Task allocation method and device based on intelligent agent, equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence and human resource management, and discloses an agent-based task allocation method, device and equipment and a storage medium, a post agent reflecting the dependency relationship of each post in an organization is constructed, tasks can be comprehensively allocated from the dependency relationship of each post when the post agent allocates the tasks, and the task allocation efficiency is improved. Therefore, the task cannot be executed due to the conflict after the task is allocated, and the matching degree between the task and the post is improved.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and human resource management technology, and in particular to a task allocation method, apparatus, device and storage medium based on intelligent agents. Background Technology

[0002] With the development of artificial intelligence and distributed systems, intelligent agents have become one way to automate task allocation. In current task management systems, an intelligent agent is first configured based on task parameters, creating a set of intelligent agents adaptable to multiple task allocations. Then, the tasks to be allocated are input into the intelligent agent set for analysis and matching, directly outputting the corresponding executor. While existing intelligent agents can identify and allocate tasks, this is only from the perspective of an individual agent. When dealing with complex task allocations involving overlapping roles and permissions, they cannot analyze execution conflicts, leading to inaccurate allocation. Summary of the Invention

[0003] This invention provides a task allocation method, apparatus, device, and storage medium based on intelligent agents, aiming to solve the problems of allocation conflicts and low accuracy in existing task allocation schemes when assigning tasks with overlapping roles.

[0004] The first aspect of this invention provides a task allocation method based on intelligent agents, applied to a multi-position organizational management system, the method comprising: The historical tasks of each position in the target organization are obtained, and a three-dimensional organizational relationship model reflecting the dependencies between positions is analyzed based on the organizational structure of the target organization to construct a multi-position intelligent agent. Determine the assigned target tasks and analyze the information of multiple related target positions for the target tasks; The target task and multiple target job information are input into multiple job intelligent agents, and the assignment information of each target job is generated based on the corresponding three-dimensional organizational relationship model in the intelligent agent; The target task is assigned based on the aforementioned delegation information.

[0005] In one feasible implementation, the step of acquiring the historical tasks of each position in the target organization, analyzing the organizational structure of the target organization to generate a three-dimensional organizational relationship model reflecting the dependencies between positions, and constructing a multi-position intelligent agent includes: Obtain the historical tasks of each position in the target organization, and extract the permission information, coordination relationships and control relationships between positions based on the historical tasks of each position. Using the authority information of each position, the coordination and control relationships between positions as constraints, a three-dimensional organizational relationship model corresponding to all positions in the target organization is constructed based on the organizational structure of the target organization. Based on the permission information of each position and the historical tasks of each position, an initial intelligent agent for each position is abstracted. The model parameters of all the initial agents are initialized using the organizational three-dimensional relationship model to obtain the target agents for each position.

[0006] In one feasible implementation, the step of constructing a three-dimensional organizational relationship model corresponding to all positions in the target organization based on the organizational structure of the target organization, using the permission information of each position, the coordination relationship between positions, and the control relationship as constraints, includes: Based on the organizational structure of the target organization, a formal definition is performed to construct a three-dimensional job relationship network of the target organization, wherein the three-dimensional job relationship network (OS) is as follows:

[0007] A set of job roles; A collection of intelligent agents; Indicates permission information; Indicates a coordinating relationship; Indicates a control relationship; This represents the mapping relationship between job positions and intelligent agents; Based on the permission information of each position, the coordination and control relationships between positions, and the control information extracted from the historical tasks of each position, the parameters in the three-dimensional relationship network of the positions are adjusted to obtain the three-dimensional relationship type of the organization corresponding to all positions in the target organization.

[0008] In one feasible implementation, the step of initializing the model parameters of all the initial agents using the organizational three-dimensional relationship model to obtain the target agents for each position includes: Based on the authority information, coordination relationships, and control relationships in the organizational three-dimensional relationship model, the organizational awareness triplet for each position is derived. The organizational awareness triplet includes: goal O, norm N, and information domain I. Goal O is derived from the power relationships in the organizational three-dimensional relationship model. Norm N is obtained by mapping the coordination relationships in the organizational three-dimensional relationship model. Information domain I is determined by filtering the control relationships in the organizational three-dimensional relationship model. Initialize the model parameters of the initial agents for each position, bind each initial agent to the position, and load the organizational consciousness triple to obtain the target agents for each position.

[0009] In one feasible implementation, after acquiring the historical tasks of each position in the target organization, analyzing the organizational structure of the target organization to generate a three-dimensional organizational relationship model reflecting the dependencies between positions, and constructing a multi-position intelligent agent, the method further includes: Based on the responsibilities and permissions of each position, the communication behaviors between the agents corresponding to each position are determined through semantic analysis. The communication behaviors include task delegation, status notification, and execution supervision. Based on the aforementioned communication behavior, the coordination and control relationships among the intelligent agents in each role are adjusted.

[0010] In one feasible implementation, the step of inputting the target task and multiple target job information into multiple job-specific intelligent agents, and generating assignment information for each target job based on the corresponding organizational three-dimensional relationship model in the intelligent agents, includes: Based on the target task and multiple target job information, the task transmission path of the target task among the target jobs is determined by utilizing the dependency relationships between the corresponding jobs in the multiple job intelligent agents. Based on the task delivery path and combined with the permission information of each target position, assignment information for each target position, including task execution requirements, time limits, and resource allocation, is generated.

[0011] In one feasible implementation, determining the task delivery path of the target task among the target positions based on the target task and multiple target position information, and utilizing the dependency relationships between the corresponding positions in the multiple position agents, includes: Starting with the target task, a directed graph for task transfer is constructed according to the coordination and control relationships between the corresponding positions in the multi-position intelligent agents. The graph algorithm searches the directed graph for the shortest path from the starting point to each target position, which serves as the task transfer path between the target positions.

[0012] A second aspect of the present invention provides a task allocation device based on an intelligent agent, applied to a multi-position organizational management system, the device comprising: The module is used to acquire the historical tasks of each position in the target organization, and analyze the organizational structure of the target organization to generate a three-dimensional organizational relationship model that reflects the dependencies between positions, and construct a multi-position intelligent agent. The analysis module is used to determine the assigned target task and analyze multiple related target job information of the target task; The generation module is used to input the target task and multiple target job information into multiple job intelligent agents, and generate the assignment information of each target job based on the corresponding three-dimensional organizational relationship model in the intelligent agent; The allocation module is used to allocate the target tasks based on the assigned information.

[0013] A third aspect of the present invention provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the electronic device to perform the above-described agent-based task allocation method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described agent-based task allocation method.

[0015] The technical solution provided by this invention involves: acquiring historical tasks of each position in a target organization; analyzing the organizational structure of the target organization to generate a three-dimensional organizational relationship model reflecting the dependencies between positions; constructing a multi-position intelligent agent; determining the assigned target tasks; analyzing multiple associated target position information of the target tasks; inputting the target tasks and multiple target position information into multiple position intelligent agents; generating assignment information for each target position based on the corresponding three-dimensional organizational relationship model in the intelligent agents; and performing the assignment operation of the target tasks based on the assignment information.

[0016] In this embodiment of the invention, by constructing a job-specific intelligent agent that reflects the dependency relationships between various positions in an organization, the job-specific intelligent agent can comprehensively allocate tasks based on the dependency relationships between various positions when allocating tasks, so that there will be no conflict in the execution of the allocated tasks, thereby improving the matching degree between tasks and positions. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of an embodiment of the agent-based task allocation method of the present invention; Figure 2 This is a schematic diagram of one embodiment of constructing a multi-position intelligent agent in this invention; Figure 3 A diagram illustrating role definition and task breakdown; Figure 4 A structural diagram of the three-dimensional relational model; Figure 5 Flowchart for agent initialization and task delegation information transmission; Figure 6 This is a schematic diagram of the task monitoring and recovery mechanism; Figure 7 A schematic diagram illustrating dynamic updates and closed-loop optimization of organizational structure; Figure 8 This is a schematic diagram of an embodiment of generating delegation information in this invention. Figure 9 This is a schematic diagram of one embodiment of the agent-based task allocation device of the present invention; Figure 10 This is a schematic diagram of one embodiment of the electronic device in this invention. Detailed Implementation

[0018] This invention provides a task allocation method, apparatus, device, and storage medium based on intelligent agents, enabling intelligent and structured optimization of task allocation in multi-position organizational management systems. The method constructs a three-dimensional organizational relationship model, formally modeling the permissions, coordination, and control relationships between positions. This allows the task delegation process to dynamically adapt to dependencies within the real organizational structure, effectively solving the problem of inefficient collaboration caused by the lack of organizational structure constraints in traditional task allocation. Simultaneously, the position relationship model extracted from historical task data, combined with the intelligent agent's organizational awareness triple (goal-norm-information domain), ensures that task allocation decisions consider both position responsibility boundaries and resource competence, significantly improving task execution success rates and organizational operational stability. This provides verifiable technical support for automated decision-making in complex enterprise scenarios.

[0019] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] It is understood that the executing entity of this invention can be an agent-based task allocation device, a terminal, or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as the executing entity as an example.

[0021] It should be noted that all information acquisition under this plan shall be conducted only with the explicit authorization of the user and in strict compliance with data privacy and security regulations.

[0022] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the task allocation method based on intelligent agents in this invention includes: 101. Obtain the historical tasks of each position in the target organization, and based on the organizational structure of the target organization, analyze the three-dimensional organizational relationship model that reflects the dependencies between positions, and construct a multi-position intelligent agent; In this step, the historical tasks of each position in the target organization are obtained, and multi-dimensional features of each historical task are extracted to obtain the permission information, task coordination features and hierarchical relationship of each position. Then, based on the organizational structure of the target organization and the extracted features, a three-dimensional organizational relationship model is constructed, and then a job intelligent agent is constructed based on the three-dimensional organizational relationship model.

[0023] It should be noted that when constructing job-specific intelligent agents based on the three-dimensional organizational relationship model, the organizational consciousness of the job is actually derived from the three-dimensional organizational relationship model. The job-specific intelligent agents are then constructed based on the organizational consciousness, enabling them to have the automatic reasoning ability to cooperate and negotiate among job roles.

[0024] 102. Determine the assigned target tasks and analyze the information of multiple related target positions for the target tasks; Specifically, semantic parsing is performed on the target task to extract key task elements; based on preset job and task element matching rules, the key task elements are compared with the responsibilities of each job to determine multiple related target job information that match the key task elements.

[0025] Specifically, natural language processing technology is used to parse the target task description and extract technical keywords and business constraints. The keywords are matched with positions with execution authority through permission information in the organizational three-dimensional relationship model, and the constraints are filtered to select positions with collaborative capabilities through coordination and control relationships.

[0026] 103. Input the target task and multiple target job information into the multi-job intelligent agent, and generate the assignment information of each target job based on the corresponding three-dimensional organizational relationship model in the intelligent agent; In this embodiment, the target task and multiple target job information are input into each job-specific intelligent agent. Each job-specific intelligent agent extracts the corresponding task content from the target task and outputs information related to the responsibilities of the corresponding job to generate assignment information.

[0027] Specifically, based on the permission information of the target position information, a delegating semantic action is first generated, which includes a task identifier, an execution period, and quality standard parameters. The task identifier is used to mark the position target inherited from the intelligent agent's organizational consciousness model. The execution period is generated based on the timeliness of information transmission in the coordination relationship. The quality standard parameters are set according to the task completion quality assessment in the control relationship. Then, the control relationship is used to detect whether the delegated semantic action violates the inter-position supervision responsibility chain; if there is a conflict, the task execution authority is reassigned according to the permission information, and the information transmission path is adjusted through the coordination relationship and control relationship to form the corrected delegation information.

[0028] 104. Assign target tasks based on the information assigned.

[0029] Specifically, each assignment information is sent to the corresponding target position intelligent agent; task acceptance confirmation information is received from each target position intelligent agent; if no confirmation information is received within a preset time, the assignment information is regenerated and sent.

[0030] In this embodiment of the invention, by constructing a three-dimensional organizational relationship model, the permissions, coordination and control relationships between positions are formally modeled, enabling the task assignment process to dynamically adapt to the dependencies in the real organizational structure. This solves the problems of assignment conflicts and low accuracy in existing task allocation schemes when assigning tasks with overlapping positions.

[0031] In alternative embodiments, such as Figure 2 The diagram illustrates the process of constructing a multi-role intelligent agent. The method for constructing a multi-role intelligent agent includes the following steps: 1011. Obtain the historical tasks of each position in the target organization, and extract the permission information, coordination relationship and control relationship between each position based on the historical tasks of each position.

[0032] Specifically, the historical tasks of each position are classified and organized, and labeled according to task type, task source, and task execution result; based on the labeled historical tasks, data mining algorithms are used to extract the permission information of each position, the coordination relationship between positions, and the control relationship.

[0033] It should be noted that Power refers to defining who has the authority to delegate tasks to whom; Coordination refers to defining the information flow and collaboration chain; and Control refers to defining the responsibilities for supervision and assessment.

[0034] 1012. Using the authority information of each position, the coordination and control relationships between positions as constraints, construct a three-dimensional organizational relationship model corresponding to all positions in the target organization based on the organizational structure of the target organization.

[0035] In this step, a formal definition is performed based on the organizational structure of the target organization, and a three-dimensional job relationship network (OS) of the target organization is constructed, wherein the three-dimensional job relationship network (OS) is as follows:

[0036] A set of job roles; A collection of intelligent agents; Indicates permission information; Indicates a coordinating relationship; Indicates a control relationship; This represents the mapping relationship between job positions and intelligent agents; Based on the permission information of each position, the coordination and control relationships between positions, and the control information extracted from the historical tasks of each position, the parameters in the three-dimensional relationship network of the positions are adjusted to obtain the three-dimensional relationship type of the organization corresponding to all positions in the target organization.

[0037] 1013. Based on the permission information of each position and the historical tasks of each position, the initial intelligent agent of each position is abstracted.

[0038] 1014. Use the three-dimensional relationship model of the organization to initialize the model parameters of all initial agents to obtain the target agents for each position.

[0039] In this embodiment, when constructing the target intelligent agent for each position, the organizational consciousness triplet for each position is derived based on the permission information, coordination relationship, and control relationship in the organizational three-dimensional relationship model. The organizational consciousness triplet includes: target O, norm N, and information domain I. The target O is generated through the power relationship in the organizational three-dimensional relationship model, the norm N is obtained by mapping the coordination relationship in the organizational three-dimensional relationship model, and the information domain I is determined by filtering the control relationship in the organizational three-dimensional relationship model. The model parameters of the initial intelligent agent for each position are initialized, each initial intelligent agent is bound to the position, and the organizational consciousness triplet is loaded to obtain the target intelligent agent for each position.

[0040] For example, each job role is assigned based on job information from historical tasks and the organizational three-dimensional relationship model. The organizational consciousness triplet is defined, and the resulting formula is as follows: , in: Indicate job objectives; Indicates behavioral norms; This indicates the information domain accessible to the agent. Playing a role Time (recorded as) Its internal state satisfies:

[0041] This allows for the construction of a job-specific intelligent agent possessing "goal awareness, normative constraints, and information sharing." Indicates the intelligent agent for the job position Business goals pursued That is, the task outcome that it needs to actively achieve; Indicates the intelligent agent for the job position Knowing clearly that you have an obligation to complete the action This fulfills their obligations, thereby enabling them to proactively fulfill their responsibilities in subsequent actions; Indicates the intelligent agent for the job position Information has been received This information can be used for task judgment and decision-making reasoning.

[0042] Then, based on the permission information, coordination relationship and control relationship in the organizational three-dimensional relationship model, the formula for updating the organizational awareness triplet of each position is derived to obtain the target intelligent agent of each position.

[0043] In this embodiment, after acquiring the historical tasks of each position in the target organization, analyzing the organizational structure of the target organization to generate a three-dimensional organizational relationship model reflecting the dependencies between positions, and constructing a multi-position intelligent agent, the method further includes: Based on the responsibilities and permissions of each position, the communication behaviors between the agents corresponding to each position are determined through semantic analysis. The communication behaviors include task delegation, status notification, and execution supervision. Based on the aforementioned communication behavior, the coordination and control relationships among the intelligent agents in each role are adjusted.

[0044] Specifically, the communication behavior between job-specific intelligent agents is defined by the following three types of actions: , , , During semantic analysis, its semantics satisfy the following logical constraints:

[0045] That is: when the position For the position When an agent has the authority to give instructions, if the agent... With the role To intelligent agents (Role Assign tasks Then the system generates obligations. i.e., intelligent agent Must fulfill the task When a coordination channel exists between roles, the information status notification inform(a,b,ϕ) can ensure that the receiving agent... Obtain task status or environment information This enables knowledge sharing among different positions; when a position right Having a controlling relationship, and They each play these two roles, and if monitored... Execute action Afterwards, the result If not, then the monitor Automatically generate compensation obligations .

[0046] In this embodiment, steps 1011-1014 can be implemented in three steps: role definition and task decomposition, establishment of a three-dimensional organizational relationship model, and agent initialization and role binding. The specific implementation of these three steps is as follows: S1, Role Definition and Task Breakdown.

[0047] This step abstracts the responsibilities and tasks of different positions within the organization by using historical figures associated with each position, and maps each position to an intelligent agent role. And identify the roles' objectives, norms, and information needs. Organize overall goals. Decomposed into several sub-targets Each sub-goal is mapped to a corresponding role. Task decomposition also considers the dependencies between tasks, forming a directed acyclic graph (DAG) to ensure that each sub-task has a clear responsible person and execution order, specifically as follows: Figure 3 As shown, each sub-goal also includes necessary resource constraints, priorities, and execution conditions, such as the time window required for the task, information dependencies, and role capability matching.

[0048] In this step, the system can also automatically generate a role responsibility matrix, mapping each sub-goal, dependent task, required information, and execution specification to a specific role, providing data structure support for subsequent agent initialization, task delegation, and information transmission.

[0049] S2, establish a three-dimensional relationship model of Power, Coordination, and Control.

[0050] Based on the role and task information obtained in step S1, establish a three-dimensional organizational relationship: Power, Coordination, and Control, to ensure that tasks can be executed efficiently and reliably, specifically as follows: Figure 4 vegetarian diet.

[0051] Power (Permissions Information): Determines the legitimacy of task delegation between roles. If a role For the role Having power, then the intelligent agent Can target Delegated to intelligent agents :

[0052] The assigned person automatically generates corresponding obligations. This forms a formal chain of tasks and responsibilities.

[0053] Coordination: Ensures timely transmission of information related to task execution. (If role) With the character If a coordination relationship exists, the information transmission action inform(a,b,φ) will cause the receiving agent to... Gain task knowledge :

[0054] The coordination path ensures that the delegated agent obtains the context information required to perform the task.

[0055] Control (Control Relationship): Used to monitor task completion and trigger remedial measures in case of failure. If role t has a control relationship with role s, the monitoring action monitor(c, DONE(α(b))) will generate a new obligation if the task is not completed. In this step, the system can also set power levels, information synchronization strategies, and monitoring priorities, so that the organizational structure not only conforms to the actual reporting chain, but also enables efficient collaboration in a multi-agent system.

[0056] S3, agent initialization and role binding.

[0057] After the organizational structure is established, each agent is initialized, bound to a corresponding role, and a Goal Base, Knowledge Base, and Obligation Base are generated:

[0058] The agent initialization process also includes resource allocation, capability parameter setting, and information interface registration to ensure that the agent can perform delegation, coordination, and monitoring actions. Figure 5 It illustrates the process of agent initialization and role binding, including the establishment and mapping of the internal state of each agent.

[0059] In this embodiment, as Figure 8 As shown, step 103, which involves inputting the target task and multiple target job information into multiple job-specific intelligent agents, and generating assignment information for each target job based on the corresponding organizational three-dimensional relationship model in the intelligent agents, includes the following steps: 1031. Based on the target task and multiple target job information, utilize the dependency relationships between the corresponding jobs in the multi-job intelligent agent to determine the task transmission path of the target task between the target jobs.

[0060] In this step, starting from the target task, a directed graph for task transmission is constructed according to the coordination and control relationships between the corresponding positions in the multi-position intelligent agents. The shortest path from the starting point to each target position is searched in the directed graph using a graph algorithm, which serves as the task transmission path of the target task between the target positions.

[0061] 1032. Based on the task delivery path and combined with the permission information of each target position, generate assignment information for each target position that includes task execution requirements, time limits, and resource allocation.

[0062] In this step, the core responsibilities of each target position are determined based on their role in the organizational three-dimensional relationship model along the task transmission path. Then, the scope of operable data, the system functions that can be invoked, and the collaboration methods are clarified by combining the permission information, and specific task execution requirements are formed.

[0063] Then, based on the total time requirement of the task delivery path, the available time is initially allocated according to the order of each position in the task process and the proportion of the expected processing time; combined with the permission information, the time is increased for positions with high permissions and great impact, and the time is reduced for positions with low permissions and auxiliary functions; the start and end times and key nodes of each position's task are determined to form time limit information.

[0064] Furthermore, analyze and calculate the total amount of human, material, and technical resources required for the target task; prioritize resources for positions with higher authority based on workload and importance of responsibilities for each position, combined with access information; allocate human resources according to job skills and number of personnel, allocate material resources according to actual needs and inventory, and allocate technical resources according to technical capabilities and requirements, thus forming resource allocation information.

[0065] In this embodiment, after the allocation operation of the target task based on each of the delegation information, the method further includes: When an abnormality is detected in the execution of the target task, a remedial obligation is generated based on a preset task remedial rule. The remedial obligation includes the remedial task and the permission information and collaborative resources of the responsible intelligent agent executing the remedial task. Based on the aforementioned remedial obligation, the intelligent agents of each position are invoked to reallocate the target task, and the result is matched with the previous allocation to determine the reallocation node for reallocation.

[0066] In practical applications, after tasks are assigned, they are executed. During task execution, the progress of task execution is evaluated in real time, specifically including the following steps: Dynamic simulation evaluation steps: A time-series simulation engine is used to maintain a dynamic organizational model DM(T)=(OS(T),E_d,E_i,E_m); where OS(T) is the organizational structure state at time T, E_d is the task delegation event set, E_i is the information notification event set, and E_m is the monitoring event set; the organizational structure evolution and task transfer process are simulated through an update function; the simulation results are correlated with the parameter optimization of the three-dimensional organizational relationship model.

[0067] The indicator calculation steps are as follows: Generate an organizational structure topology time-series evolution diagram, a task delegation path and execution rate matrix, and an agent obligation fulfillment and conflict statistics table. These indicators are used to evaluate task allocation efficiency, information flow, and control loop stability. Task allocation efficiency is calculated using the task delegation path and execution rate matrix; information flow is analyzed using the agent obligation fulfillment statistics table; and control loop stability is evaluated using the conflict statistics table. The evaluation results of this step are linked to parameter adjustments in the organizational three-dimensional relationship model. The optimization and adjustment steps involve dynamically adjusting the parameters of the three-dimensional relationship model based on the evaluation results. These adjustments include power relationship weight correction (based on task assignment success rate), coordination path optimization (based on information transmission timeliness), and control threshold update (based on task completion quality assessment). The adjustment results of this step are fed back to the multi-position agent generation stage to update agent attributes.

[0068] Remedial Obligation Generation Steps: When an abnormal task execution is detected, the system automatically generates a remedial obligation Comp(R_x,R_y,T') based on the control relationship; the obligation clarifies the remedial task T' and the responsible agent R_x; the responsible agent R_x determines the execution authority through the power relationship matrix and matches collaborative resources through the coordination relationship matrix; the remedial obligation generated in this step is related to the organization's three-dimensional relationship model; Closed-loop verification steps: After the responsible intelligent agent executes the remedial task, it verifies the consistency of the organizational system through the dynamic logic reasoning module; the module is based on the Kripke structural state machine model to ensure the complete execution of task delegation, information transmission and supervision chain; the verification results are fed back to the organizational three-dimensional relationship model to correct the relationship parameters, forming a closed-loop optimization of the entire task allocation process.

[0069] The following is combined Figure 5 The scenario provides a detailed explanation of the steps mentioned above, including task allocation, assessment of task completion, and remedial task execution.

[0070] S4, Assign tasks.

[0071] During task execution, the intelligent agent delegates tasks according to the Power relationship: Check the conditions for the Power relationship to be established; Execute the delegation action delegate(a,b,φ) to generate a new obligation. ; Perform the information transmission action inform(a,b,Obφ) to ensure that the receiving agent obtains task-related information. .

[0072] Figure 5 The process of task delegation, information transmission, and monitoring is displayed. In this step, each task also includes a timestamp, resource usage information, and execution conditions to ensure the traceability and scheduling rationality of multi-task execution.

[0073] S5, task monitoring and recovery mechanism.

[0074] The controlling agent monitors the assigned tasks. If a task is not completed, a remedial obligation is generated. This triggers remedial measures, such as reassignment, resource adjustment, or notification to the superior agent.

[0075]

[0076] Figure 6 The system displays the DAG (Directed Acyclic Graph) and agent execution paths for tasks, ensuring a reasonable task sequence, timely information synchronization, and complete obligation updates. In this step, the system can parameterize the monitoring frequency, remedial strategies, and exception handling logic to ensure that the agent can still execute tasks efficiently even with a large workload or dynamic environmental changes.

[0077] S6 enables dynamic updates and iterative execution.

[0078] As tasks are executed, delegated, and monitored, the agent's target base, knowledge base, and obligation base are updated in real time. , , ; This indicates that the job-specific intelligent agent is in the time sequence In between, when a new task is received → a new goal is added When new information is received → New facts are added When new responsibilities → new obligations are generated .

[0079] The organizational structure can dynamically adjust the relationships between Power, Coordination, and Control based on task completion, achieving adaptive optimization. Figure 7 The process of optimizing organizational structure is shown as follows: task execution → performance evaluation → structural adjustment → agent state update → next round of task iteration.

[0080] S7, Organizational Structure Assessment and Optimization.

[0081] The system periodically evaluates the organizational structure and agent performance, including metrics such as task completion rate, information transmission latency, and control coverage. Based on the evaluation results, it dynamically adjusts the three-dimensional relationships and updates the agent states, forming a closed-loop optimization: task execution and monitoring; performance evaluation (completion rate, information latency, control coverage); three-dimensional relationship optimization; agent state update; and the next task iteration. Through closed-loop optimization, the organizational structure achieves adaptive evolution, agent collaboration efficiency is improved, and the overall goal is ensured to be achieved efficiently.

[0082] In summary, this embodiment achieves multi-agent collaborative management based on organizational architecture through a systematic approach. It fully integrates task decomposition, role binding, task delegation, information synchronization, monitoring and remediation, and dynamic optimization loops to construct an efficient, controllable, and adaptive multi-agent execution framework. During implementation, each agent not only clearly defines its own goals and obligations but also collaborates with other agents based on the organization's three-dimensional relationship model (Power, Coordination, Control), achieving orderly task allocation and execution. By establishing a DAG task dependency graph and generating a role responsibility matrix, this invention ensures clear task responsibilities, explicit dependencies, and a reasonable execution order, while avoiding information silos and task conflicts.

[0083] Furthermore, in the task execution phase, the intelligent agent can delegate tasks based on power relationships, transmit information through coordination relationships, and implement monitoring and remedial measures through control relationships, thus forming a closed-loop operation process of "delegation—information—monitoring." With task iteration and real-time updates to the agent's state, the system can dynamically adjust its organizational structure and three-dimensional relationships to achieve adaptive optimization, improving overall task completion rate and organizational collaboration efficiency. Simultaneously, through real-time maintenance of the target database, knowledge base, and obligation database, the intelligent agent achieves closed-loop management from "should know - should do" to "known - done" during execution, ensuring the stable operation of multi-agent systems in complex environments.

[0084] The above describes the agent-based task allocation method in the embodiments of the present invention. The following describes the agent-based task allocation device in the embodiments of the present invention. Please refer to [link to documentation]. Figure 9 One embodiment of the task allocation device based on intelligent agents in this invention includes: The construction module 910 is used to obtain the historical tasks of each position in the target organization, and analyze the organizational three-dimensional relationship model of the organization based on the organizational structure of the target organization to reflect the dependency relationship between each position, and construct a multi-position intelligent agent. Analysis module 920 is used to determine the assigned target task and analyze multiple related target job information of the target task; The generation module 930 is used to input the target task and multiple target job information into multiple job intelligent agents, and generate the assignment information of each target job based on the corresponding three-dimensional organizational relationship model in the intelligent agent; The allocation module 940 is used to perform the allocation operation of the target task based on the delegation information.

[0085] Optionally, the building module 910 is specifically used for: Obtain the historical tasks of each position in the target organization, and extract the permission information, coordination relationships and control relationships between positions based on the historical tasks of each position. Using the authority information of each position, the coordination and control relationships between positions as constraints, a three-dimensional organizational relationship model corresponding to all positions in the target organization is constructed based on the organizational structure of the target organization. Based on the permission information of each position and the historical tasks of each position, an initial intelligent agent for each position is abstracted. The model parameters of all the initial agents are initialized using the organizational three-dimensional relationship model to obtain the target agents for each position.

[0086] Optionally, the building module 910 is specifically used for: Based on the organizational structure of the target organization, a formal definition is performed to construct a three-dimensional job relationship network of the target organization, wherein the three-dimensional job relationship network (OS) is as follows:

[0087] A set of job roles; A collection of intelligent agents; Indicates permission information; Indicates a coordinating relationship; Indicates a control relationship; This represents the mapping relationship between job positions and intelligent agents; Based on the permission information of each position, the coordination and control relationships between positions, and the control information extracted from the historical tasks of each position, the parameters in the three-dimensional relationship network of the positions are adjusted to obtain the three-dimensional relationship type of the organization corresponding to all positions in the target organization.

[0088] Optionally, the building module 910 is specifically used for: Based on the authority information, coordination relationships, and control relationships in the organizational three-dimensional relationship model, the organizational meaning triplet for each position is derived. The organizational meaning triplet includes: target O, norm N, and information domain I. The target O is generated through the power relationships in the organizational three-dimensional relationship model. The norm N is obtained by mapping through the coordination relationships in the organizational three-dimensional relationship model. The information domain I is determined by filtering through the control relationships in the organizational three-dimensional relationship model. Initialize the model parameters of the initial agents for each position, bind each initial agent to the position, and load the organizational consciousness triple to obtain the target agents for each position.

[0089] Optionally, the building module 910 is further configured to: Based on the responsibilities and permissions of each position, the communication behaviors between the agents corresponding to each position are determined through semantic analysis. The communication behaviors include task delegation, status notification, and execution supervision. Based on the aforementioned communication behavior, the coordination and control relationships among the intelligent agents in each role are adjusted.

[0090] Optionally, the generation module 930 is specifically used for: Based on the target task and multiple target job information, the task transmission path of the target task among the target jobs is determined by utilizing the dependency relationships between the corresponding jobs in the multiple job intelligent agents. Based on the task delivery path and combined with the permission information of each target position, assignment information for each target position, including task execution requirements, time limits, and resource allocation, is generated.

[0091] Optionally, the generation module 930 is specifically used for: Starting with the target task, a directed graph for task transfer is constructed according to the coordination and control relationships between the corresponding positions in the multi-position intelligent agents. The graph algorithm searches the directed graph for the shortest path from the starting point to each target position, which serves as the task transfer path between the target positions.

[0092] In this embodiment of the invention, by constructing a job-specific intelligent agent that reflects the dependency relationships between various positions in an organization, the job-specific intelligent agent can comprehensively allocate tasks based on the dependency relationships between various positions when allocating tasks, so that there will be no conflict in the execution of the allocated tasks, thereby improving the matching degree between tasks and positions.

[0093] above Figure 9 The task allocation device based on intelligent agents in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The electronic device in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0094] See Figure 10 As shown, the electronic device includes a processor 1000 and a memory 1001. The memory 1001 stores machine-executable instructions that can be executed by the processor 1000. The processor 1000 executes the machine-executable instructions to implement the above-described agent-based task allocation method.

[0095] Furthermore, Figure 10 The electronic device shown also includes a bus 1002 and a communication interface 1003. The processor 1000, the communication interface 1003 and the memory 1001 are connected via the bus 1002.

[0096] The memory 1001 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 1003 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 1002 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 10 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0097] The processor 1000 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 1000 or by instructions in software form. The processor 1000 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this disclosure can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 1001. Processor 1000 reads information from memory 1001 and, in conjunction with its hardware, completes the method steps of the aforementioned embodiments.

[0098] The present invention also provides an electronic device, the computer device including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor performs the steps of the agent-based task allocation method described in the above embodiments. The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer performs the steps of the agent-based task allocation method.

[0099] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0100] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0101] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A task allocation method based on intelligent agents, applied to a multi-position organizational management system, characterized in that, The method includes: The historical tasks of each position in the target organization are obtained, and a three-dimensional organizational relationship model reflecting the dependencies between positions is analyzed based on the organizational structure of the target organization to construct a multi-position intelligent agent. Determine the assigned target tasks and analyze the information of multiple related target positions for the target tasks; The target task and multiple target job information are input into multiple job intelligent agents, and the assignment information of each target job is generated based on the corresponding three-dimensional organizational relationship model in the intelligent agent; The target task is assigned based on the aforementioned delegation information.

2. The agent-based task allocation method according to claim 1, characterized in that, The process involves acquiring historical tasks for each position within the target organization, analyzing the organizational structure to create a three-dimensional organizational relationship model reflecting the dependencies between positions, and constructing a multi-position intelligent agent, including: Obtain the historical tasks of each position in the target organization, and extract the permission information, coordination relationships and control relationships between positions based on the historical tasks of each position. Using the authority information of each position, the coordination and control relationships between positions as constraints, a three-dimensional organizational relationship model corresponding to all positions in the target organization is constructed based on the organizational structure of the target organization. Based on the permission information of each position and the historical tasks of each position, an initial intelligent agent for each position is abstracted. The model parameters of all the initial agents are initialized using the organizational three-dimensional relationship model to obtain the target agents for each position.

3. The agent-based task allocation method according to claim 2, characterized in that, The process uses the authority information of each position, the coordination and control relationships between positions as constraints, and constructs a three-dimensional organizational relationship model corresponding to all positions in the target organization based on the organizational structure of the target organization. This includes: Based on the organizational structure of the target organization, a formal definition is performed to construct a three-dimensional job relationship network of the target organization, wherein the three-dimensional job relationship network (OS) is as follows: A set of job roles; A collection of intelligent agents; Indicates permission information; Indicates a coordinating relationship; Indicates a control relationship; This represents the mapping relationship between job positions and intelligent agents; Based on the permission information of each position, the coordination and control relationships between positions, and the control information extracted from the historical tasks of each position, the parameters in the three-dimensional relationship network of the positions are adjusted to obtain the three-dimensional relationship type of the organization corresponding to all positions in the target organization.

4. The agent-based task allocation method according to claim 2, characterized in that, The process of initializing the model parameters of all initial agents using the organizational three-dimensional relationship model to obtain the target agents for each position includes: Based on the authority information, coordination relationships, and control relationships in the organizational three-dimensional relationship model, the organizational awareness triplet for each position is derived. The organizational awareness triplet includes: goal O, norm N, and information domain I. Goal O is derived from the power relationships in the organizational three-dimensional relationship model. Norm N is obtained by mapping the coordination relationships in the organizational three-dimensional relationship model. Information domain I is determined by filtering the control relationships in the organizational three-dimensional relationship model. Initialize the model parameters of the initial agents for each position, bind each initial agent to the position, and load the organizational consciousness triple to obtain the target agents for each position.

5. The agent-based task allocation method according to any one of claims 1-4, characterized in that, After acquiring the historical tasks of each position in the target organization, analyzing the organizational structure of the target organization to generate a three-dimensional organizational relationship model reflecting the dependencies between positions, and constructing a multi-position intelligent agent, the process further includes: Based on the responsibilities and permissions of each position, the communication behaviors between the agents corresponding to each position are determined through semantic analysis. The communication behaviors include task delegation, status notification, and execution supervision. Based on the aforementioned communication behavior, the coordination and control relationships among the intelligent agents in each role are adjusted.

6. The agent-based task allocation method according to claim 1, characterized in that, The step of inputting the target task and multiple target job information into multiple job-specific intelligent agents, and generating assignment information for each target job based on the corresponding three-dimensional organizational relationship model in the intelligent agents, includes: Based on the target task and multiple target job information, the task transmission path of the target task among the target jobs is determined by utilizing the dependency relationships between the corresponding jobs in the multiple job intelligent agents. Based on the task delivery path and combined with the permission information of each target position, assignment information for each target position, including task execution requirements, time limits, and resource allocation, is generated.

7. The agent-based task allocation method according to claim 6, characterized in that, The step of determining the task transmission path of the target task among the target positions based on the target task and multiple target position information, and utilizing the dependency relationships between the corresponding positions in the multiple position agents, includes: Starting with the target task, a directed graph for task transfer is constructed according to the coordination and control relationships between the corresponding positions in the multi-position intelligent agents. The graph algorithm searches the directed graph for the shortest path from the starting point to each target position, which serves as the task transfer path between the target positions.

8. A task allocation device based on intelligent agents, applied to a multi-position organizational management system, characterized in that, The device includes: The module is used to acquire the historical tasks of each position in the target organization, and analyze the organizational structure of the target organization to generate a three-dimensional organizational relationship model that reflects the dependencies between positions, and construct a multi-position intelligent agent. The analysis module is used to determine the assigned target task and analyze multiple related target job information of the target task; The generation module is used to input the target task and multiple target job information into multiple job intelligent agents, and generate the assignment information of each target job based on the corresponding three-dimensional organizational relationship model in the intelligent agent; The allocation module is used to allocate the target tasks based on the assigned information.

9. An electronic device, characterized in that, The electronic device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the electronic device to execute the agent-based task allocation method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the agent-based task allocation method as described in any one of claims 1-7.

Images