Book a Medium-sized Meeting"; Figure 6 The "Book a large meeting" node in the target sub-process shown, as well as the subsequent "Prepare agenda and materials" and "Assistant-Prepare tea break" process nodes are split and reorganized and replaced with "Sub-process 2-Book a large meeting"; Figure 6The process node of "Book a small meeting" in the target sub-process shown is split and reorganized and replaced with "Sub-process 3-Book a small meeting"; for Figure 6 The process nodes of the independent process on the left side of the target sub-process are not processed. The title of the process node of the replacement sub-process (eg, "Sub-process 1 - Book a Medium-sized Meeting") can be generated by the model based on the content of the target sub-process.

[0148] For example, based on the strategy of splitting by role, Figure 6 By splitting the target subprocess shown in Figure 7B The replacement sub-process 2 is shown. Figure 7B Schematic diagram of a replacement sub-process obtained by splitting according to roles provided in an embodiment of the present application. Figure 7B As shown, according to Figure 6 The target business process shown can be determined Figure 7B The three roles shown in the dotted boxes are "Human Resources", "Assistant" and "Finance", and the process nodes included in the target sub-process are matched with the roles. The relative positions of the roles and the process nodes corresponding to the roles in the replacement sub-process are determined according to the execution order between the processes, and reorganized according to the relative positions. Among them, the process nodes corresponding to the roles are horizontally aligned with the dotted boxes on the right side of each dotted box ( Figure 7B In the figure, each role and the corresponding task node are separated by a dotted horizontal line, and the process nodes are connected by sequence flows (broken lines) with arrows to indicate the execution and logical order, where the dotted sequence flows represent optional ones, that is, after "booking a meeting room or classroom", the process node of "preparing fruit cuts and drinks" may be executed, or the process node of "preparing a tea break" may be executed, or the process node of "preparing small gifts" may be executed; in addition, the three roles are generated through model analysis based on the target business process and target sub-process.

[0149] The role-based splitting strategy can be split and reorganized across processes, for example Figure 7B The replacement subprocess shown would Figure 6 The task nodes belonging to two relatively independent processes in the target sub-process shown are reorganized in the replacement sub-process 2, that is, it cannot be obviously seen that the replacement sub-process contains the task nodes of two relatively independent processes. In this way, the whole process can be taken into consideration for process optimization, and after the target sub-process is split and reorganized, the overall target business process can be adjusted according to the replacement sub-process after the split and reorganization, so that the overall target business process is also optimized according to the roles, making the target business process smoother and more harmonious.

[0150] Exemplarily, based on the strategy of splitting according to the common sequence, Figure 6 The target sub-process shown in the figure can be split into Figure 7CThe replacement sub-process 3 is shown. Figure 7C Schematic diagram of a replacement sub-process obtained by splitting a common sequence according to an embodiment of the present application. Figure 7C As shown, splitting by common sequence is similar to splitting by role, and can also be split and reorganized across processes. Figure 6 The two relatively independent processes in the target business process shown can be reorganized into the same replacement sub-process 3 according to the common sequence. Figure 7C As shown, replace sub-process 3, Figure 6 In the target sub-process shown, "booking a classroom" - "preparing promotional materials" - "assistant - preparing small gifts", "booking a medium-sized conference room" - "assistant - preparing cut fruits and drinks", "booking a large meeting" - "preparing an agenda and materials" - "assistant preparing tea breaks" are extracted, split and reorganized to determine a common sequence: "booking medium and large conference rooms and classrooms" - "preparation actions", where the preparation actions include the assistant preparing tea breaks, cut fruits and drinks, preparing agendas and materials (including promotional materials) and other preparation actions; for booking a small conference room, no preparation actions are required and the process ends directly.

[0151] also, Figure 6 The target business process shown involves relatively simple businesses, so the strategy of splitting by business is not adopted (that is, the target splitting strategy set does not include the strategy of splitting by business).

[0152] It should be noted that the above Figure 6 , Figure 7A-7C For the purpose of exemplary display of the embodiments of the present application, the target process (referred to as the "target business process" above), the target sub-process and the replacement sub-process may be of other styles, which are not limited by the embodiments of the present application.

[0153] Further, in order to illustrate the strategy of splitting by business, Figure 8 and Fig. 9 Provide specific instructions.

[0154] Figure 8 is another target flow diagram provided in an embodiment of the present application, and Figure 8 The target process shown is also a business process, so the following Figure 8 The target process shown is called the "target business process". Combining user input and analysis of the target business process, it can be found that the target business process mainly involves the design, production, sales and improvement of two products: mobile phones and home appliances. In the target process, each task node is directly connected and executed in a logical order, and it is impossible to distinguish the business parts (for example, which steps are for design and production, and which steps are for after-sales service). The readability of the process is poor and the process nodes are too cumbersome. Through analysis, it can be determined that Figure 8The target business process shown includes three target sub-processes (target sub-process 1, target sub-process 2 and target sub-process 3), as shown in FIG. Figure 8 As shown, one or more process nodes in each dotted box constitute a target sub-process.

[0155] Fig. 9 Schematic diagram of a replacement sub-process generated according to business splitting and reorganization provided by an embodiment of the present application, such as Fig. 9 As shown, Figure 8 The three target sub-processes shown can be split and reorganized into three corresponding replacement sub-processes, where Figure 8 The target sub-process 1 shown can be split and reorganized into a replacement sub-process 1, Figure 8 The target sub-process 2 shown can be split and reorganized into a replacement sub-process 2, Figure 8 The target sub-process 3 shown can be split and reorganized into replacement sub-process 3, so that design production, sales, and after-sales support can be distinguished and solidified into reusable sub-processes, which can be reused when building other processes or optimizing this process. Fig. 9 and Figure 8 The comparison can clearly reflect the strategy of splitting according to business.

[0156] S304: Replace the split and reorganization results back into the target process to obtain a replacement process.

[0157] Through the above steps, the splitting and reorganization results corresponding to one or more target splitting strategy sets can be obtained (the splitting and reorganization results corresponding to each target splitting strategy set include the splitting and reorganization results of N target sub-processes in the target process), and the target process can be replaced according to the splitting results of one or more target splitting strategy sets. Specifically, the N replacement sub-processes corresponding to the target splitting strategy set in the above step S303 can be replaced back to the target process.

[0158] For example, for Figure 6 The target process shown and Figure 7A-7C The replacement sub-process shown in the figure can be obtained after the replacement process Figure 7D-7F Replacement process shown. Fig.7D is a replacement flow diagram provided in an embodiment of the present application, such as Fig.7D As shown, Fig. 7A After the replacement subprocess 1 shown is replaced back to the target process, the replacement process 1 can be obtained; Fig. 7E is another replacement flow diagram provided in the embodiment of the present application, such as Fig. 7E As shown, Figure 7B After the replacement sub-process 2 shown is replaced back to the target process, the replacement process 2 can be obtained, in which the sequence flow in the form of a dotted line ( Fig. 7EThe broken line connecting two process nodes in the figure represents optional, that is, after executing the process node of "booking a meeting room or classroom", you can choose to execute the next step of "preparing fruit cuts and drinks", "preparing a tea break", "preparing small gifts" or "ending" according to the execution result; Figure 7F is another replacement flow diagram provided in the embodiment of the present application, such as Figure 7F As shown, Figure 7C After the replacement sub-process 3 shown is replaced back to the target process, the replacement process 3 can be obtained; similarly, Fig.10 is another replacement flow diagram provided in the embodiment of the present application, Fig. 9 Replace sub-process 1, replace sub-process 2, replace sub-process 3 and replace Figure 8 The target process shown can be obtained Fig.10 Replacement process shown.

[0159] Further, after generating one or more replacement processes, corresponding titles and related descriptions can be generated for the one or more replacement processes through the large language model. Fig.11 A detailed description will be given. Fig.11 It is a schematic diagram of a process for generating a title and description for a replacement process provided by an embodiment of the present application.

[0160] In a possible implementation, for each replacement process: first, obtain the replacement process and the process context of the target sub-process corresponding to the replacement sub-process in the replacement process in the target process, wherein the process context includes adjacent process blocks (including one or more process nodes) in a logical order of the target sub-process corresponding to the replacement sub-process and information provided by the adjacent process blocks; construct a prompt word based on the obtained process context of the replacement process and the target sub-process corresponding to the replacement sub-process in the target business process, and the prompt word is used to prompt the large language model with the process context of the replacement sub-process and the parameter information of the replacement process, which can guide the large language model to generate a title and related description that is more in line with the content of the replacement process; the prompt word, the replacement process and the corresponding process context are input into the large language model, and the title and related description are generated through analysis by the large language model.

[0161] After generating the corresponding titles and descriptions, one or more replacement processes and the corresponding titles and descriptions may be saved.

[0162] For example, for Fig.7DThe replacement process 1 shown in the figure can generate a title through the analysis of the above large language model: split by decision point "branch"; and generate relevant descriptions: split and reorganize the three subsequent processes corresponding to the decision point "branch" into three sub-processes, namely "booking a large meeting", "booking a medium-sized meeting", and "booking a small meeting", which simplifies the complexity, realizes the reuse and modularization of the process, and realizes the parallel processing of the process;

[0163] for Fig. 7E The replacement process 2 shown in the figure can generate a title through the analysis of the above large language model: split by role; and can generate relevant descriptions: the analysis determines the three functional roles involved in the process, and divides the tasks by role, improves the process, clarifies the role tasks, and strengthens departmental cooperation;

[0164] for Figure 7F The replacement process 3 shown in the figure can generate a title through the analysis of the above large language model: split according to the common sequence; generate relevant descriptions: comprehensively analyze the overall target business process, integrate similar task nodes of "booking XXX" and "preparing actions" in the two processes, and establish a unified standard for easy reuse and promotion;

[0165] for Fig.10 The replacement process shown can generate a title through the analysis of the above large language model: Split by business; and can generate relevant descriptions: Splitting more complex businesses into relatively independent processes can enable functional personnel to focus more on their own business, improve operational efficiency, and reduce internal communication costs.

[0166] Further, after saving one or more replacement processes and corresponding titles and descriptions, the saved one or more replacement processes and corresponding titles and description information may be output so that the user may select at least one replacement process from the saved one or more replacement processes to complete the optimization of the target process.

[0167] For example, for Fig.7D , Fig. 7E , Figure 7F The replacement process shown and the above Fig.7D , Fig. 7E , Figure 7F The title and description can be generated Fig. 12A The card-style control shown, Fig. 12A Schematic diagram of an optimization strategy output interface provided by an embodiment of the present application. Fig. 12A As shown, according to Fig.7D , Fig. 7E , Figure 7FThe replacement process shown in the output interface displays three card controls, each of which represents an optimization plan. Each optimization plan is generated by a corresponding replacement process and the title and description corresponding to the replacement process. The card controls include a text title of "Optimization Plan 1\2\3", a title and description below "Optimization Plan 1\2\3", a "Comparison Chart Before and After Optimization" viewing control, and a confirmation control. The title and description are the title and description of the replacement process corresponding to the card control (that is, the title and related description generated by the large language model above). Click the "Comparison Chart Before and After Optimization" viewing control in each card control to view the comparison chart of the target process (before optimization) and the replacement process (after optimization). The display interface will be shown as follows: Fig. 12B The comparison diagram before and after optimization is shown in the figure, which is clearer after enlarging. In the comparison diagram before and after optimization, the upper part is the target process and the lower part is the replacement process. Fig. 12B This is a comparison diagram before and after optimization provided in the embodiment of the present application. Click Fig. 12B The picture shown can be exited to view the "before and after optimization comparison picture"; after the user confirms the optimization plan, he can click Fig. 12A Click the OK control shown to select the replacement process.

[0168] In addition, you can also control the number of alternative options presented to users by setting the number of card controls.

[0169] Furthermore, after receiving the user's selection input, the replacement process selected by the user is saved to complete the optimization of the target process.

[0170] Fig.13 A schematic diagram of the structure of a business process management device provided in an embodiment of the present application is shown in FIG. Fig.13 The business process management device 40 shown includes:

[0171] A receiving unit 401 is used to receive a target process input by a user;

[0172] The splitting unit 402 is used to parse the target process input by the user to obtain process characteristics; based on the process characteristics, determine a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies; based on the target splitting strategy set, split and reorganize the target process to generate a replacement process corresponding to the target process;

[0173] The storage unit 403 is used to store the replacement process.

[0174] Furthermore, the receiving unit 401 is further configured to receive a user feature input by a target user and a desired splitting strategy input by the target user, wherein the target process is constructed by the target user and the user feature is a user portrait of the target user;

[0175] The splitting unit 402 is specifically used to: determine the target sub-process to be split; analyze the expected splitting strategy to generate the user intention, and determine the target sub-splitting strategy set to be adopted for the target sub-process from the preset multiple process splitting strategies based on the user intention, user characteristics, and process characteristics; determine the target splitting strategy set based on the target sub-splitting strategy set, and split and reorganize the target process to obtain the corresponding replacement sub-process; generate a corresponding title and description for the replacement process;

[0176] The storage unit 403 is specifically used to obtain and store the replacement process and the title and description information corresponding to the replacement process.

[0177] Specifically, the process management device can achieve: obtaining a target process, and performing feature extraction based on the target process to obtain process features; based on the process features, determining a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies; based on the target splitting strategy set, splitting and reorganizing the target process to generate a replacement process corresponding to the target process.

[0178] Exemplarily, performing feature extraction based on the target process to obtain process features may include: receiving user features input by a target user, and performing feature extraction based on the user features and the target process to obtain process features; wherein the target process is constructed by the target user, and the user features are user portraits of the target user.

[0179] In a possible implementation, determining a target splitting strategy set to be adopted for the target process may include: determining N target sub-processes that need to be split in the target process, where N is an integer greater than or equal to 1; based on the process characteristics, determining a target sub-splitting strategy set to be adopted for each of the target sub-processes from a plurality of preset process splitting strategies; determining a target splitting strategy set based on the target sub-splitting strategy set of the N target sub-processes; further, based on the process characteristics, determining a target sub-splitting strategy set to be adopted for each of the target sub-processes from a plurality of preset process splitting strategies, including: receiving an expected splitting strategy input by a user, analyzing the expected splitting strategy, generating a user intent, wherein the input form of the expected splitting strategy includes one or more of text, image, audio, and reference business processes; based on the user intent and the process characteristics, and the plurality of preset process splitting strategies, determining a target sub-splitting strategy set to be adopted for each of the target sub-processes. strategy set; further, any one of the target sub-splitting strategy sets includes M preset process splitting strategies; the determination of the target sub-splitting strategy set to be adopted for each of the target sub-processes may include: for any target sub-process, scoring the corresponding M preset process splitting strategies based on the user intention and the process characteristics to obtain M confidences, where M is an integer greater than or equal to 1; sorting the M preset process splitting strategies from high to low according to the M confidences to obtain the sorting results of the M preset process splitting strategies; generating a target sub-splitting strategy set corresponding to the target sub-process based on the sorting results; determining the target splitting strategy set based on the target sub-splitting strategy set of the N target sub-processes may include: based on the sorting results of the confidences, determining at least one process splitting strategy from each target sub-splitting strategy set corresponding to the N target sub-processes to form one or more target splitting strategy sets.

[0180] Further, in a possible implementation, based on the target splitting strategy set, splitting and reorganizing the target process to generate a replacement process corresponding to the target process may include: based on the one or more target splitting strategy sets, splitting and reorganizing the target process to generate one or more replacement processes corresponding to the target splitting strategy sets; further, after obtaining the replacement process, based on the one or more replacement processes and the process features, a large language model can be used to generate titles and description information for the one or more replacement processes respectively; the one or more replacement processes and the corresponding titles and description information are saved; further, the one or more replacement processes and the corresponding titles and description information saved can be output to allow a user to select at least one replacement process; after receiving the user's selection input, the replacement process selected by the user is saved.

[0181] The above-mentioned preset process splitting strategies include: one or more of splitting strategies according to a common sequence, splitting strategies according to a business, splitting strategies according to a role, and splitting strategies according to a structure.

[0182] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A process management method, characterized in that: include: Acquire a target process, and perform feature extraction based on the target process to acquire process features; Based on the process characteristics, determining a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies; Based on the target splitting strategy set, the target process is split and reorganized to generate a replacement process corresponding to the target process.

2. The method according to claim 1, characterized in that The extracting features based on the target process to obtain process features includes: Receive user features input by a target user, and perform feature extraction based on the user features and the target process to obtain process features; the target process is constructed by the target user, and the user features are a user portrait of the target user.

3. The method according to claim 2, characterized in that The determining of a target splitting strategy set to be adopted for the target process includes: Determine the N target sub-processes that need to be split in the target process, where N is an integer greater than or equal to 1; Based on the process characteristics, determining a target sub-splitting strategy set to be adopted for each of the target sub-processes from a plurality of preset process splitting strategies; A target splitting strategy set is determined based on the target sub-splitting strategy sets of the N target sub-processes.

4. The method according to claim 3, characterized in that The step of determining a target sub-splitting strategy set to be adopted for each target sub-process from a plurality of preset process splitting strategies based on the process characteristics includes: Receive an expected splitting strategy input by a user, analyze the expected splitting strategy, and generate a user intent, wherein the input form of the expected splitting strategy includes one or more of text, image, audio, and reference business process; Based on the user intention and the process characteristics, and the preset multiple process splitting strategies, a target sub-splitting strategy set to be adopted for each of the target sub-processes is determined.

5. The method according to claim 4, characterized in that Any of the target sub-splitting strategy sets includes M preset process splitting strategies; The step of determining a target sub-splitting strategy set to be adopted for each target sub-process includes: For any target sub-process, score the corresponding M preset process splitting strategies respectively based on the user intention and the process characteristics to obtain M confidences, where M is an integer greater than or equal to 1; Sorting the M preset process splitting strategies from high to low according to the M confidences to obtain a sorting result of the M preset process splitting strategies; A target sub-splitting strategy set corresponding to the target sub-process is generated based on the sorting result.

6. The method according to claim 3, characterized in that The determining of the target splitting strategy set based on the target sub-splitting strategy sets of the N target sub-processes includes: Based on the ranking result of the confidence, at least one process splitting strategy is respectively determined from each target sub-splitting strategy set corresponding to the N target sub-processes to form one or more target splitting strategy sets.

7. The method according to claim 3, characterized in that The step of splitting and reorganizing the target process based on the target splitting strategy set to generate a replacement process corresponding to the target process includes: Based on the one or more target splitting strategy sets, the target process is split and reorganized to generate one or more replacement processes corresponding to the target splitting strategy sets.

8. The method according to claim 7, characterized in that , the method further comprises: Based on the one or more replacement processes and the process features, respectively generate titles and description information for the one or more replacement processes by using a large language model; The one or more replacement processes and corresponding titles and description information are saved.

9. The method according to claim 6, characterized in that , the method further comprises: Outputting the saved one or more alternative processes and corresponding titles and description information, so that a user can select at least one alternative process; After receiving the user's selection input, the replacement process selected by the user is saved.

10. The method according to claim 1, characterized in that The multiple process splitting strategies include: one or more of a splitting strategy according to a common sequence, a splitting strategy according to a business, a splitting strategy according to a role, and a splitting strategy according to a structure.

11. A process management device, characterized in that: The device comprises: A receiving unit, used for receiving a target process input by a user; A splitting unit is used to parse the target process input by the user to obtain process characteristics; based on the process characteristics, determine a target splitting strategy set to be adopted for the target process from a plurality of preset process splitting strategies; based on the target splitting strategy set, split and reorganize the target process to generate a replacement process corresponding to the target process; The storage unit is used to store the replacement process.

12. The device according to claim 11, characterized in that The receiving unit is further used to receive user features input by a target user and a desired splitting strategy input by the target user, the target process is constructed by the target user, and the user features are a user portrait of the target user.

13. The device according to claim 11, characterized in that The splitting unit is specifically used to determine the target sub-process that needs to be split; analyze the expected splitting strategy to generate the user intention, and determine the target sub-splitting strategy set to be adopted for the target sub-process from the preset multiple process splitting strategies based on the user intention, the user characteristics, and the process characteristics; determine the target splitting strategy set based on the target sub-splitting strategy set, and split and reorganize the target process to obtain the corresponding replacement sub-process; Generate a corresponding title and description for the replacement process.

14. The device according to claim 11, characterized in that The storage unit is specifically used to obtain and store the replacement process and the title and description information corresponding to the replacement process.

15. A server, characterized in that: include: A memory and one or more processors; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the process management method described in any one of claims 1-10.

16. An electronic device, comprising: memory, one or more processors; The memory is coupled to the one or more processors, and the memory is used to store computer program code, wherein the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the process management method described in any one of claims 1-10.

17. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and when the computer program is executed by a processor, the process management method described in any one of claims 1 to 10 is implemented.

18. A computer program product, comprising instructions, which, when executed by a computer, enable the computer to execute the process management method according to any one of claims 1 to 10.

Images