Task readiness assessment method and related apparatus

Abstract

Embodiments of the present application relate to the technical field of artificial intelligence, and disclose a task intelligent readiness assessment method and a related device, which comprises: constructing a task description model; inputting business information of a to-be-assessed business into the task description model to obtain structured description information corresponding to each task node in the to-be-assessed business; quantitatively evaluating each task node from multiple preset task evaluation dimensions according to the structured description information to obtain a quantitative evaluation score of each task node; determining an automation degree of each task node according to the quantitative evaluation score of each task node and an automation level determination rule; and determining a role suggestion corresponding to each task node of the to-be-assessed business according to the automation degree, wherein the role suggestion comprises an intelligent agent role, a human role or a hybrid role. Through the above manner, the embodiments of the present application realize effective allocation suggestion of task roles in enterprise workflows.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a method and apparatus for assessing the intelligent readiness of a task. Background Technology

[0002] With the development of large language models and AI agent technologies, organizations face a core decision-making problem: among numerous business tasks, which tasks are suitable for AI to automate, which still require human leadership, and to what extent should automation be implemented? However, existing technologies have several shortcomings in addressing this issue.

[0003] First, existing automation maturity models (such as the CMMI Capability Maturity Model and the SPICE Process Assessment Model) primarily focus on organizational-level process capability maturity assessment. The assessment object is the organization's overall process management level, rather than the automation readiness level of a single task. These models can answer the question of which level an organization's process management is at, but they cannot answer the question of whether a specific task should be performed by AI or a human.

[0004] Second, the feasibility assessment methods in the field of Robotic Process Automation (RPA) mainly evaluate whether a task is suitable for automation from two dimensions: rule determinism and the degree of data structuring. However, in the era of large language models, AI has the ability to process unstructured data and non-deterministic tasks, and the two-dimensional assessment framework of RPA is no longer sufficient to cover the decision space of AI automation. Dimensions such as the cognitive complexity of the task, interaction characteristics, and compliance constraints are all ignored.

[0005] Third, existing AI feasibility assessments are mostly qualitative judgments, producing a binary result of feasible or infeasible, lacking a quantitative grading mechanism. In real-world scenarios, the degree of automation is a continuous spectrum—there are multiple intermediate states between full automation and purely human execution.

[0006] Fourth, when it comes to industries such as finance and healthcare, compliance constraints usually have a more significant impact on automated decision-making than other factors, but existing assessment methods lack a mechanism to differentiate and weight compliance dimensions.

[0007] Therefore, there is an urgent need for an effective intelligent task readiness assessment method to provide effective suggestions for the allocation of task roles in enterprise workflows. Summary of the Invention

[0008] In view of the above problems, embodiments of the present invention provide a task intelligence readiness assessment method and related apparatus to solve the problem of the lack of an effective task intelligence readiness assessment method in the prior art.

[0009] According to one aspect of the present invention, a method for assessing task intelligence readiness is provided, the method comprising:

[0010] A task description model is constructed; the task description model is obtained by modeling using a graphical business process modeling language; the graphical business process modeling language includes task identifier, input artifact set, execution logic, output artifact set, responsibility role, constraint set, cognitive level, and deterministic classification;

[0011] The business information of the business to be evaluated is input into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level and target determinism classification corresponding to the business to be evaluated;

[0012] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension.

[0013] The degree of automation of each task node is determined based on the quantitative evaluation score of each task node and the automation level determination rules.

[0014] Based on the degree of automation, role suggestions are determined for each task node of the service to be evaluated; wherein, the role suggestions include intelligent agent roles, human roles, or hybrid roles.

[0015] In one alternative approach, constructing the task description model includes:

[0016] A graphical business process modeling language is established. This language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The input artifact set contains at least one input artifact, each with an artifact identifier, description, and data format attributes. The execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions. The constraint set includes time constraints, access constraints, quality constraints, and audit constraints, with each constraint associated with a policy identifier and policy level.

[0017] A three-layer recursive process model is constructed, comprising a process domain layer, an activity layer, and a task layer; wherein each process domain contains at least one activity, and each activity contains at least one task.

[0018] Define role executors for the process model; the role executors include human roles, large language model intelligent agent roles, system roles, and hybrid roles; wherein, the large language model intelligent agent roles are associated with intelligent agent specification parameters, and the hybrid roles are configured with both intelligent agent specification parameters and human ability descriptions;

[0019] Each task in the process model is formally described using the graphical business process modeling language to obtain the task description model.

[0020] In one optional approach, the business information includes business requirement documents, requirement review management systems, requirement development process standards, code review and quality standards, and information security and compliance management regulations; the step of inputting the business information of the business to be evaluated into the task description model to obtain structured description information corresponding to each task node in the business to be evaluated includes:

[0021] Based on the business information, the task description model is used to instantiate the business to be evaluated, thereby obtaining the structured description information corresponding to each task node in the business to be evaluated.

[0022] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node, including:

[0023] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0024] Based on the weights corresponding to each preset task evaluation dimension, the quantitative evaluation sub-scores of each preset task evaluation dimension are calculated by weighted average to obtain the quantitative evaluation score of each task node.

[0025] In one alternative approach, the cognitive complexity dimension includes the following levels: memory and retrieval level, understanding and interpretation level, application and preliminary analysis level, in-depth analysis and synthesis level, and evaluation and creation level;

[0026] The data dependency dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams.

[0027] The interaction diversity dimension includes the following levels: no interaction or pure API call, one-way notification or message push, form and tool interaction, multi-party communication and coordination and multi-party negotiation.

[0028] The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory laws and regulations.

[0029] The innovation requirements include the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making.

[0030] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation sub-score for each task node in each preset task evaluation dimension, including:

[0031] By calling the intelligent agent to perform quantitative evaluation on each task node using multiple preset task evaluation dimensions, the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension are obtained.

[0032] Receive the user's correction of the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, and obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0033] In one alternative approach, determining the role recommendations corresponding to each task node of the service to be evaluated based on the degree of automation includes:

[0034] When the automation level of the target task node is fully automated, the role corresponding to the target task node is determined to be an intelligent agent role; the target task node is one of the task nodes of the service to be evaluated.

[0035] When the degree of automation of the target task node is within the range of full automation, the role corresponding to the target task node is determined to be an intelligent agent role;

[0036] When the degree of automation of the target task node is within the range of assisted automation level, the role corresponding to the target task node is determined to be a hybrid role; the hybrid role includes human role and intelligent agent role;

[0037] When the automation level of the target task node is within the range of purely human level, the role corresponding to the target task node is determined to be a human role.

[0038] According to another aspect of the present invention, a task intelligent readiness assessment device is provided, comprising:

[0039] A construction module is used to build a task description model; the task description model is obtained by modeling using a graphical business process modeling language; the graphical business process modeling language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications;

[0040] The structured module is used to input the business information of the business to be evaluated into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target deterministic classification corresponding to the business to be evaluated;

[0041] The quantitative evaluation module is used to quantitatively evaluate each task node from multiple preset task evaluation dimensions based on the structured description information, and obtain the quantitative evaluation score of each task node; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension.

[0042] The level determination module is used to determine the degree of automation of each task node based on the quantitative evaluation score of each task node and the automation level determination rules.

[0043] The role suggestion module is used to determine the role suggestions corresponding to each task node of the service to be evaluated based on the degree of automation; wherein, the role suggestions include intelligent agent roles, human roles, or hybrid roles.

[0044] According to another aspect of the present invention, a computer device is provided, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0045] The memory is used to store at least one executable instruction that causes the processor to perform the operation of the intelligent task readiness assessment method.

[0046] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction, which, when executed on a computer device, causes the computer device to perform the operation of the intelligent task readiness assessment method.

[0047] This invention constructs a task description model, inputting business information of the business to be evaluated into the model to obtain structured description information corresponding to each task node in the business to be evaluated. Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. Based on the quantitative evaluation scores and automation level determination rules, the automation degree of each task node is determined; based on the automation degree, role suggestions corresponding to each task node in the business to be evaluated are determined. The task description model is obtained using a graphical business process modeling language; this language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension; role suggestions include intelligent agent role, human role, or hybrid role.

[0048] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0049] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0050] Figure 1 A flowchart illustrating the intelligent task readiness assessment method provided in an embodiment of the present invention is shown.

[0051] Figure 2 This diagram illustrates a preset task evaluation dimension in a task intelligent readiness evaluation method provided by an embodiment of the present invention.

[0052] Figure 3 A schematic diagram of the degree of automation in a task intelligent readiness assessment method provided by an embodiment of the present invention is shown;

[0053] Figure 4 A schematic diagram of the structure of the intelligent task readiness assessment device provided in an embodiment of the present invention is shown;

[0054] Figure 5A schematic diagram of the structure of a computer device provided in an embodiment of the present invention is shown. Detailed Implementation

[0055] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0056] Figure 1 A flowchart of a task intelligence readiness assessment method provided in an embodiment of the present invention is shown. This method is executed by a computer device. The computer device can be a computer, a smart terminal, a distributed device, a robot, a wearable device, etc., and the embodiments of the present invention do not impose specific limitations. Figure 1 As shown, the method includes the following steps:

[0057] Step 110: Construct the task description model.

[0058] The task description model is obtained by using a graphical business process modeling language.

[0059] In one embodiment of the present invention, the task description model is obtained by modeling based on BPMN (Business Process Model and Notation) and using a graphical business process modeling language.

[0060] The specific construction process of the task description model is as follows:

[0061] Step 1101: Establish a graphical business process modeling language.

[0062] The graphical business process modeling language is derived by extending the semantics of BPMN, specifically including task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The input artifact set contains at least one input artifact, each with an artifact identifier, description, and data format attributes. The execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions. The constraint set includes time constraints, access constraints, quality constraints, and audit constraints, each associated with a policy identifier and policy level.

[0063] Specifically, based on the standard BPMN 2.0 semantics, specialized extensions are made to form a graphical business process modeling language suitable for human-machine collaboration and intelligent execution. This unifies and standardizes the task description dimensions and metadata structure, ensuring that the model can be parsed, scheduled, and monitored by computers. The specific extension elements of the graphical business process modeling language include:

[0064] Task Identifier: A globally unique ID used for locating, retrieving, and scheduling tasks.

[0065] Input workpiece set: contains at least one input workpiece. Each workpiece is configured with workpiece identifier, text description, data format, data source, data version, and data validity rule attributes. It supports multiple workpiece types such as structured data, unstructured text, files, and interface parameters.

[0066] Execution logic: includes a sequential sequence of steps, a list of tool references, branch execution conditions, loop rules, exception handling logic, and completion determination definition, clearly defining "how to execute, when to complete, and how to tolerate errors" for the task.

[0067] Output artifact set: Defines the result data, files, status information, etc. produced after the task is completed, and specifies the output format, storage location and push rules.

[0068] Responsible Role: The type of entity that performs the task, limiting the scope of roles that can perform the task.

[0069] The constraint set includes five categories: time constraints, access constraints, quality constraints, compliance constraints, and audit constraints. Each constraint is associated with a system identifier, system source, constraint level, alarm policy, and enforcement rules to ensure that task execution complies with business specifications.

[0070] Cognitive level: The depth of knowledge and judgment level required for the labeling task, used to match the ability level of the large language model agent.

[0071] Deterministic classification: Tasks are divided into deterministic, probabilistic, fuzzy decision-making, and open-ended creative types to distinguish between mechanically executed tasks and flexible tasks that require large models to understand, reason, and generate.

[0072] This modeling language retains the graphical symbols and layout specifications of the BPMN standard, while also supporting the visual editing and storage of the aforementioned extended metadata, balancing readability for business users and system executability.

[0073] Step 1102: Construct a three-layer recursive process model, which includes a process domain layer, an activity layer, and a task layer. Each process domain contains at least one activity, and each activity contains at least one task. This process model can be a BPMN model.

[0074] In this embodiment of the invention, a three-layer recursive process model is constructed based on the aforementioned extended modeling language, enabling hierarchical decomposition of business processes from macro to micro levels, making complex processes decomposable, manageable, and executable. This process model is compatible with the standard BPMN model structure and supports recursive nesting, making it suitable for large-scale, multi-level, and cross-departmental business processes.

[0075] The three-layer structure is specifically defined as follows:

[0076] Process domain layer: This is the highest level and corresponds to a complete business domain or end-to-end business process, such as customer account opening process, transaction risk control process, and operation and maintenance emergency process. A process domain can contain one or more interrelated activities.

[0077] Activity layer: The middle level, corresponding to the independent business links in the process. An activity consists of at least one task, and multiple activities can be combined in sequence, parallel, branching, looping and other logical combinations.

[0078] Task layer: The finest-grained execution unit, corresponding to the smallest indivisible execution action, is the basic unit for direct scheduling, execution and monitoring of the model.

[0079] The three layers satisfy a recursive relationship: process domains can nest subprocess domains, activities can nest subactivities, and tasks are atomic nodes that cannot be further divided, ensuring the model's universal adaptability to both simple and complex processes.

[0080] Step 1103: Define the role executor for the process model.

[0081] In this embodiment of the invention, the role executor includes a human role, a large language model intelligent agent role, a system role, and a hybrid role; wherein, the large language model intelligent agent role is associated with intelligent agent specification parameters, and the hybrid role is configured with both intelligent agent specification parameters and human ability descriptions.

[0082] Specifically, the character types and configuration rules are as follows:

[0083] Human roles: These are performed by natural persons such as business personnel, administrators, and operators, and are bound to organizational structure, job permissions, and operational qualifications.

[0084] Large Language Model Agent Role: An AI agent driven by a large model, associated with agent specification parameters, including model version, context length, call temperature, capability tags, call limits, and security policies.

[0085] System roles: executed by pure software systems such as business systems, interface services, scheduled tasks, and automated scripts, requiring no manual or AI intervention, and supporting automatic triggering and completion;

[0086] Hybrid role: It is executed collaboratively by humans and large language model intelligent agents or by the system and large language model intelligent agents. It also configures intelligent agent specification parameters and human ability descriptions, and supports collaborative modes of AI assisting human and system calling AI.

[0087] This invention enables standardized management of task execution entities through role definition, supporting the flexible undertaking of the same task by different roles in different scenarios.

[0088] Step 1104: Formalize each task in the process model according to the graphical business process modeling language to obtain the task description model.

[0089] Using the extended graphical business process modeling language of step 1101 as the syntax standard, each atomic task in the three-layer process model of step 1102 is given a complete, unified, and parsable formal description, which ultimately forms a task description model that can be directly used for process execution and scheduling.

[0090] The specific description process of the task description model includes:

[0091] Assign a unique task identifier to each task and bind it to its process domain, activity, and hierarchical path;

[0092] Enter the input and output work sets for the task, and clarify the rules for data inflow and outflow;

[0093] Fill in the execution logic, including the sequence of steps, tool list, branch conditions, and completion definition;

[0094] Bind responsibility roles, specifying humans, intelligent agents, systems, or hybrid actors;

[0095] Load the constraint set and configure time, permission, quality, compliance, and audit rules;

[0096] The cognitive level and deterministic classification are labeled for matching agent capabilities and selecting execution strategies.

[0097] The above information is then solidified in a structured format to form a storable, readable, and schedulable task description model. The structured format can be XML, JSON, BPMN extension files, etc.

[0098] The task description model constructed through this step retains the advantages of BPMN's graphical and easy-to-understand nature, while also possessing the technical capabilities to support large-scale intelligent agents, multi-role collaboration, strong constraint control, and full-process auditability, providing core model support for subsequent process automation and intelligent execution.

[0099] In this embodiment of the invention, the modeling process of the task description model is based on the BPMN model and is constructed using the extended language described above. The process is largely the same as the modeling process of the BPMN model.

[0100] Step 120: Input the business information of the business to be evaluated into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated.

[0101] The business information includes business requirement documents, requirement review management systems, requirement development process standards, code review and quality standards, and information security and compliance management regulations. Specifically, this business information serves as a comprehensive basis for the business to be evaluated, including: Business requirement documents: specifying business objectives, business scenarios, functional requirements, process steps, and expected results; Requirement review management systems: specifying management rules and process requirements for requirement proposal, review, modification, and confirmation; Requirement development process standards: specifying requirement breakdown, development specifications, phase division, and deliverable requirements; Code review and quality standards: specifying code inspection, testing specifications, quality access, and acceptance criteria; Information security and compliance management regulations: specifying data security, access control, operation logging, compliance red lines, and audit requirements. This business information is input in the form of text, policy clauses, process specifications, and standard documents, and the task description model automatically maps and fills in unstructured information into structured information.

[0102] The structured description information includes at least the task name, input / output data description, processing logic description, execution role, and applicable institutional constraints. In an optional embodiment, the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated.

[0103] In this embodiment of the invention, after obtaining the task description model, the task description of the business to be evaluated can be instantiated using the task description model based on the business information, thereby obtaining structured description information corresponding to each task node in the business to be evaluated. Specifically, this includes:

[0104] Step 1201: Analyze and extract elements from the business information.

[0105] The process involves unified parsing of multi-source business information to extract core elements relevant to task execution: task name, business scenario, processing logic, and input / output data from business requirement documents; execution sequence, stage requirements, and completion conditions from requirement reviews and development process standards; quality constraints, acceptance rules, and verification logic from code reviews and quality standards; and access constraints, time constraints, audit constraints, and compliance clauses from information security and compliance management regulations. The extraction process supports text matching, rule mapping, and keyword positioning to ensure that business information is fully incorporated into the metadata structure defined in the model.

[0106] Step 1202: Based on the extracted elements, perform instantiation matching based on the task description model.

[0107] Specifically, using the BPMN extended task description model constructed in step 110 as a template, the extracted business elements are instantiated and populated according to the eight standard structures defined in the model, forming task instances that correspond one-to-one with the business to be evaluated: the task identifier in the model is instantiated as the target task identifier, which is globally unique and bound to the business number; the input artifact set in the model is instantiated as the target input artifact set, specifying the data format, source, and validity; the execution logic in the model is instantiated as the target execution logic, including the step sequence, tools, branch conditions, and completion definition; the output artifact set in the model is instantiated as the target output artifact set, specifying the result format, storage, and push rules; the responsibility role in the model is instantiated as the target responsibility role, specifying human, large model intelligent agent, system, or hybrid role; the constraint set in the model is instantiated as the target constraint set, associating specific institutional clauses and constraint levels; the cognitive level in the model is instantiated as the target cognitive level, matching the knowledge depth required for the task; and the deterministic classification in the model is instantiated as the target deterministic classification, distinguishing between mechanical execution and fuzzy decision-making tasks.

[0108] Step 1203: Generate structured description information for task nodes.

[0109] After instantiation, complete, standardized, and directly usable structured description information is generated for each task node in the business process to be evaluated. The structured description information includes at least: task name and task identifier, input and output data description (input artifacts, output artifacts), processing logic description (execution steps, conditions, completion rules), execution role (human, intelligent agent, system, or hybrid), and institutional constraints (time, permissions, quality, compliance, audit).

[0110] In a preferred embodiment of the present invention, the structured description information fully includes the following eight standardized contents: target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target deterministic classification.

[0111] Step 1204: Perform structured information verification and output to obtain the structured description information corresponding to each task node in the service to be evaluated.

[0112] Specifically, this includes: verifying the completeness and compliance of the generated structured description information: checking whether the required fields are fully filled; checking whether the execution roles, constraints and business systems are consistent; checking whether the input and output artifact formats conform to the model definition; and outputting the information in the form of an instance file after the verification is passed, for use by the subsequent evaluation module.

[0113] Step 130: Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation score of each task node.

[0114] Among them, such as Figure 2 As shown, the multiple preset task evaluation dimensions include cognitive complexity, data dependence, interaction diversity, compliance sensitivity, and innovation needs. The evaluation scores for different data in each dimension can be pre-classified based on expert experience, then scored by an intelligent agent, and finally adjusted by experts to obtain the final score.

[0115] Specifically, such as Figure 3 As shown, the cognitive complexity dimension includes the following levels: memory and retrieval level, understanding and interpretation level, application and preliminary analysis level, deep analysis and synthesis level, and evaluation and creation level. This invention defines a non-linear compression mapping rule for the cognitive complexity dimension from six cognitive levels to a five-point rating. Cognitive educational psychology divides cognitive activities into six levels from low to high: memory, understanding, application, analysis, evaluation, and creation. This invention compresses and maps these six levels into five rating levels: 1 point = memory and retrieval level, such as querying configuration item status; 2 points = understanding and interpretation level, such as interpreting requirements documents; 3 points = application and preliminary analysis level (such as code conflict detection); 4 points = deep analysis and synthesis level, such as change impact analysis; 5 points = evaluation and creation level, such as architecture design decisions. There is an overlap between the application and analysis levels (3-4 points), which needs to be determined based on the specific characteristics of the task: if the task only requires application within a predetermined framework, it is 3 points; if it requires active decomposition and synthesis of multiple information sources, it is 4 points.

[0116] The data dependency dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams. Specifically, 1 point = structured single-source data, such as single-table queries; 2 points = structured multi-source data, such as cross-table joins; 3 points = semi-structured multi-source integration, such as API and document extraction; 4 points = unstructured multi-source integration, which can be emails, minutes, and code comments; and 5 points = unstructured real-time streams, such as real-time monitoring of multiple information sources and dynamic decision-making.

[0117] The diversity of interactions includes the following levels: no interaction or pure API call, one-way notification or push notification, form and tool interaction, multi-party communication and coordination, and multi-party negotiation. In one example, 1 point = no interaction or pure API call; 2 points = one-way notification or push notification; 3 points = form and tool interaction; 4 points = multi-party communication and coordination; 5 points = multi-party negotiation, such as mediation of conflicts of interest.

[0118] The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory legal and regulatory constraints. 1 point = no compliance requirements, 2 points = industry practice constraints, 3 points = internal organizational system constraints, 4 points = industry regulatory requirements, and 5 points = mandatory legal and regulatory constraints. This dimension is associated with the three-tiered classification of regulations, with different scores corresponding to different types of regulations. The default weight W4 = 1.5, higher than the 1.0 of other dimensions. The rationale for this design is that compliance constraints often have a veto power in actual business operations. That is, even if the other four dimensions are at their lowest scores, if D4 = 5 (mandatory regulations), the total score will be raised to (1+1+1+7.5+1) / 5.5≈2.27, exceeding the L1 threshold, ensuring that compliance-sensitive tasks are not classified as fully automated.

[0119] The innovation requirements dimensions include the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making. Specifically, 1 point = standard process execution, 2 points = parameterized adjustment, 3 points = limited flexibility (e.g., flexible responses within a rule framework), 4 points = solution selection (e.g., weighing multiple solutions), and 5 points = open-ended creativity and strategic decision-making.

[0120] Specifically, the steps include the following:

[0121] Step 1301: Based on the structured description information, quantitatively evaluate each task node from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension. In one embodiment of the present invention, by calling an agent to quantitatively evaluate each task node using multiple preset task evaluation dimensions, the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension are obtained. During agent evaluation, multiple scoring agents can independently score the tasks, with at least two scoring agents scoring independently. When the difference between two scores in any dimension exceeds 2 points, it is marked as "disagreement pending" and negotiation is triggered. After negotiation, a consensus value is taken; otherwise, the average value is taken. The system receives corrections from users regarding the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, thus obtaining the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension. The user can be an expert user.

[0122] Step 1302: Based on the weights corresponding to each preset task evaluation dimension, perform a weighted average calculation on the quantitative evaluation sub-scores of each preset task evaluation dimension to obtain the quantitative evaluation score of each task node.

[0123] In this embodiment of the invention, the weights corresponding to each preset task evaluation dimension are set according to the specific actual scenario. In one example, the default weight vector can be represented as W=[1.0,1.0,1.0,1.5,1.0], with a sum of weights of 5.5. Weights can be customized at the organizational level: financial institutions can increase D4 to 2.0, and innovative enterprises can increase D5 to 1.5. The weighted average formula is as follows: .in, This represents the evaluation dimension of the i-th preset task. This represents the weight corresponding to the i-th preset task evaluation dimension. This indicates the number of preset task evaluation dimensions.

[0124] For example, in a computational instance, such as in the field of software engineering: for a code conflict detection task, if D1=2, D2=2, D3=1, D4=1, D5=1, the total score is (2+2+1+1.5+1) / 5.5=7.5 / 5.5≈1.36, which belongs to the L1 level, which is the fully automated level.

[0125] For example, in another calculation instance, such as in the field of software engineering: change impact analysis task, D1=4, D2=3, D3=3, D4=3, D5=3, total score=(4+3+3+4.5+3) / 5.5≈3.18, which belongs to level L3, which is human-dominated level.

[0126] For example, in another calculation instance, such as in the financial field: for a compliance review report task, D1=5, D2=4, D3=4, D4=5, D5=4, the total score = (5+4+4+7.5+4) / 5.5≈4.45, which belongs to L4, the level of pure human. It is important to note here that D4=5, with a weight of 1.5, contributed 7.5 points, accounting for 30.6% of the weighted total score, reflecting the veto effect of the compliance dimension.

[0127] Step 140: Determine the degree of automation of each task node based on the quantitative evaluation score of each task node and the automation level determination rules.

[0128] The automation level determination rules include multi-level rules. In one embodiment, four levels of rules are set: L1 Fully Automation (totalScore ≤ 2.0), which suggests an AI agent role; L2 Assisted Automation (2.0-3.0), which suggests a hybrid role; L3 Human-Dominated (3.0-4.0), which suggests a human role; and L4 Purely Human (greater than 4.0), which requires a human role. The default thresholds of 2.0 / 3.0 / 4.0 are divided at equal intervals. Empirical verification on 25 tasks shows that this threshold division has a 92% consistency with the human judgment of domain experts. The thresholds can be customized by organizations.

[0129] In one implementation of a financial risk control approach, the transaction monitoring task is evaluated as follows: D1=4 (comprehensive analysis of multi-dimensional transaction characteristics), D2=5 (unstructured real-time transaction flow), D3=3 (invoking rule engines and manual review tools), D4=5 (subject to mandatory regulatory constraints), D5=3 (flexible judgment within the rule framework). The total score is (4+5+3+7.5+3) / 5.5≈4.09, which is considered to fall into Level 4, the purely human level. With equal weighting, the total score of (4+5+3+5+3) / 5=4.0 is only at the L3 / L4 boundary. However, with compliance weighting, the total score rises to 4.09, crossing into L4, demonstrating the veto effect of compliance.

[0130] Step 150: Based on the degree of automation, determine the role recommendations corresponding to each task node of the business to be evaluated.

[0131] The role suggestions in this embodiment of the invention include intelligent agent roles, human roles, or hybrid roles.

[0132] Specifically, when the automation level of the target task node is fully automated, the role corresponding to the target task node is determined to be an intelligent agent role; the target task node is one of the task nodes of the service to be evaluated.

[0133] When the degree of automation of the target task node is within the range of full automation, the role corresponding to the target task node is determined to be an intelligent agent role.

[0134] When the degree of automation of the target task node is within the range of assisted automation level, the role corresponding to the target task node is determined to be a hybrid role; the hybrid role includes human role and intelligent agent role.

[0135] When the automation level of the target task node is within the range of purely human level, the role corresponding to the target task node is determined to be a human role.

[0136] This invention constructs a task description model, inputting business information of the business to be evaluated into the model to obtain structured description information corresponding to each task node in the business to be evaluated. Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. Based on the quantitative evaluation scores and automation level determination rules, the automation degree of each task node is determined; based on the automation degree, role suggestions corresponding to each task node in the business to be evaluated are determined. The task description model is obtained using a graphical business process modeling language; this language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension; role suggestions include intelligent agent role, human role, or hybrid role.

[0137] Figure 4 A schematic diagram of the structure of the intelligent task readiness assessment device provided in an embodiment of the present invention is shown. Figure 4 As shown, the device 300 includes:

[0138] Module 310 is used to construct a task description model; the task description model is obtained by modeling using a graphical business process modeling language. The graphical business process modeling language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications.

[0139] The structured module 320 is used to input the business information of the business to be evaluated into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target deterministic classification corresponding to the business to be evaluated.

[0140] The quantitative evaluation module 330 is used to quantitatively evaluate each task node from multiple preset task evaluation dimensions based on the structured description information, and obtain a quantitative evaluation score for each task node. The multiple preset task evaluation dimensions include cognitive complexity, data dependence, interaction diversity, compliance sensitivity, and innovation demand.

[0141] The level determination module 340 is used to determine the degree of automation of each task node based on the quantitative evaluation score of each task node and the automation level determination rules.

[0142] The role suggestion module 350 is used to determine the role suggestions corresponding to each task node of the business to be evaluated based on the degree of automation; wherein, the role suggestions include intelligent agent roles, human roles, or hybrid roles.

[0143] The specific working process of each module of the device 300 in this embodiment of the invention is generally consistent with the specific method steps of the aforementioned method embodiment, and will not be repeated here.

[0144] This invention constructs a task description model, inputting business information of the business to be evaluated into the model to obtain structured description information corresponding to each task node in the business to be evaluated. Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. Based on the quantitative evaluation scores and automation level determination rules, the automation degree of each task node is determined; based on the automation degree, role suggestions corresponding to each task node in the business to be evaluated are determined. The task description model is obtained using a graphical business process modeling language; this language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension; role suggestions include intelligent agent role, human role, or hybrid role.

[0145] Figure 5 The diagram shows a structural schematic of a computer device provided in an embodiment of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the computer device.

[0146] like Figure 5 As shown, the computer device may include: a processor 402, a communications interface 404, a memory 406, and a communications bus 408.

[0147] The processor 402, communication interface 404, and memory 406 communicate with each other via communication bus 408. Communication interface 404 is used to communicate with other network elements such as clients or other servers. The processor 402 executes program 410, specifically performing the relevant steps described in the embodiment of the task intelligent readiness assessment method.

[0148] Specifically, program 410 may include program code, which includes computer-executable instructions.

[0149] Processor 402 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The computer device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0150] Memory 406 is used to store program 410. Memory 406 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0151] Specifically, program 410 can be called by processor 402 to cause the computer device to perform the following operations:

[0152] A task description model is constructed; the task description model is obtained by modeling using a graphical business process modeling language; the graphical business process modeling language includes task identifier, input artifact set, execution logic, output artifact set, responsibility role, constraint set, cognitive level, and deterministic classification;

[0153] The business information of the business to be evaluated is input into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level and target determinism classification corresponding to the business to be evaluated;

[0154] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension.

[0155] The degree of automation of each task node is determined based on the quantitative evaluation score of each task node and the automation level determination rules.

[0156] Based on the degree of automation, role recommendations are determined for each task node of the service to be evaluated; wherein, the role recommendations include intelligent agent roles, human roles, or hybrid roles.

[0157] In one alternative approach, constructing the task description model includes:

[0158] A graphical business process modeling language is established. This language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The input artifact set contains at least one input artifact, each with an artifact identifier, description, and data format attributes. The execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions. The constraint set includes time constraints, access constraints, quality constraints, and audit constraints, with each constraint associated with a policy identifier and policy level.

[0159] A three-layer recursive process model is constructed, comprising a process domain layer, an activity layer, and a task layer; wherein each process domain contains at least one activity, and each activity contains at least one task.

[0160] Define role executors for the process model; the role executors include human roles, large language model intelligent agent roles, system roles, and hybrid roles; wherein, the large language model intelligent agent roles are associated with intelligent agent specification parameters, and the hybrid roles are configured with both intelligent agent specification parameters and human ability descriptions;

[0161] Each task in the process model is formally described using the graphical business process modeling language to obtain the task description model.

[0162] In one optional approach, the business information includes business requirement documents, requirement review management systems, requirement development process standards, code review and quality standards, and information security and compliance management regulations; the step of inputting the business information of the business to be evaluated into the task description model to obtain structured description information corresponding to each task node in the business to be evaluated includes:

[0163] Based on the business information, the task description model is used to instantiate the business to be evaluated, thereby obtaining the structured description information corresponding to each task node in the business to be evaluated.

[0164] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node, including:

[0165] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0166] Based on the weights corresponding to each preset task evaluation dimension, the quantitative evaluation sub-scores of each preset task evaluation dimension are calculated by weighted average to obtain the quantitative evaluation score of each task node.

[0167] In one alternative approach, the cognitive complexity dimension includes the following levels: memory and retrieval level, understanding and interpretation level, application and preliminary analysis level, in-depth analysis and synthesis level, and evaluation and creation level;

[0168] The data dependency dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams.

[0169] The interaction diversity dimension includes the following levels: no interaction or pure API call, one-way notification or message push, form and tool interaction, multi-party communication and coordination and multi-party negotiation.

[0170] The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory laws and regulations.

[0171] The innovation requirements include the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making.

[0172] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation sub-score for each task node in each preset task evaluation dimension, including:

[0173] By calling the intelligent agent to perform quantitative evaluation on each task node using multiple preset task evaluation dimensions, the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension are obtained.

[0174] Receive the user's correction of the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, and obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0175] In one alternative approach, determining the role recommendations corresponding to each task node of the service to be evaluated based on the degree of automation includes:

[0176] When the automation level of the target task node is fully automated, the role corresponding to the target task node is determined to be an intelligent agent role; the target task node is one of the task nodes of the service to be evaluated.

[0177] When the degree of automation of the target task node is within the range of full automation, the role corresponding to the target task node is determined to be an intelligent agent role;

[0178] When the degree of automation of the target task node is within the range of assisted automation level, the role corresponding to the target task node is determined to be a hybrid role; the hybrid role includes human role and intelligent agent role;

[0179] When the automation level of the target task node is within the range of purely human level, the role corresponding to the target task node is determined to be a human role.

[0180] This invention constructs a task description model, inputting business information of the business to be evaluated into the model to obtain structured description information corresponding to each task node in the business to be evaluated. Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. Based on the quantitative evaluation scores and automation level determination rules, the automation degree of each task node is determined; based on the automation degree, role suggestions corresponding to each task node in the business to be evaluated are determined. The task description model is obtained using a graphical business process modeling language; this language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension; role suggestions include intelligent agent role, human role, or hybrid role.

[0181] This invention provides a computer-readable storage medium storing at least one executable instruction that, when executed on a computer device, causes the computer device to perform the task intelligent readiness assessment method described in any of the above method embodiments.

[0182] Executable instructions can be used to cause computer devices to perform the following operations:

[0183] A task description model is constructed; the task description model is obtained by modeling using a graphical business process modeling language; the graphical business process modeling language includes task identifier, input artifact set, execution logic, output artifact set, responsibility role, constraint set, cognitive level, and deterministic classification;

[0184] The business information of the business to be evaluated is input into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level and target determinism classification corresponding to the business to be evaluated;

[0185] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension.

[0186] The degree of automation of each task node is determined based on the quantitative evaluation score of each task node and the automation level determination rules.

[0187] Based on the degree of automation, role suggestions are determined for each task node of the service to be evaluated; wherein, the role suggestions include intelligent agent roles, human roles, or hybrid roles.

[0188] In one alternative approach, constructing the task description model includes:

[0189] A graphical business process modeling language is established. This language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The input artifact set contains at least one input artifact, each with an artifact identifier, description, and data format attributes. The execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions. The constraint set includes time constraints, access constraints, quality constraints, and audit constraints, with each constraint associated with a policy identifier and policy level.

[0190] A three-layer recursive process model is constructed, comprising a process domain layer, an activity layer, and a task layer; wherein each process domain contains at least one activity, and each activity contains at least one task.

[0191] Define role executors for the process model; the role executors include human roles, large language model intelligent agent roles, system roles, and hybrid roles; wherein, the large language model intelligent agent roles are associated with intelligent agent specification parameters, and the hybrid roles are configured with both intelligent agent specification parameters and human ability descriptions;

[0192] Each task in the process model is formally described using the graphical business process modeling language to obtain the task description model.

[0193] In one optional approach, the business information includes business requirement documents, requirement review management systems, requirement development process standards, code review and quality standards, and information security and compliance management regulations; the step of inputting the business information of the business to be evaluated into the task description model to obtain structured description information corresponding to each task node in the business to be evaluated includes:

[0194] Based on the business information, the task description model is used to instantiate the business to be evaluated, thereby obtaining the structured description information corresponding to each task node in the business to be evaluated.

[0195] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node, including:

[0196] Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0197] Based on the weights corresponding to each preset task evaluation dimension, the quantitative evaluation sub-scores of each preset task evaluation dimension are calculated by weighted average to obtain the quantitative evaluation score of each task node.

[0198] In one alternative approach, the cognitive complexity dimension includes the following levels: memory and retrieval level, understanding and interpretation level, application and preliminary analysis level, in-depth analysis and synthesis level, and evaluation and creation level;

[0199] The data dependency dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams.

[0200] The interaction diversity dimension includes the following levels: no interaction or pure API call, one-way notification or message push, form and tool interaction, multi-party communication and coordination and multi-party negotiation.

[0201] The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory laws and regulations.

[0202] The innovation requirements include the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making.

[0203] In one optional approach, based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation sub-score for each task node in each preset task evaluation dimension, including:

[0204] By calling the intelligent agent to perform quantitative evaluation on each task node using multiple preset task evaluation dimensions, the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension are obtained.

[0205] Receive the user's correction of the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, and obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

[0206] In one alternative approach, determining the role recommendations corresponding to each task node of the service to be evaluated based on the degree of automation includes:

[0207] When the automation level of the target task node is fully automated, the role corresponding to the target task node is determined to be an intelligent agent role; the target task node is one of the task nodes of the service to be evaluated.

[0208] When the degree of automation of the target task node is within the range of full automation, the role corresponding to the target task node is determined to be an intelligent agent role;

[0209] When the degree of automation of the target task node is within the range of assisted automation level, the role corresponding to the target task node is determined to be a hybrid role; the hybrid role includes human role and intelligent agent role;

[0210] When the automation level of the target task node is within the range of purely human level, the role corresponding to the target task node is determined to be a human role.

[0211] This invention constructs a task description model, inputting business information of the business to be evaluated into the model to obtain structured description information corresponding to each task node in the business to be evaluated. Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. Based on the quantitative evaluation scores and automation level determination rules, the automation degree of each task node is determined; based on the automation degree, role suggestions corresponding to each task node in the business to be evaluated are determined. The task description model is obtained using a graphical business process modeling language; this language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target certainty classification corresponding to the business to be evaluated; the multiple preset task evaluation dimensions include cognitive complexity dimension, data dependence dimension, interaction diversity dimension, compliance sensitivity dimension, and innovation demand dimension; role suggestions include intelligent agent role, human role, or hybrid role.

[0212] This invention provides a task intelligent readiness assessment device for performing the above-described task intelligent readiness assessment method.

[0213] This invention provides a computer program that can be called by a processor to enable a computer device to execute the task intelligent readiness assessment method in any of the above method embodiments.

[0214] This invention provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions, which, when executed on a computer, cause the computer to perform the task intelligent readiness assessment method described in any of the above method embodiments.

[0215] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0216] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0217] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim.

[0218] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0219] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for assessing the intelligent readiness of a task, characterized in that, The method includes: Constructing a task description model includes: establishing a graphical business process modeling language; constructing a three-level recursive process model; defining executor roles for the process model; and formally describing each task in the process model using the graphical business process modeling language to obtain the task description model. The task description model is obtained by modeling using the graphical business process modeling language. The graphical business process modeling language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications. The input artifact set contains at least one input artifact, and each artifact has an artifact identifier, description, and data. The format attributes include: the execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions; the constraint set includes time constraints, permission constraints, quality constraints, and audit constraints, with each constraint associated with a policy identifier and policy level; the process model includes a process domain layer, an activity layer, and a task layer; each process domain contains at least one activity, and each activity contains at least one task; the executor roles include human roles, large language model intelligent agent roles, and hybrid roles; the large language model intelligent agent role is associated with intelligent agent specification parameters, and the hybrid role is configured with both intelligent agent specification parameters and human ability descriptions. The business information of the business to be evaluated is input into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level and target determinism classification corresponding to the business to be evaluated; Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node. These preset task evaluation dimensions include cognitive complexity, data dependency, interaction diversity, compliance sensitivity, and innovation demand. The cognitive complexity dimension includes the following levels: memory and retrieval, understanding and interpretation, application and preliminary analysis, in-depth analysis and synthesis, and evaluation and creation. The data dependency dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams. The interaction diversity dimension includes the following levels: no interaction or pure API calls, one-way notifications or message pushes, form and tool interactions, multi-party communication and coordination, and multi-party negotiation. The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory legal and regulatory constraints. The innovation demand dimension includes the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making. The degree of automation of each task node is determined based on the quantitative evaluation score of each task node and the automation level determination rules. Based on the degree of automation, role recommendations are determined for each task node of the service to be evaluated; wherein, the role recommendations include intelligent agent roles, human roles, or hybrid roles.

2. The method according to claim 1, characterized in that, The business information includes business requirement documents, requirement review management system, requirement development process standards, code review and quality standards, and information security and compliance management regulations. The step of inputting the business information of the business to be evaluated into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated includes: Based on the business information, the task description model is used to instantiate the business to be evaluated, thereby obtaining the structured description information corresponding to each task node in the business to be evaluated.

3. The method according to any one of claims 1-2, characterized in that, Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain a quantitative evaluation score for each task node, including: Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension. Based on the weights corresponding to each preset task evaluation dimension, the quantitative evaluation sub-scores of each preset task evaluation dimension are calculated by weighted average to obtain the quantitative evaluation score of each task node.

4. The method according to claim 1, characterized in that, Based on the structured description information, each task node is quantitatively evaluated from multiple preset task evaluation dimensions to obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, including: By calling the intelligent agent to perform quantitative evaluation on each task node using multiple preset task evaluation dimensions, the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension are obtained. Receive the user's correction of the estimated quantitative evaluation sub-scores of each task node in each preset task evaluation dimension, and obtain the quantitative evaluation sub-scores of each task node in each preset task evaluation dimension.

5. The method according to any one of claims 1-2, characterized in that, The step of determining the role recommendations corresponding to each task node of the business to be evaluated based on the degree of automation includes: When the automation level of the target task node is fully automated, the role corresponding to the target task node is determined to be an intelligent agent role; the target task node is one of the task nodes of the service to be evaluated. When the degree of automation of the target task node is within the range of full automation, the role corresponding to the target task node is determined to be an intelligent agent role; When the degree of automation of the target task node is within the range of assisted automation level, the role corresponding to the target task node is determined to be a hybrid role; the hybrid role includes human role and intelligent agent role; When the automation level of the target task node is within the range of purely human level, the role corresponding to the target task node is determined to be a human role.

6. A task intelligent readiness assessment device, characterized in that, The device includes: The construction module is used to build a task description model, including: establishing a graphical business process modeling language; constructing a three-level recursive process model; defining role executors for the process model; formally describing each task in the process model according to the graphical business process modeling language to obtain the task description model; the task description model is obtained by modeling using the graphical business process modeling language; the graphical business process modeling language includes task identifiers, input artifact sets, execution logic, output artifact sets, responsibility roles, constraint sets, cognitive levels, and deterministic classifications; wherein, the input artifact set contains at least one input artifact, and each artifact has an artifact identifier, description, and other characteristics. The description and data format attributes; the execution logic includes a sequentially arranged sequence of steps, a tool reference list, execution conditions, and completion definitions; the constraint set includes time constraints, permission constraints, quality constraints, and audit constraints, each constraint being associated with a system identifier and system level; the process model includes a process domain layer, an activity layer, and a task layer; wherein each process domain contains at least one activity, and each activity contains at least one task; wherein the executor roles include human roles, large language model intelligent agent roles, and hybrid roles; the large language model intelligent agent roles are associated with intelligent agent specification parameters, and the hybrid roles are configured with both intelligent agent specification parameters and human ability descriptions; The structured module is used to input the business information of the business to be evaluated into the task description model to obtain the structured description information corresponding to each task node in the business to be evaluated; the structured description information includes the target task identifier, target input artifact set, target execution logic, target output artifact set, target responsibility role, target constraint set, target cognitive level, and target deterministic classification corresponding to the business to be evaluated; The quantitative evaluation module is used to quantitatively evaluate each task node from multiple preset task evaluation dimensions based on the structured description information, and obtain the quantitative evaluation score of each task node. The multiple preset task evaluation dimensions include cognitive complexity, data dependence, interaction diversity, compliance sensitivity, and innovation demand. The cognitive complexity dimension includes the following levels: memory and retrieval level, understanding and interpretation level, application and preliminary analysis level, in-depth analysis and synthesis level, and evaluation and creation level. The data dependence dimension includes structured single-source data, structured multi-source data, semi-structured multi-source integration, unstructured multi-source integration, and unstructured real-time streams. The interaction diversity dimension includes the following levels: no interaction or pure API call, one-way notification or message push, form and tool interaction, multi-party communication and coordination, and multi-party negotiation. The compliance sensitivity dimension includes the following levels: no compliance requirements, industry practice constraints, internal organizational system constraints, industry regulatory requirements, and mandatory legal and regulatory constraints. The innovation demand dimension includes the following levels: standard process execution, parameterized adjustment, limited flexibility, solution selection, open-ended creativity, and strategic decision-making. The level determination module is used to determine the degree of automation of each task node based on the quantitative evaluation score of each task node and the automation level determination rules. The role suggestion module is used to determine the role suggestions corresponding to each task node of the service to be evaluated based on the degree of automation; wherein, the role suggestions include intelligent agent roles, human roles, or hybrid roles.

7. A computer device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation of the task intelligent readiness assessment method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on a computer device, causes the computer device to perform the operation of the task intelligent readiness assessment method as described in any one of claims 1-5.

Images