A hierarchical multi-agent construction progress simulation method and system

Abstract

The application discloses a layered multi-agent construction progress simulation method and system, and relates to the technical field of intelligent construction, which comprises the following steps: importing construction project data into a corresponding modeling tool to obtain a construction information model corresponding to the construction project; analyzing construction process tasks based on a layered multi-agent structure to obtain standardized construction task units and constraint characteristic information; executing the layered multi-agent structure in the construction information model to make collaborative decisions on the standardized construction task units and output construction state characteristics; inputting the construction state characteristics and the constraint characteristic information into the layered multi-agent structure for iterative adjustment to output a construction progress plan; and the method realizes unified scheduling of tasks, resources and constraints through the collaborative linkage of the construction information model and the layered multi-agent, and improves the dynamic adjustment precision of the construction progress simulation and the reliability of the plan output.

Description

Technical Field

[0001] This invention relates to the field of intelligent construction technology, specifically to a hierarchical multi-agent construction progress simulation method and system. Background Technology

[0002] Construction progress management involves various types of information, including process organization, resource allocation, site constraints, and progress feedback. As project scale expands and professional collaboration increases, relying solely on Gantt charts, network plans, or manual experience to revise progress plans is no longer sufficient to meet the dynamic control needs of complex construction scenarios. Building Information Modeling (BIM) technology can integrate information such as components, spaces, processes, and resources into a unified model environment and form a four-dimensional progress simulation through time-dimensional correlation, enabling the construction process to be expressed based on model objects. Multi-agent technology can abstract construction process tasks, construction resources, site constraints, and scheduling decisions into different autonomous units, and handle complex construction organization problems through state interaction and rule-based collaboration. With the combination of BIM and multi-agent technology, construction progress simulation is gradually evolving from static display to dynamic deduction, conflict identification, and plan adjustment, providing a digital foundation for construction organization verification, resource allocation, and work surface coordination.

[0003] Existing construction progress simulation methods still have many shortcomings. While some methods incorporate Building Information Modeling (BIM), the model data often remains at the level of component display and schedule linkage. There is a lack of unified indexing relationships between construction objects, construction procedures, construction resources, and work areas, leading to a disconnect between model data and scheduling data. Some multi-agent methods lack clear division of agent responsibilities, and there is a lack of clear data transfer boundaries between task parsing, resource allocation, environmental assessment, and global scheduling. Changes in construction status are difficult to achieve continuous linkage between different decision-making units. When faced with on-site disturbances such as procedure delays, resource conflicts, weather changes, and limited workspace, common approaches are often local revisions, lacking a mechanism for overall judgment based on task chains and constraints. The progress simulation process also often focuses on the initial plan playback, lacking closed-loop iteration between simulation deviations, on-site feedback, and subsequent adjustments. This easily leads to asynchrony between the model display status, task execution status, and resource plan status. Therefore, the field of construction progress simulation still needs further improvement in model data organization, agent division of labor and collaboration, conflict state expression, and dynamic iteration mechanisms. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing construction progress simulation methods suffer from insufficient linkage between construction model data and scheduling decision data, unclear division of functions among multiple agents leading to low efficiency of collaborative decision-making, difficulty in continuously transmitting construction deviation and conflict states in the task chain, and the problem of how to realize construction process task parsing, constraint matching, collaborative adjustment, and iterative output of progress plan based on construction information model.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a hierarchical multi-agent construction progress simulation method, comprising: importing construction project data into a corresponding modeling tool to obtain a construction information model corresponding to the construction project; analyzing construction process tasks based on a hierarchical multi-agent structure to obtain standardized construction task units and constraint feature information; executing collaborative decision-making of standardized construction task units within the construction information model using the hierarchical multi-agent structure and outputting construction state features; inputting the construction state features and constraint feature information into the hierarchical multi-agent structure for iterative adjustment and outputting a construction progress plan.

[0007] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method described in this invention, the following steps are included: obtaining the construction information model corresponding to the construction project; objectifying and parsing the construction project data; generating task node records corresponding to the construction procedures according to the task identifiers of the construction procedures; generating a construction task topology based on the execution dependencies between the task node records; extracting a lightweight progress data layer from the construction task topology for the hierarchical multi-agent structure to call; configuring progress interaction fields for the task node records in the lightweight progress data layer, the progress interaction fields including call lock status and write-back version status; when the hierarchical multi-agent structure reads the task node records, updating the corresponding task node records to the call lock status; when the hierarchical multi-agent structure outputs construction status features or construction progress plans, verifying whether the current write-back data corresponds to the task node records based on the write-back version status, and writing the corresponding data into the construction object in the construction information model after the verification is passed.

[0008] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method described in this invention, the hierarchical multi-agent structure includes a task agent, a resource agent, an environmental constraint agent, and a collaborative scheduling agent. The task agent reads the task operation profile from the standardized construction task unit and generates the task operation status of the corresponding construction procedure. The resource agent generates the resource allocation status of the corresponding construction procedure based on the standardized construction task unit and constraint feature information. The environmental constraint agent performs constraint matching on the execution conditions of the standardized construction task unit based on constraint feature information and generates an environmental executable status. The collaborative scheduling agent associates the task operation status, resource allocation status, and environmental executable status according to the task identifier of the same construction procedure and generates a construction collaborative decision result.

[0009] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method of the present invention, the standardized construction task unit and constraint feature information include: extracting the corresponding task operation profile from the construction task topology according to the construction procedure task identifier; generating a construction task chain based on the task sequence in the task operation profile; writing the task operation profile into a preset task template to obtain a standardized construction task unit; and generating constraint feature information bound to the standardized construction task unit by matching the corresponding construction constraint rules based on the task operation profile. The task operation profile is used to characterize the execution object, task sequence, and executable conditions of a single construction procedure task in the construction information model.

[0010] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method described in this invention, the following steps are taken: A collaborative scheduling agent establishes a construction status determination record according to the construction procedure task identifier; the construction status determination record includes a planned status field, a task operation status field, an environment executable status field, and a model mapping field; based on the time difference and status difference between the planned status field and the task operation status field, a construction deviation status containing the deviation type and deviation amount is generated; based on the rule matching result between the environment executable status field and the execution condition field of the standardized construction task unit, a construction conflict status containing the conflict type, restricted time period, and recoverable time period is generated; the construction deviation status and construction conflict status are written into the task status record of the corresponding construction object in the construction information model according to the model mapping field, thereby obtaining construction status features for subsequent collaborative adjustments.

[0011] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method described in this invention, the collaborative decision-making includes: the collaborative scheduling agent generating a construction conflict set based on the construction deviation state and the construction conflict state; sorting the construction conflict set according to the constraint influence level and the task influence range, and generating a collaborative adjustment result; the collaborative adjustment result includes task adjustment parameters for adjusting the execution relationship of standardized construction task units, and constraint adjustment parameters for adjusting the executable state of the environment.

[0012] As a preferred embodiment of the hierarchical multi-agent construction progress simulation method described in this invention, the iterative adjustment includes: transmitting the collaborative adjustment result along the construction task chain to the affected task set and updating the construction state features; inputting the updated construction state features and constraint feature information back into the hierarchical multi-agent structure until the construction deviation state and construction conflict state meet the preset convergence conditions; determining the construction collaborative decision result that meets the preset convergence conditions as the construction progress plan, and binding it to the construction information model according to the construction procedure task identifier to form a 4D progress simulation result.

[0013] Another objective of this invention is to provide a hierarchical multi-agent construction progress simulation system. This system can establish a continuous interactive link between the construction information model, standardized construction task units, constraint feature information, and construction status features through data transfer between the model generation module, collaborative decision-making module, and progress output module. This solves the problems in current construction progress simulation systems, such as the separation of model data and scheduling data, unclear division of functions among multiple agents, difficulty in collaboratively resolving construction conflicts, and difficulty in iteratively writing the progress plan back to the model.

[0014] As a preferred embodiment of the hierarchical multi-agent construction progress simulation system described in this invention, it includes: a model generation module, a collaborative decision-making module, and a progress output module; the model generation module is used to establish a construction information model indexed by construction process task identifiers, and to provide a unified data call and write-back interface for standardized construction task units, constraint feature information, and construction progress plans; the collaborative decision-making module is used to configure task agents, resource agents, environmental constraint agents, and collaborative scheduling agents, and to perform associated scheduling of task running status, resource allocation status, and environmental executable status according to the same construction process task identifier, generating construction collaborative decision results; the progress output module is used to receive construction collaborative decision results and form construction status features, trigger iterative transmission of collaborative adjustment results based on construction status features and constraint feature information, and bind construction collaborative decision results that meet preset convergence conditions to the construction information model to form a construction progress plan.

[0015] Another object of the present invention is to provide a hierarchical multi-agent construction progress simulation device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the hierarchical multi-agent construction progress simulation method.

[0016] Another object of the present invention is to provide a hierarchical multi-agent construction progress simulation storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the hierarchical multi-agent construction progress simulation method.

[0017] The beneficial effects of this invention are as follows: The hierarchical multi-agent construction progress simulation method provided by this invention organizes construction project data into a construction information model indexed by construction process task identifiers. This model carries construction task topology, standardized construction task units, and constraint feature information, enabling construction objects, task sequences, resource constraints, and site constraints to be invoked and written back within the same data link, reducing the disconnect between the BIM model and schedule scheduling data. By associating task agents, resource agents, environmental constraint agents, and collaborative scheduling agents with the same task identifier, the task running status, resource allocation status, and environmental executable status can be uniformly aggregated to form a construction collaborative decision-making result, reducing scheduling conflicts caused by the homogenization of multi-agent functions and the disconnect between local decisions. By mapping construction deviation states and construction conflict states to the construction information model and transmitting the collaborative adjustment results along the construction task chain to the affected task set, schedule adjustment can be transformed from single-process correction to continuous iterative correction oriented towards the task chain. This invention achieves better results in terms of consistency between the construction model and scheduling data, hierarchical collaborative decision-making capabilities of multi-agents, and the accuracy of dynamic iteration in construction progress simulation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 The above is an overall flowchart of a hierarchical multi-agent construction progress simulation method provided in Embodiment 1 of the present invention.

[0020] Figure 2 This is an overall timing diagram of a hierarchical multi-agent construction progress simulation method provided in Embodiment 1 of the present invention.

[0021] Figure 3This is a hierarchical multi-agent structure diagram of a hierarchical multi-agent construction progress simulation method provided in Embodiment 4 of the present invention.

[0022] Figure 4 This is a system schematic diagram of a hierarchical multi-agent construction progress simulation method provided in Embodiment 6 of the present invention. Detailed Implementation

[0023] To make the objectives, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0024] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0026] The "construction project data" mentioned in this invention refers to the set of basic engineering data used to construct the construction information model 100 and support the simulation of construction progress. Specifically, it includes construction drawing data, construction organization design data, construction procedure data, construction resource data, construction site data, and construction constraint data. Among them, construction drawing data is used to determine the construction object, construction organization design data is used to determine the construction procedure and procedure sequence, construction resource data is used to determine the manpower, machinery, and material conditions required for each construction procedure, construction site data is used to determine the work area and space occupancy relationship, and construction constraint data is used to determine the resource, environmental, and site condition restrictions that the construction task is subject to during execution.

[0027] The "construction information model 100" mentioned in this invention refers to a model carrier formed after the construction project data is imported into the corresponding modeling tool, which is used to carry construction objects, construction tasks and construction progress data. The construction information model 100 includes at least a construction object identifier, a construction process task identifier, a construction task topology, a lightweight progress data layer and a progress write-back interface, so that the hierarchical multi-agent structure 200 can read the corresponding construction task data according to the construction process task identifier and write back the construction progress plan 400 to the corresponding construction object.

[0028] The "construction task identifier" mentioned in this invention refers to a data index used to uniquely represent a single construction process task. The construction task identifier can be formed by combining floor number, professional category, component number and process number. For example, "F03-STR-COL-REBAR-001" represents the first group of column reinforcement binding task in the three-story structure profession. The construction task identifier is used to associate construction object, task operation profile, constraint feature information 102, construction status feature 300 and construction schedule plan 400.

[0029] The "construction task topology" described in this invention refers to a task relationship structure formed by using construction task identifiers as nodes and the sequential execution relationship between construction tasks as connecting edges. In the construction task topology, each node corresponds to a construction procedure task, and each connecting edge represents the dependency relationship between the preceding and succeeding tasks. For example, the "column reinforcement binding" node is connected to the "column formwork installation" node, and the "column formwork installation" node is connected to the "column concrete pouring" node. The construction task topology is used to support the generation of standardized construction task units 101, the formation of construction task chains, and the transmission of collaborative adjustment results.

[0030] The "lightweight progress data layer" mentioned in this invention refers to a data layer that is extracted from the construction information model 100 and retains data related to construction progress simulation. Specifically, it includes construction task identifiers, construction object identifiers, task sequence relationships, planned execution intervals, task status fields, constraint index fields, and progress write-back fields. The lightweight progress data layer does not retain component fine textures, rendering materials, or auxiliary display data required for non-progress simulation, so that the layered multi-agent structure 200 can directly call data related to construction progress.

[0031] The “layered multi-agent structure 200” described in this invention refers to a four-layer collaborative decision-making structure consisting of a task agent 201, a resource agent 202, an environmental constraint agent 203, and a collaborative scheduling agent 204. The task agent 201 is used to generate the task running state, the resource agent 202 is used to generate the resource allocation state, the environmental constraint agent 203 is used to generate the environmental executable state, and the collaborative scheduling agent 204 is used to associate the above states according to the same construction task identifier and generate construction collaborative decision results.

[0032] The “task operation profile” mentioned in this invention refers to a data package extracted from the construction task topology and used to describe the operation conditions of a single construction procedure task. The task operation profile includes at least the construction task identifier, corresponding construction object, task sequence, planned execution interval, operation location, execution conditions, and task status fields. After the task operation profile is written into the preset task template, it forms a standardized construction task unit 101.

[0033] The "standardized construction task unit 101" mentioned in this invention refers to a construction procedure task data unit generated according to a preset task template. Each standardized construction task unit 101 corresponds to a construction procedure task and is bound to the construction object in the construction information model 100 through a construction task identifier, so that the task intelligent agent 201, resource intelligent agent 202 and environmental constraint intelligent agent 203 can read and process them respectively.

[0034] The “constraint feature information 102” mentioned in this invention refers to a set of data that is bound to the standardized construction task unit 101 and used to limit the executable conditions of the construction task. The constraint feature information 102 is generated by matching the task operation profile with construction constraint rules, and may specifically include resource availability conditions, resource occupancy conditions, weather restrictions, safety restrictions, and work area restrictions.

[0035] The “construction state feature 300” mentioned in this invention refers to the task state expression formed in the construction information model 100 based on the collaborative decision-making results of the hierarchical multi-agent structure 200, specifically including construction deviation state and construction conflict state; the construction state feature 300 is mapped to the corresponding construction object according to the construction task identifier for subsequent iterative adjustment.

[0036] The “construction deviation status” mentioned in this invention refers to the status data formed by comparing the task operation status with the planned status of the standardized construction task unit 101. For example, when the actual completion time is later than the planned completion time, the task execution status does not reach the planned status, or the task duration exceeds the planned execution interval, a corresponding construction deviation status is generated.

[0037] The “construction conflict state” mentioned in this invention refers to the state data formed by matching the environmental executable state with the execution conditions of the standardized construction task unit 101. For example, if the execution conditions of the construction task require concrete pouring under no-rainfall conditions, and the environmental executable state shows that there are rainfall restrictions during the current working period, a corresponding construction conflict state is generated.

[0038] The "construction task chain" mentioned in this invention refers to the construction task transmission link generated based on the task sequence in the task operation profile. The construction task chain is used to represent the sequential influence relationship between construction tasks, so that after the current construction task deviates or conflicts, the collaborative adjustment result can be transmitted along the construction task chain to the affected task set.

[0039] The “affected task set” mentioned in this invention refers to the set of subsequent construction tasks that need to be updated synchronously due to the adjustment of the current construction task, as determined along the construction task chain. For example, after the “column concrete pouring” task is postponed, the “beam reinforcement binding”, “beam formwork installation” and “beam concrete pouring” tasks that have subsequent dependencies on it can all be included in the affected task set.

[0040] The “cooperative adjustment result” mentioned in this invention refers to the adjustment data generated by the cooperative scheduling agent 204 based on the construction deviation state and the construction conflict state, specifically including task adjustment parameters and constraint adjustment parameters; the task adjustment parameters are used to adjust the execution relationship of the standardized construction task unit 101, and the constraint adjustment parameters are used to adjust the executable state of the environment, so that the set of affected tasks in the construction task chain can be updated synchronously.

[0041] The "preset convergence condition" mentioned in this invention refers to the set of conditions used to determine whether the iterative adjustment has ended. Specifically, it can be set as follows: the construction deviation state is lower than the preset deviation threshold, the construction conflict state is eliminated, the affected task set no longer adds new tasks, or the construction collaborative decision result remains stable within a consecutive preset round. When the preset convergence condition is met, the current construction collaborative decision result is determined as the construction schedule plan 400.

[0042] The “construction schedule plan 400” mentioned in this invention refers to the result of construction task execution formed after the preset convergence conditions are met. The construction schedule plan 400 includes at least the construction task identifier, execution sequence, execution interval, construction object binding relationship and adjustment record. The construction schedule plan 400 is bound to the construction information model 100 according to the construction procedure task identifier.

[0043] The “4D progress simulation result” mentioned in this invention refers to the four-dimensional construction progress expression result formed after binding the construction progress plan 400 to the construction objects in the construction information model 100 according to the time dimension; the 4D progress simulation result can display the task status, planned execution interval and adjusted construction sequence of the corresponding construction objects according to the construction time nodes.

[0044] Example 1, referring to Figures 1-2 This is the first embodiment of the present invention, which provides a hierarchical multi-agent construction progress simulation method to illustrate the overall operation process of the present invention from the input of construction project data to the output of the construction progress plan 400.

[0045] In this embodiment, the construction project can be a residential building project with 18 floors above ground and 2 floors underground. The construction project data includes architectural construction drawings, structural construction drawings, mechanical and electrical construction drawings, construction organization design documents, construction sequence division documents, construction resource configuration files, and site layout documents. For example, the construction resource configuration file records 2 tower cranes, 50 construction workers, 1000 tons of steel reinforcement, and 8000 m³ of concrete; the site layout document records a construction site that is 80m long and 50m wide, with temporary facilities located on the north side of the site; and the construction organization design document records the sequence arrangement of foundation engineering, main structure, secondary structure, and decoration and finishing.

[0046] First, import the above construction project data into the corresponding modeling tool to obtain the construction information model 100 corresponding to the construction project.

[0047] In practice, architectural, structural, and MEP construction drawings are imported into a Building Information Modeling (BIM) tool to create model objects including beams, slabs, columns, walls, and MEP components. Then, the construction procedures in the construction organization design documents are mapped to these model objects, ensuring that construction procedures such as "column reinforcement binding," "column formwork installation," "column concrete pouring," "beam reinforcement binding," "beam formwork installation," "beam concrete pouring," "floor slab reinforcement binding," "floor slab formwork installation," and "floor slab concrete pouring" can find their corresponding construction objects in the construction information model 100. In this embodiment, the construction information model 100 is not merely a 3D display model, but a model carrier that includes construction objects, construction procedures, construction resources, work areas, and progress tracking relationships.

[0048] Furthermore, based on the hierarchical multi-agent structure 200, the construction process tasks are analyzed to obtain standardized construction task units 101 and constraint feature information 102.

[0049] For example, the hierarchical multi-agent structure 200 reads the construction procedures of the main structure stage from the construction information model 100, identifies "three-story column reinforcement binding" as a construction procedure task, and converts it into a standardized construction task unit 101. The construction object corresponding to this standardized construction task unit 101 is the three-story column component. The corresponding prerequisite task can be the task after the completion of the two-story main structure, and the corresponding subsequent task can be the three-story column formwork installation task. For the "three-story column concrete pouring" task, the hierarchical multi-agent structure 200 will also combine concrete supply, tower crane or pump truck use, pouring personnel configuration and weather restrictions to generate constraint feature information 102 bound to the task.

[0050] Furthermore, in the construction information model 100, a hierarchical multi-agent structure 200 is executed to make collaborative decisions on the standardized construction task units 101 and output construction status features 300.

[0051] In practical implementation, the hierarchical multi-agent structure 200 reads each standardized construction task unit 101 and constraint feature information 102 to comprehensively judge the execution status, resource allocation status, and on-site executable status of the construction process. For example, during the main structure construction stage, if the task of binding the steel bars of the three-story columns is planned to take 5 days, but on-site feedback shows that it has not been completed by the 6th day, then the task corresponds to a construction deviation; if the task of pouring concrete for the three-story columns is planned to be executed within the next 3 days, but on-site weather data indicates that there will be moderate to heavy rain in the next 3 days and trigger the constraint rule of "prohibiting outdoor concrete construction during rainy weather", then the task corresponds to a construction conflict. The hierarchical multi-agent structure 200 maps the above deviations or conflicts to the three-story column components and their associated construction processes in the construction information model 100, forming construction status features 300.

[0052] It should be noted that the construction status feature 300 and constraint feature information 102 are input into the layered multi-agent structure 200 for iterative adjustment, and the construction schedule plan 400 is output. For example, when the concrete pouring task of the third-floor column cannot be carried out as originally planned due to rainfall conditions, the layered multi-agent structure (200) adjusts the concrete pouring task of the third-floor column to the executable period after the rainfall ends, based on the construction status feature 300 and the corresponding constraint feature information 102 of the task, and simultaneously adjusts the execution time of subsequent beam reinforcement binding, beam formwork installation, beam concrete pouring, floor slab reinforcement binding, floor slab formwork installation and floor slab concrete pouring; if tower crane occupation conflicts or insufficient construction personnel occur after adjustment, the updated construction status feature 300 is input into the layered multi-agent structure 200 for the next round of adjustment.

[0053] When the adjusted construction task arrangement no longer generates construction deviations or conflicts exceeding the preset range, the hierarchical multi-agent structure 200 outputs the current collaborative decision-making result as a construction schedule plan 400. The construction schedule plan 400 can include the execution sequence, execution time, corresponding construction object, and adjustment records of each construction process task, and is bound to the construction information model 100 according to the construction process task identifier. For example, the final construction schedule plan 400 can record that the third-floor column rebar binding will be executed in 5 days, the third-floor column formwork installation in 3 days, and the third-floor column concrete pouring in 2 days, while binding the adjustment record of the concrete pouring task being postponed due to rainfall to the corresponding column component.

[0054] Therefore, this embodiment forms a complete process from importing construction project data, generating construction information model 100, forming standardized construction task units 101 and constraint feature information 102, outputting construction status features 300, to iteratively generating construction schedule plan 400, enabling those skilled in the art to implement the method according to construction project data, construction model objects, construction process tasks and on-site constraints.

[0055] Example 2 is the second embodiment of the present invention. Based on Example 1, this embodiment limits the generation method of the construction information model 100 and adds task node recording, call lock state, write-back version state and write-back verification process. This solves the technical problem that the status of task nodes, construction objects and write-back results are easily inconsistent during multi-round iteration or multi-agent parallel call, and achieves the technical effect of keeping the task data reading and progress data write-back consistent in the construction information model 100.

[0056] In this embodiment, after importing the construction project data into the corresponding modeling tool, the construction project data is parsed in an object-oriented manner, and task node records for the corresponding construction procedures are generated according to the task identifiers of the construction procedures. The task node records are used to represent the callable data units of a single construction procedure task in the construction information model 100, so that the subsequent hierarchical multi-agent structure 200 can read the corresponding task data according to the same construction procedure task identifier.

[0057] In practical implementation, taking the main structure construction stage as an example, beams, slabs, columns, and other components in the construction drawings are analyzed as construction objects. The content in the construction organization design, such as column reinforcement binding, column formwork installation, column concrete pouring, and beam and slab construction, is analyzed as construction procedures. Construction procedure task identifiers are then generated according to floor, component category, and procedure type. For example, "F03-Column-Concrete Pouring" represents the task of pouring concrete for the third-floor column. Based on this construction procedure task identifier, the corresponding three-floor column components, their sequential construction relationships, and planned execution information are written into the same task node record.

[0058] Furthermore, a construction task topology is formed based on the execution dependencies between task node records, and a lightweight progress data layer is extracted from the construction task topology for the hierarchical multi-agent structure 200 to call. The lightweight progress data layer retains the task node records, the construction task topology, and the interaction data related to progress simulation, so that when the hierarchical multi-agent structure 200 performs task parsing and collaborative decision-making, it does not need to repeatedly traverse the complete construction information model 100, but can directly read the corresponding task node records according to the construction procedure task identifier.

[0059] Furthermore, in the lightweight progress data layer, progress interaction fields are configured for task node records. These progress interaction fields include a call lock status and a write-back version status. The call lock status indicates whether the current task node record is being called by the hierarchical multi-agent structure 200; the write-back version status indicates the version information of the task node record most recently received from the construction status feature 300 or the construction progress plan 400.

[0060] When the hierarchical multi-agent structure 200 reads the task node record, the construction information model 100 first locates the corresponding task node record according to the task identifier of the construction procedure, and determines the call lock status of the task node record. If the call lock status is idle, the task node record is updated to the call lock status and the task node record is output to the hierarchical multi-agent structure 200. If the call lock status is occupied, the output of the task node record is temporarily suspended until the current call process is completed before the corresponding call lock status is released.

[0061] For example, when task agent 201 reads the task node record corresponding to "F03-Column-Concrete Pouring", the construction information model 100 updates the call lock status of the task node record to "occupied". When resource agent 202 and environmental constraint agent 203 read the task node record under the same construction process task identifier, they both judge the resource allocation status and environmental executable status based on the same call version, avoiding different agents forming different judgment results under different data versions.

[0062] Furthermore, when the hierarchical multi-agent structure 200 outputs construction status feature 300 or construction schedule plan 400, the construction information model 100 verifies the correspondence between the current write-back data and the task node record based on the write-back version status. If the construction procedure task identifier carried by the current write-back data is consistent with the task node record, and the version corresponding to the current write-back data is not lower than the write-back version status in the task node record, then the verification passes, and the construction status feature 300 or construction schedule plan 400 is written into the corresponding construction object in the construction information model 100; if the verification fails, the write-back is rejected, and the original task node record is retained.

[0063] For example, the collaborative scheduling agent 204 determines that the "F03-Column-Concrete Pouring" task needs to be adjusted from the original plan of day 6-7 to day 9-10 due to rainfall limitations. After receiving the construction schedule plan 400, the construction information model 100 first verifies whether the task identifier of the construction process is "F03-Column-Concrete Pouring", and then verifies whether the write-back result corresponds to the current write-back version status. After the verification is successful, the adjusted execution interval, the reason for the adjustment, and the version update result are written to the task node record corresponding to the three-story column component, and the call lock status of the task node record is released.

[0064] Through the above processing, after generating the lightweight progress data layer, the construction information model 100 can not only provide task node records for the hierarchical multi-agent structure 200, but also control the task data reading process by calling the lock state, and verify the write-back process of the construction status feature 300 and the construction progress plan 400 by writing back the version status, thereby ensuring that the construction tasks, model objects and write-back results remain consistent during the multi-agent collaborative decision-making and multi-round iterative adjustment process.

[0065] Example 3, referring to Figure 3 This is the third embodiment of the present invention. Based on embodiment 2, this embodiment describes the specific operation process of obtaining the standardized construction task unit 101 and constraint feature information 102.

[0066] In this embodiment, a lightweight progress data layer has been formed in embodiment 2. The lightweight progress data layer stores a construction task topology indexed by the construction process task identifier. After the hierarchical multi-agent structure 200 is started, the task agent 201 first reads the task operation profile in the construction task topology item by item according to the construction process task identifier. The task operation profile is a data record that can be read by the agent. It includes at least the construction process task identifier, the corresponding construction object, the preceding task, the succeeding task, the planned start time, the planned end time, the work area, the execution conditions, and the current task status.

[0067] For example, for the task of "three-story column concrete pouring", the task operation profile read by task agent 201 from the construction task topology can be represented as follows: the construction process task is identified as "F03-column-concrete pouring", the corresponding construction object is the three-story column component, the preceding task is "three-story column formwork installation", the following task is "three-story beam reinforcement binding", the planned execution time is 2 days, the work area is the three-story main structure construction area, the execution conditions include the completion of formwork installation, the availability of concrete resources, the availability of pouring equipment, and the on-site weather meeting the pouring requirements, and the current task status is not started.

[0068] After reading the task execution profile, task agent 201 writes the profile into a preset task template to obtain a standardized construction task unit 101. The preset task template can use a unified field structure, including fields such as task identifier, construction object, preceding and following tasks, planned time period, execution conditions, and task status. After writing these fields, task agent 201 generates a standardized construction task unit 101 that can be jointly recognized by resource agent 202, environmental constraint agent 203, and collaborative scheduling agent 204.

[0069] For the main structure stage, the task agent 201 can sequentially generate multiple standardized construction task units 101, such as "three-story column reinforcement binding", "three-story column formwork installation", "three-story column concrete pouring", "three-story beam reinforcement binding", "three-story beam formwork installation", "three-story beam concrete pouring", "three-story floor slab reinforcement binding", "three-story floor slab formwork installation" and "three-story floor slab concrete pouring".

[0070] While generating standardized construction task units 101, the task agent 201 generates a construction task chain based on the task sequence in the task operation profile.

[0071] The construction task chain uses the task identifier of the construction process as the node and the dependency relationship between the preceding and succeeding tasks as the connection relationship. For example, "F03-Column-Reinforcement Binding" is connected to "F03-Column-Formwork Installation", "F03-Column-Formwork Installation" is connected to "F03-Column-Concrete Pouring", and "F03-Column-Concrete Pouring" is connected to "F03-Beam-Reinforcement Binding". This construction task chain is subsequently used to coordinate and adjust the transmission of results.

[0072] After receiving the standardized construction task unit 101, the resource agent 202 reads the planned time period field, work area field, and execution condition field, and extracts the available resources for the corresponding time period from the resource database. The resource database can record resource type, resource number, resource quantity, resource status, occupied time period, and location. For example, the resource database records that tower crane 1 is idle, tower crane 2 is idle, 20 steelworkers are available, 15 formwork workers are available, 15 concrete workers are available, 300 tons of steel reinforcement are available, and 1000 m³ of concrete is available. After matching the resource requirements of the standardized construction task unit 101 with the resource database, the resource agent 202 generates a resource allocation status.

[0073] For example, for the task of "three-story column rebar tying", resource agent 202 matches 10 rebar workers, 1 tower crane, and 50 tons of rebar, and writes the above results into the resource allocation status corresponding to the task; for the task of "three-story column formwork installation", resource agent 202 matches 8 formwork workers, and writes the time period occupied by the formwork workers into the resource allocation status; for the task of "three-story column concrete pouring", resource agent 202 matches concrete workers, concrete materials, and pouring equipment, and records whether the resources meet the requirements of the planned execution time period.

[0074] After receiving the constraint feature information 102, the environmental constraint agent 203 reads the execution conditions of the standardized construction task unit 101 and retrieves the on-site environmental data corresponding to the work area and planned time period. The on-site environmental data can come from the meteorological platform, on-site sensors or on-site inspection records, and can specifically include information such as rainfall, wind force, temperature, work surface occupancy, safety hazards and on-site interference. The environmental constraint agent 203 matches the on-site environmental data with the construction constraint rules to generate an environmental executable state.

[0075] For example, for the task of "three-story column concrete pouring", the constraint feature information 102 is configured with the construction constraint rule "outdoor concrete construction is prohibited in rainy weather". When the environmental constraint agent 203 reads that there is moderate to heavy rain during the planned execution period, it marks the environmental executable status of the task as restricted execution and records the reason for the restriction and the adjustable period. For the task of "three-story column rebar binding", if the work area is not occupied and the wind force does not exceed the limit, the environmental constraint agent 203 marks the environmental executable status of the task as allowed execution.

[0076] The collaborative scheduling agent 204 receives the task running status output by the task agent 201, the resource allocation status output by the resource agent 202, and the environment executable status output by the environment constraint agent 203. It then associates these statuses according to the task identifier of the same construction process. Specifically, the collaborative scheduling agent 204 establishes a task status record table, with each record using the construction process task identifier as the primary key and including the task running status, resource allocation status, and environment executable status. Once all three types of statuses under the same construction process task identifier have been written, the collaborative scheduling agent 204 generates the construction collaborative decision result for that standardized construction task unit 101.

[0077] For example, for the task of "three-story column concrete pouring", if the collaborative scheduling agent 204 reads that the task's running status is pending execution, the resource allocation status is that the concrete resources and pouring equipment meet the requirements, but the environment's executable status is restricted execution, then the collaborative scheduling agent 204 records the construction collaborative decision result of the task as temporarily not to be executed and proceeds to subsequent conflict handling; for the task of "three-story column formwork installation", if the task's running status is executable, the resource allocation status is that the requirements are met, and the environment's executable status is that execution is allowed, then the collaborative scheduling agent 204 records the construction collaborative decision result of the task as executed according to plan.

[0078] In this embodiment, the task agent 201, resource agent 202, and environmental constraint agent 203 establish bidirectional communication links based on the TCP / IP protocol with the collaborative scheduling agent 204. The task agent 201 sends task running status and task deviation information to the collaborative scheduling agent 204 through these bidirectional communication links. The resource agent 202 sends resource allocation status and resource warning information to the collaborative scheduling agent 204 through these bidirectional communication links. The environmental constraint agent 203 sends environmental executable status and constraint adjustment suggestions to the collaborative scheduling agent 204 through these bidirectional communication links. The collaborative scheduling agent 204 issues task adjustment parameters to the task agent 201, resource adjustment parameters to the resource agent 202, and constraint update parameters to the environmental constraint agent 203 through these bidirectional communication links, enabling each agent to update its corresponding status based on the construction collaborative decision-making results.

[0079] Furthermore, the data transmitted between the intelligent agents adopts a standardized JSON format. This standardized JSON format includes at least the following fields: construction process task identifier, data source intelligent agent, target receiving intelligent agent, state type, state value, conflict type, adjustment suggestion, and update time. Task intelligent agent 201, resource intelligent agent 202, and environmental constraint intelligent agent 203 send status data to the collaborative scheduling intelligent agent 204 according to a preset period. When task deviations, resource shortages, or environmental constraints occur, the corresponding intelligent agents immediately send abnormal status data through a two-way communication link. After receiving the abnormal status data, the collaborative scheduling intelligent agent 204 generates a construction collaborative decision result and sends the corresponding adjustment parameters back to the relevant intelligent agents.

[0080] Through the above process, the construction task topology formed in Example 2 is further converted into standardized construction task units 101 and constraint feature information 102. The task agent 201, resource agent 202, environmental constraint agent 203 and collaborative scheduling agent 204 complete the state generation and state association according to the same construction procedure task identifier, providing a data foundation for the formation of subsequent construction state features 300 and iterative adjustment process.

[0081] Example 4 is the fourth embodiment of the present invention. Based on the task running state, resource allocation state and environment executable state already formed in Example 3, this embodiment further explains how the construction state feature 300 is generated and how the collaborative scheduling agent 204 forms a collaborative adjustment result based on the construction state feature 300.

[0082] In this embodiment, after receiving the task running status, resource allocation status, and environment executable status under the same construction process task identifier, the collaborative scheduling agent 204 first establishes a construction status determination table. The construction status determination table uses the construction process task identifier as the primary key, and each status record includes at least the plan status field, task running status field, environment executable status field, construction deviation field, construction conflict field, and model mapping field.

[0083] Among them, the plan status field comes from the standardized construction task unit 101, the task operation status field comes from the task agent 201, the environment executable status field comes from the environment constraint agent 203, and the model mapping field is used to point to the corresponding construction object in the construction information model 100.

[0084] Specifically, the collaborative scheduling agent 204 first compares the task's running status with the planned status to generate a construction deviation status. During the comparison, the collaborative scheduling agent 204 reads the planned start time, planned completion time, and planned task status from the standardized construction task unit 101, and then reads the actual start time, actual completion time, and current task status reported by the task agent 201. When the current task status is inconsistent with the planned task status, or the actual completion time is later than the planned completion time, the corresponding construction procedure task identifier is written into the construction deviation field. The construction deviation status can include the deviation type, deviation duration, and deviation source. For example, if the planned completion time for the "F03-Column-Reinforcement Binding" task is 18:00 on day 5, and the task agent 201 reports that the task status is still in progress at 08:00 on day 6, then the collaborative scheduling agent 204 marks the task as a delayed status and records the deviation duration as 14 hours.

[0085] Furthermore, the collaborative scheduling agent 204 matches the executable state of the environment with the execution conditions of the standardized construction task unit 101 to generate a construction conflict state. During matching, the collaborative scheduling agent 204 reads the execution condition field in the standardized construction task unit 101 and then reads the executable state of the environment output by the environmental constraint agent 203. When the executable state of the environment is restricted execution or unexecutable, and the reason for the restriction matches the restriction rule in the execution condition, the corresponding construction procedure task identifier is written into the construction conflict field. The construction conflict state can include conflict type, conflict reason, restricted period, and recoverable period. For example, the execution conditions of the "F03-Column-Concrete Pouring" task include that outdoor concrete pouring must not be performed during periods of moderate to heavy rain. The environmental constraint agent 203 outputs that there will be moderate to heavy rain from day 6 to day 8. Then, the collaborative scheduling agent 204 marks the task as an environmental constraint conflict state and records the restricted period as day 6 to day 8 and the recoverable period as day 9 and beyond.

[0086] Furthermore, the collaborative scheduling agent 204, according to the construction procedure task identifier, writes the construction deviation state and construction conflict state into the corresponding construction object in the construction information model 100, forming the construction state feature 300. For example, the construction deviation state of "F03-column-reinforcement binding" is written into the reinforcement binding task record corresponding to the three-story column component, and the construction conflict state of "F03-column-concrete pouring" is written into the concrete pouring task record corresponding to the three-story column component. The written construction state feature 300 includes at least the construction procedure task identifier, construction object identifier, state type, state source, state value, and state update time, enabling the construction information model 100 to accept construction state changes on a per-construction-object basis.

[0087] After the construction state feature 300 is formed, the collaborative scheduling agent 204 generates a construction conflict set based on the construction deviation state and the construction conflict state. The construction conflict set is not a simple list of abnormal tasks, but a collection of deviation tasks and conflict tasks according to the task identifier of the construction procedure. Each construction conflict record includes the conflict task identifier, the conflict starting task, the conflict type, the constraint influence level, the task influence range, and the direction to be adjusted. Among them, the constraint influence level is determined according to whether the construction task can continue to be executed; the task influence range is determined according to the number of subsequent standardized construction task units 101 affected by the task in the construction task chain.

[0088] For example, if the "F03-Column-Concrete Pouring" task cannot be performed due to moderate to heavy rain, and its subsequent tasks include multiple tasks such as "F03-Beam-Reinforcement Binding," "F03-Beam-Formwork Installation," "F03-Beam-Concrete Pouring," and "F03-Floor Slab-Reinforcement Binding," then the constraint impact level of this conflict record is set to high, and the scope of the task impact is set to the set of beam and slab construction tasks on the same floor after the column concrete pouring task; if the "F03-Column-Reinforcement Binding" task is delayed by 14 hours, but the subsequent formwork installation task has not yet started, and the resource allocation status still meets the requirements, then the constraint impact level of this deviation record is lower than the aforementioned environmental constraint conflict.

[0089] It should be noted that the collaborative scheduling agent 204 sorts the set of construction conflicts according to the constraint impact level and the task impact range. During sorting, it first processes construction conflict states that render standardized construction task unit 101 unexecutable, then processes construction deviation states that affect a large number of subsequent tasks. If two conflict records have the same constraint impact level, the conflict record with the larger task impact range is processed first; if the task impact ranges are the same, the conflict record with the earlier planned execution time is processed first. Through this sorting rule, the collaborative scheduling agent 204 can determine the adjustment priority within the set of construction conflicts.

[0090] It should also be noted that the collaborative scheduling agent 204 generates collaborative adjustment results based on the sorting results. The collaborative adjustment results include task adjustment parameters and constraint adjustment parameters. The task adjustment parameters are used to adjust the execution relationship of standardized construction task units (101), and may specifically include the adjusted start time, adjusted completion time, predecessor task update value, and successor task update value; the constraint adjustment parameters are used to adjust the executable state of the environment, and may specifically include the restricted period, recovery period, and constraint state update value.

[0091] For example, for the "F03-Column-Concrete Pouring" task, the task adjustment parameters generated by the collaborative scheduling agent 204 are as follows: the original execution interval of day 6 to day 7 is adjusted to day 9 to day 10, and the start time of its immediate successor task "F03-Beam-Reinforcement Binding" is updated to day 11; the generated constraint adjustment parameters are as follows: the executable status of the environment corresponding to day 6 to day 8 is updated to restricted execution, and the executable status of the environment corresponding to day 9 to day 10 is updated to allowed execution. For the delayed "F03-Column-Reinforcement Binding" task, the task adjustment parameters generated by the collaborative scheduling agent 204 can be: the start time of the column formwork installation task is postponed by 14 hours, or the waiting interval between column formwork installation and column concrete pouring is compressed when the resource allocation status meets the requirements.

[0092] In this embodiment, the resource allocation status does not replace the construction deviation status or construction conflict status on its own, but rather serves as a constraint reference when the collaborative scheduling agent 204 generates the collaborative adjustment result. For example, when the column concrete pouring task is adjusted to day 9 to day 10, the collaborative scheduling agent 204 reads the resource allocation status and determines whether the concrete workers, pouring equipment, and concrete materials have been occupied by other standardized construction task units 101 during this period. If there is occupation, the task adjustment parameters are updated synchronously in the collaborative adjustment result to avoid resource conflicts introduced by new task execution relationships.

[0093] Through the above processing, this embodiment further converts the task running state, resource allocation state and environment executable state generated in embodiment 3 into construction state features 300 that can be written into the construction information model 100, and forms a collaborative adjustment result through construction conflict set, constraint influence level and task influence scope, providing an executable data foundation for the transmission and iterative adjustment of construction task chain in subsequent embodiments.

[0094] Example 5 is the fifth embodiment of the present invention. Based on Example 4, this embodiment describes the iterative adjustment and output process of the construction schedule plan 400, further explaining how the collaborative adjustment results are transmitted along the construction task chain and finally form the 4D progress simulation results.

[0095] In this embodiment, after generating the collaborative adjustment result, the collaborative scheduling agent 204 first reads the task adjustment parameters and constraint adjustment parameters in the collaborative adjustment result, and locates the starting task in the construction task chain according to the construction procedure task identifier. The starting task is the standardized construction task unit 101 that generates a construction deviation state or a construction conflict state. For example, in the case of "three-story column concrete pouring" being restricted by rainfall, the starting task is "F03-column-concrete pouring".

[0096] Furthermore, the collaborative scheduling agent 204 uses the starting task as the initial node and searches along the construction task chain for standardized construction task units 101 that have a immediate successor relationship with the starting task, forming an affected task set. The affected task set is not a direct selection of all subsequent tasks, but is determined step by step according to task dependencies. For example, after the adjustment of "F03-column-concrete pouring", its immediate successor task "F03-beam-reinforcement binding" is determined first, and then "F03-beam-formwork installation", "F03-beam-concrete pouring", "F03-floor slab-reinforcement binding", "F03-floor slab-formwork installation" and "F03-floor slab-concrete pouring" are determined, thus forming an affected task set for the same floor's main structure stage.

[0097] In one specific implementation, the collaborative scheduling agent 204 establishes an adjustment transfer table for the affected task set. The adjustment transfer table includes at least the construction procedure task identifier, the original planned execution interval, the adjusted execution interval, the update values ​​of the preceding task, the update values ​​of the succeeding task, the constraint status update values, and the adjustment round field. The collaborative scheduling agent 204 writes the task adjustment parameters into the adjustment transfer table and updates the execution interval of each standardized construction task unit 101 in the affected task set item by item according to the sequence of the construction task chain.

[0098] For example, the original execution interval for "F03-Column-Concrete Pouring" was from day 6 to day 7. Due to rainfall restrictions, the collaborative adjustment result changed it to day 9 to day 10. Based on this, the collaborative scheduling agent 204 updated the start time of "F03-Beam-Reinforcement Binding" to day 11 and the start time of "F03-Beam-Formwork Installation" to after the beam reinforcement binding is completed. The subsequent beam concrete pouring and floor slab construction tasks will continue to be extended according to the construction task chain. For each updated standardized construction task unit 101, the collaborative scheduling agent 204 retains both the original planned execution interval and the adjusted execution interval in order to determine whether the construction deviation status has been eliminated.

[0099] The collaborative scheduling agent 204 updates the construction status feature 300 based on the adjustment of the transfer table.

[0100] Specifically, for the starting task, the collaborative scheduling agent 204 adjusts parameters according to constraints to update the executable state of the environment; for subsequent tasks in the affected task set, the collaborative scheduling agent 204 recalculates the construction deviation state and construction conflict state according to the updated execution interval, and the updated construction state feature 300 is still mapped to the corresponding construction object in the construction information model 100 according to the construction procedure task identifier.

[0101] For example, when "F03-Column-Concrete Pouring" is executed from day 9 to day 10, the environmental executable status is updated from restricted to allowed. "F03-Beam-Reinforcement Binding" forms a new planned status due to a delayed start time, but if the delayed execution interval still meets the resource allocation status and on-site execution conditions, its construction conflict status is marked as conflict-free. The collaborative scheduling agent 204 writes the updated status into the corresponding three-story column, beam, and floor slab construction objects in the construction information model 100.

[0102] The updated construction state feature 300 and constraint feature information 102 are then input into the hierarchical multi-agent structure 200.

[0103] During this round of re-input, the task agent 201 reads the updated task status and determines whether the task execution interval in the affected task set satisfies the task order; the resource agent 202 reads the updated execution interval and determines whether there are resource occupation conflicts within the same time period; the environmental constraint agent 203 reads the updated constraint feature information 102 and determines whether environmental restrictions still exist in the adjusted execution interval; the collaborative scheduling agent 204 summarizes the task running status, resource allocation status, and environmental executable status again to generate new construction collaborative decision results.

[0104] For example, if the "F03-Column-Concrete Pouring" task is rescheduled to day 9 or 10, and resource agent 202 determines that the pump truck is already occupied by another floor's concrete pouring task on day 9, then resource agent 202 generates a new resource allocation state. Coordination agent 204 then re-includes this state in the construction conflict set and generates a new coordination adjustment result. If resources are available on day 9 or 10 and the environment is in an executable state (allowed to execute), then this task will no longer generate a construction conflict state.

[0105] In this embodiment, iterative adjustments are performed according to simulation rounds. Each simulation round includes: reading the current construction state feature 300, matching constraint feature information 102, updating the affected task set, regenerating the construction collaborative decision result, and determining the preset convergence condition. The preset convergence condition can be set to ensure that both the construction deviation state and the construction conflict state meet preset requirements. Specifically, it can be set to ensure that there are no unexecutable tasks in the current simulation round, the deviation value corresponding to the construction deviation state is lower than the preset deviation threshold, the affected task set does not add any new standardized construction task units 101, and the construction collaborative decision results remain consistent for two consecutive rounds.

[0106] For example, the simulation step size can be set to 1 day, and the preset deviation threshold can be set to 5%. In the first iteration, the system finds that "F03-Column-Concrete Pouring" is restricted by rainfall and generates a collaborative adjustment result; in the second iteration, the system checks whether new resource conflicts or environmental restrictions occur in the adjusted beam and slab related tasks; in the third iteration, if the comprehensive deviation value corresponding to the construction deviation state drops to 3% and no new construction conflict states are added, then the preset convergence condition is met.

[0107] The construction collaborative decision-making result that meets the preset convergence conditions is determined as the construction schedule plan 400.

[0108] The construction schedule 400 includes at least the construction process task identifier, the construction object identifier, the final execution interval, the task sequence, adjustment records, and constraint status records. The collaborative scheduling agent 204 binds the construction schedule 400 to the corresponding construction object in the construction information model 100 according to the construction process task identifier.

[0109] For example, the final construction schedule 400 records the execution period for "three-story column reinforcement binding" as day 1 to day 5, "three-story column formwork installation" as day 6 to day 8, and "three-story column concrete pouring" as day 9 to day 10. It also records that the task was adjusted due to rainfall restrictions from day 6 to day 8. At the same time, the adjusted execution periods for subsequent beam and slab construction tasks are written into the corresponding construction objects.

[0110] Furthermore, the construction information model 100 generates a 4D progress simulation result based on the construction schedule plan 400. In practice, the construction information model 100 reads the construction schedule plan 400 according to the time dimension and displays the task status of the corresponding construction object at the corresponding time node. For example, day 5 shows the completion of the third-floor column rebar tying task, day 8 shows the completion of the third-floor column formwork installation task, day 10 shows the completion of the third-floor column concrete pouring task, and day 30 shows the completion of the main structure to the corresponding floor level. Each construction object is bound to the execution interval and adjustment records in the construction schedule plan 400 through a construction process task identifier.

[0111] Through the above processing, the collaborative adjustment results formed in Example 4 are further transmitted to the affected task set along the construction task chain, and the matching results of construction state feature 300 and constraint feature information 102 are updated in each round of simulation adjustment until the preset convergence condition is met, and the construction progress plan 400 is output, and a 4D progress simulation result is formed in the construction information model 100.

[0112] Example 6, refer to Figure 4 As an embodiment of the present invention, a hierarchical multi-agent construction progress simulation system is provided, including a model generation module, a collaborative decision-making module, and a progress output module.

[0113] The model generation module is used to establish a construction information model 100 indexed by construction process task identifiers, and to provide a unified data call and write-back interface for standardized construction task units 101, constraint feature information 102 and construction schedule plan 400.

[0114] The collaborative decision-making module is used to configure the task agent 201, resource agent 202, environmental constraint agent 203, and collaborative scheduling agent 204, and to associate and schedule the task running status, resource allocation status, and environmental executable status according to the task identifier of the same construction procedure, and generate construction collaborative decision-making results.

[0115] The progress output module is used to receive the construction collaborative decision-making results and form the construction status feature 300. Based on the construction status feature 300 and the constraint feature information 102, it triggers the iterative transmission of the collaborative adjustment results and binds the construction collaborative decision-making results that meet the preset convergence conditions to the construction information model 100 to form the construction progress plan 400.

[0116] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a hierarchical multi-agent construction progress simulation method as proposed in the above embodiment.

[0117] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a hierarchical multi-agent construction progress simulation method as proposed in the above embodiment.

[0118] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0120] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0121] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0122] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A hierarchical multi-agent construction progress simulation method, characterized in that, include: Import the construction project data into the corresponding modeling tool to obtain the construction information model (100) corresponding to the construction project. Based on the hierarchical multi-agent structure (200), the construction process task is analyzed to obtain standardized construction task units (101) and constraint feature information (102). In the construction information model (100), a hierarchical multi-agent structure (200) is executed to make collaborative decisions on standardized construction task units (101) and output construction status characteristics (300). The construction status features (300) and constraint feature information (102) are input into the hierarchical multi-agent structure (200) for iterative adjustment, and the construction schedule plan (400) is output. The hierarchical multi-agent structure (200) includes a task agent (201), a resource agent (202), an environmental constraint agent (203), and a cooperative scheduling agent (204). The task agent (201) reads the task operation profile in the standardized construction task unit (101) and generates the task operation status of the corresponding construction process; The resource agent (202) generates the resource allocation status of the corresponding construction process based on the standardized construction task unit (101) and constraint feature information (102); The environmental constraint agent (203) performs constraint matching on the execution conditions of the standardized construction task unit (101) based on constraint feature information (102) to generate an environmental executable state; The collaborative scheduling agent (204) generates construction collaborative decision results by associating the task running status, resource allocation status and environmental executable status with the task identifier of the same construction process; The standardized construction task unit (101) and constraint feature information (102) include extracting the corresponding task operation profile from the construction task topology according to the construction procedure task identifier; A construction task chain is generated based on the task sequence in the task execution profile; Write the task operation profile into a preset task template to obtain a standardized construction task unit (101). Based on the task operation profile, the corresponding construction constraint rules are matched to generate constraint feature information (102) that is bound to the standardized construction task unit (101). The task operation profile is used to characterize the execution object, task sequence and executable conditions of a single construction process task in the construction information model (100); The iterative adjustment includes passing the collaborative adjustment results along the construction task chain to the affected task set and updating the construction status characteristics (300). The updated construction state features (300) and constraint feature information (102) are input again into the hierarchical multi-agent structure (200) until the construction deviation state and construction conflict state meet the preset convergence conditions. The construction collaborative decision-making results that meet the preset convergence conditions are determined as the construction schedule plan (400), and are bound to the construction information model (100) according to the construction procedure task identifier to form a 4D progress simulation result.

2. The hierarchical multi-agent construction progress simulation method as described in claim 1, characterized in that: The construction information model (100) corresponding to the construction project includes, The construction project data is parsed in an object-oriented manner, and task node records corresponding to the construction process are generated according to the task identifier of the construction process. A construction task topology is generated based on the execution dependencies between task node records, and a lightweight progress data layer is extracted from the construction task topology for the hierarchical multi-agent structure (200) to call. In the lightweight progress data layer, a progress interaction field is configured for the task node record. The progress interaction field includes the call lock status and the write-back version status. When the hierarchical multi-agent structure (200) reads the task node record, it updates the corresponding task node record to the call lock state; When the hierarchical multi-agent structure (200) outputs construction status features (300) or construction schedule (400), it verifies whether the current write-back data corresponds to the task node record based on the write-back version status, and writes the corresponding data into the construction object in the construction information model (100) after the verification is passed.

3. The hierarchical multi-agent construction progress simulation method as described in claim 1, characterized in that: The collaborative scheduling agent (204) establishes a construction status determination record according to the construction procedure task identifier; The construction status determination record includes a plan status field, a task running status field, an environment executable status field, and a model mapping field. Based on the time difference and status difference between the planned status field and the task running status field, a construction deviation status containing the deviation type and deviation amount is generated. Based on the rule matching results between the environment executable status field and the execution condition field of the standardized construction task unit (101), a construction conflict status including conflict type, restricted period and recoverable period is generated. According to the model mapping fields, the construction deviation status and construction conflict status are written into the task status record of the corresponding construction object in the construction information model (100), and the construction status features (300) for subsequent collaborative adjustment are obtained.

4. The hierarchical multi-agent construction progress simulation method as described in claim 3, characterized in that: The collaborative decision-making includes, The collaborative scheduling agent (204) generates a set of construction conflicts based on the construction deviation state and the construction conflict state; The construction conflict set is sorted according to the constraint impact level and task impact scope, and collaborative adjustment results are generated; The collaborative adjustment results include task adjustment parameters for adjusting the execution relationship of standardized construction task units (101) and constraint adjustment parameters for adjusting the environmental executable state.

5. A hierarchical multi-agent construction progress simulation system, employing the hierarchical multi-agent construction progress simulation method as described in any one of claims 1 to 4, characterized in that: It includes a model generation module, a collaborative decision-making module, and a progress output module; The model generation module is used to establish a construction information model (100) indexed by the construction process task identifier, and to provide a unified data call and write-back interface for standardized construction task units (101), constraint feature information (102) and construction schedule (400); The collaborative decision-making module is used to configure the task agent (201), resource agent (202), environmental constraint agent (203) and collaborative scheduling agent (204), and to associate and schedule the task running status, resource allocation status and environmental executable status according to the task identifier of the same construction process, and generate construction collaborative decision-making results; The progress output module is used to receive construction collaborative decision results and form construction status features (300), trigger the iterative transmission of collaborative adjustment results based on construction status features (300) and constraint feature information (102), and bind construction collaborative decision results that meet preset convergence conditions to construction information model (100) to form construction progress plan (400).

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the hierarchical multi-agent construction progress simulation method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the hierarchical multi-agent construction progress simulation method as described in any one of claims 1 to 4.

Images