AI-driven construction drawing compliance examination system and method

Abstract

The invention discloses an AI-driven construction drawing compliance examination system and method, and relates to the technical field of construction drawing examination. An initialization module establishes a construction element database based on an input construction drawing and a BIM model, automatically generates a construction sequence diagram by using an AI engine, and synchronously generates simulation scene parameters; the map construction module automatically loads matched construction process specifications and process standards and then constructs a specification map, the specification map is converted into a standardized map format, the simulation review module automatically calls the specification map through a simulation engine, component behaviors and space states are compared in real time, and if construction abnormities exist in the simulation process, the simulation review module reviews the component behaviors and the space states. And automatically marking corresponding process nodes to form a violation event set. According to the review system, the normal form transformation of construction drawing review from static rule matching to dynamic process evolution is realized, and the insight ability and the real-time response ability to complex construction logic and space behaviors are remarkably enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of construction drawing review, and in particular to an AI-driven construction drawing compliance review system and method. Background Art

[0002] With the increasing complexity of construction projects, the continuous compression of design cycles, and the improvement of national requirements for project quality and safety supervision, the traditional manual construction drawing compliance review method can no longer meet the needs of efficient, accurate and standardized review. Manual review not only has the disadvantages of low review efficiency, strong subjectivity, and easy omission of problems, but also has difficulty adapting to the needs of multi-professional collaborative review and rapid matching of massive regulatory provisions. Against this background, the development of artificial intelligence technology has provided a new path for construction drawing review.

[0003] The existing technology has the following defects:

[0004] Traditional construction drawing review usually relies on manual experience and static rules to compare construction drawings item by item. This method is not only highly dependent on professional reviewers, but also prone to omissions and deviations when dealing with large-scale engineering projects or complex construction scenarios. In particular, when it comes to dynamic factors such as construction sequence, component layout, and equipment operation paths, manual review cannot accurately predict timing conflicts, site overlaps, or violations of regulations that may occur during the construction phase, which greatly restricts the comprehensiveness and foresight of the review. In addition, the specification documents are diverse, complex in expression, and cross-applicable across stages. It is difficult for humans to achieve dynamic integration and semantic understanding of multi-dimensional rules, resulting in delayed and inconsistent implementation of the specifications.

[0005] Based on this, the present invention proposes an AI-driven construction drawing compliance review system and method, which realizes the paradigm shift of construction drawing review from "static rule matching" to "dynamic process evolution", significantly enhancing the insight into complex construction logic and spatial behavior and real-time response capabilities. Summary of the Invention

[0006] The purpose of the present invention is to provide an AI-driven construction drawing compliance review system and method to address the shortcomings of the background technology.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: an AI-driven construction drawing compliance review system, comprising an initialization module, a graph construction module, a simulation review module, and a decision output module;

[0008] Initialization module: Establishes a construction element database based on the input construction drawings and BIM models, uses the AI ​​engine to automatically generate a construction sequence diagram, and simultaneously generates simulation scene parameters. The construction element database, construction sequence diagram, and simulation scene parameters are sent to the simulation review module;

[0009] Atlas construction module: automatically loads the matching construction process specifications and process standards to build a specification atlas, converts the specification atlas into a standardized atlas format, and sends the specification atlas to the simulation review module;

[0010] Simulation review module: The simulation engine automatically calls the specification map to perform real-time comparison of component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set, which will be sent to the decision output module;

[0011] Decision output module: Based on the time points and spatial locations of the violation event set, it automatically traces back and generates the violation causal path, and then outputs the recommended decision.

[0012] In a preferred embodiment, the simulation review module drives the dynamic evolution of construction components in three-dimensional space and time through a simulation engine based on construction drawings, BIM models, and a construction element database, generating a component behavior sequence, the spatial state and behavior action of the component at each time step in the sequence, and each component state is represented by a spatial envelope box;

[0013] During the simulation, the real-time specification map is used to match the behavior of the current component with the rules in the specification map. The spatial conflict judgment function is used to determine whether there is a spatial conflict between components, and the process timing constraint judgment function is used to determine whether the component construction meets the dependency order.

[0014] In a preferred embodiment, the decision output module extracts the component number, violation time point, spatial coordinates, violation type, and violation rule number information, and reconstructs it into an event graph with a topological structure;

[0015] Based on the event graph structure, a time-backtracking analysis algorithm is used to trace the impact path of each violation event, identify the causal path, and generate a violation score for the causal path using a path scoring function.

[0016] Once the causal path of the violation is established, an avoidance plan is automatically generated based on the rule entries in the specification map and combined with the rule mapping logic engine of the AI ​​engine.

[0017] In a preferred embodiment, the decision output module generates a violation score for the causal path using a path scoring function, and the calculation expression is: Where: φ is the violation score of the violation causal path, α j represents the impact factor of component j in the causal path, Risk(C j ) represents component C j The risk probability of triggering a violation in historical data, n represents the number of upstream components involved in the causal path before the component violates the rule.

[0018] In a preferred embodiment, the decision output module calls the time backtracking analysis algorithm to trace back the impact path of each violation event and identify its causal path. Based on the topological sorting and path identification method, the violation causal path is recursively generated according to the time dimension and construction logic. Suppose a certain component C i At time t, a violation occurs and the causal path is expressed as: P i ={C i-n ,…,C i-1 ,C i},t i-n <… <t i , where: P i Represents component C i The causal path of the violation, C i-n Indicates the upstream process component that first caused the current violation, and the components meet The time dependency of n is the number of upstream components involved in the causal path before the component violates the rule. represents the simulation start time of process i, represents the completion time of process j.

[0019] In a preferred embodiment, the simulation review module advances the behavior simulation of each component based on time in the dynamic simulation evolution, and automatically compares the current state of the behavior node with the spatial specifications and process sequence map rules at each time step. If a violation is detected, the behavior node is marked as a violation process and a violation event is generated. The violation event records the component ID, timestamp, violation type, involved object, violation rule number and severity level.

[0020] In a preferred embodiment, the graph construction module loads the construction process specification text and process standard manual of the project type and stage, and performs semantic analysis to extract the elements in the standard, including process name, construction conditions, equipment requirements and space restrictions;

[0021] The extraction results are converted into nodes and edges in a graph structure. Nodes represent construction entities, and edges represent the process logic, spatial constraints, sequence, or dependency rules between entities.

[0022] In a preferred embodiment, the initialization module performs semantic recognition and object parsing on the input two-dimensional construction drawings and three-dimensional BIM models, including component types, sizes, materials, positions, and connection relationships;

[0023] The AI ​​engine automatically constructs a construction sequence diagram based on the structural dependencies between components and the process annotations in the drawings. In the construction sequence diagram, nodes represent components and arrows represent the construction sequence dependencies.

[0024] Based on the construction sequence diagram and the geometric information of the components, the scenario parameters for construction simulation are automatically generated. The scenario parameters include path planning, component entry time window, equipment layout area, and stacking location.

[0025] In a preferred embodiment, the operation of the AI ​​engine includes the following steps:

[0026] The structural dependency assignment, spatial overlap and construction process matching between components are obtained, and the structural dependency assignment, spatial overlap and construction process matching are normalized so that the value range of the structural dependency assignment, spatial overlap and construction process matching is mapped to [0,1]. The normalized value of the structural dependency assignment, the normalized value of the spatial overlap and the normalized value of the construction process matching are obtained, and the normalized value of the structural dependency assignment, the normalized value of the spatial overlap and the normalized value of the construction process matching are summed to obtain the dependency weight between components.

[0027] An AI-driven construction drawing compliance review method, the review method comprising the following steps:

[0028] The review system establishes a construction element database based on the input construction drawings and BIM models, uses the AI ​​engine to automatically generate construction sequence diagrams, and simultaneously generates simulation scenario parameters;

[0029] Automatically load the matching construction process specifications and process standards to build a specification map, and convert the specification map into a standardized map format;

[0030] The simulation engine automatically calls the specification map to perform real-time comparison between component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set.

[0031] Based on the time points and spatial locations of the violation event set, the violation causal path is automatically traced back and the recommended decision is output.

[0032] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0033] The present invention establishes a construction element database based on the input construction drawings and BIM models through the initialization module, automatically generates a construction sequence diagram using the AI ​​engine, and simultaneously generates simulation scenario parameters. The map construction module automatically loads the matching construction process specifications and process standards to construct a specification map, and converts the specification map into a standardized map format. The simulation review module automatically calls the specification map through the simulation engine and performs real-time comparison of component behavior and spatial status. If there are construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set. The decision output module automatically traces the violation causal path based on the time point and spatial location of the violation event set and outputs a recommended decision. This review system realizes the paradigm shift of construction drawing review from "static rule matching" to "dynamic process evolution", significantly enhancing the insight into complex construction logic and spatial behavior and the real-time response capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0035] Figure 1 This is a system architecture diagram of the review system of the present invention.

[0036] Figure 2 This is a mind map of the review system of the present invention.

[0037] Figure 3 The present invention is a flowchart of the method for reviewing the present invention.

[0038] Figure 4 A mind map of the present invention's review method. DETAILED DESCRIPTION

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0040] Example 1: Please refer to Figure 1-Figure 2 and Figure 4 As shown, this embodiment provides an AI-driven construction drawing compliance review system, including an initialization module, a graph construction module, a simulation review module, and a decision output module;

[0041] Initialization module: Establishes a construction element database based on the input construction drawings and BIM models, uses the AI ​​engine to automatically generate a construction sequence diagram, and simultaneously generates simulation scene parameters. The construction element database, construction sequence diagram, and simulation scene parameters are sent to the simulation review module;

[0042] Atlas construction module: automatically loads the matching construction process specifications and process standards to build a specification atlas, converts the specification atlas into a standardized atlas format, and sends the specification atlas to the simulation review module;

[0043] Simulation review module: The simulation engine automatically calls the specification map to perform real-time comparison of component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set, which will be sent to the decision output module;

[0044] Decision output module: Based on the time points and spatial locations of the violation event set, it automatically traces back and generates the violation causal path, and then outputs the recommended decision.

[0045] This application establishes a construction element database based on the input construction drawings and BIM models through the initialization module, automatically generates a construction sequence diagram using the AI ​​engine, and simultaneously generates simulation scene parameters. The graph construction module automatically loads the matching construction process specifications and process standards to construct a specification graph, and converts the specification graph into a standardized graph format. The simulation review module automatically calls the specification graph through the simulation engine and performs a real-time comparison of component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set. The decision output module automatically traces the violation causal path based on the time point and spatial position of the violation event set and outputs a recommended decision. This review system realizes the paradigm shift of construction drawing review from "static rule matching" to "dynamic process evolution", significantly enhancing the insight into complex construction logic and spatial behavior and the real-time response capability.

[0046] Example 2: The initialization module establishes a construction element database based on the input construction drawings and BIM models, uses the AI ​​engine to automatically generate a construction sequence diagram, and simultaneously generates simulation scene parameters. The construction element database, construction sequence diagram and simulation scene parameters are sent to the simulation review module.

[0047] The initialization module is the starting point of the AI-driven construction drawing compliance review system. Its core goal is to automatically convert the input construction drawings and BIM models into computable and simulatable construction element data structures, and provide clear construction sequence logic and three-dimensional scene parameter support for subsequent simulation and review processes.

[0048] First, semantic recognition and object parsing are performed on the input 2D construction drawings (such as CAD drawings) and 3D BIM models (such as Revit models), mainly including properties such as component type, size, material, location, and connection relationship.

[0049] Table 1: Construction element database structure (simplified example)

[0050]

[0051] Table 1 is automatically generated by the initialization module. The component information is derived from BIM element attributes and spatial relationship reasoning. The component status is initially "to be installed" or "installed" and is used for the subsequent generation of construction sequence logic.

[0052] The initialization module uses the AI ​​engine to automatically build a construction sequence diagram based on the structural dependencies between components (such as columns before beams, foundations before walls) and the process annotations in the drawings (such as construction section divisions). The AI ​​engine operates in the following steps:

[0053] The structural dependency assignment, spatial overlap, and construction process matching between components are obtained, and the structural dependency assignment, spatial overlap, and construction process matching are normalized so that their value ranges are mapped to [0,1]. The normalized values ​​of the structural dependency assignment, spatial overlap, and construction process matching are obtained, and the normalized values ​​of the structural dependency assignment, spatial overlap, and construction process matching are summed to obtain the dependency weight between components. Assuming that the dependency weight from component i to component j is larger, it means that the construction of component j is more dependent on the completion of the construction of component i.

[0054] Structural Dependency Assignment D ij Reflects the support relationship between components in mechanical or structural systems, such as columns supporting beams and beams supporting plates. Data source:

[0055] Component connection information in the BIM model (such as Revit's structural connection constraints);

[0056] Calculation logic for IFC file structural relationships (such as IfcRelConnectsStructuralElement): Uses Boolean mapping + weight scoring mechanism:

[0057]

[0058] Example:

[0059] Column (C001) supporting beam (C002) → D C001,C002 =1.0;

[0060] Beam (C002) indirectly supports the ceiling → D C002,Cx =0.7.

[0061] Structural dependency assignments characterize the mechanical attachment relationships and load transfer sequence between components within a structural system. They reflect whether a component relies on another for structural stability or installation location. This dependency is a strong constraint that must be prioritized when generating construction sequence diagrams.

[0062] For example, in a multi-story frame structure, beams must be installed on top of the columns or walls to which they are attached. Otherwise, problems such as the structure being unable to support loads or the lifting points being incorrectly positioned will arise. Ignoring structural dependency assignments in the AI ​​engine can easily lead to a construction sequence that violates mechanical principles, distorting simulation results and even incorrectly determining that certain "premature installation" operations are legitimate. Therefore, structural dependency assignments are the most core, foundational constraints in the construction sequence diagram, determining the load-bearing order of components.

[0063] Furthermore, this parameter can be used to dynamically assess the impact of construction schedule adjustments caused by missing components or design changes. For example, if the construction of a key column is delayed, the system can automatically determine the scope of the impact based on the dependency weight graph and adjust the installation time of subsequent dependent components.

[0064] Spatial overlap S ij Quantify the degree of spatial overlap between component i and component j, reflecting the potential for construction conflict and the closeness of dependency. Data source:

[0065] Component 3D coordinate bounding box (Bounding-Box);

[0066] When there is spatial overlap between components: Where V overlap Represents the overlapping volume of two components, V i ,V j Example of the bounding box volume representing components i and j:

[0067] Beam (C002) and slab (C003) are laid together → S = 1.0;

[0068] The distance between the column (C001) and the far wall (C005) is large → S = 0.2.

[0069] Spatial overlap describes the proximity or potential interference between components in three-dimensional space. It critically impacts accessibility, safety, and construction path planning. Even if components have no structural dependencies, close proximity or overlap can lead to installation conflicts, operational interference, or overlapping equipment paths during construction.

[0070] When generating construction sequence diagrams or simulating the construction process, if the AI ​​engine fails to consider spatial overlap, it may schedule the simultaneous construction of multiple adjacent components, causing severe congestion on the construction site and even creating safety hazards (such as intersecting lifting paths). Therefore, spatial overlap is a core parameter that affects the "rationality of the construction operation sequence" and is widely used to: determine whether there are spatial overlap installation conflicts, predict construction interference caused by spatial proximity between components, and optimize construction paths and component stacking logic (such as the stacking order of component loading and unloading areas). In addition, spatial overlap can also be used for construction site resource scheduling and equipment deployment, such as determining whether multiple lifting operations are allowed to proceed simultaneously to avoid resource conflicts.

[0071] The construction process matching degree expresses the degree of fit between the construction sequence of components and the standard process.

[0072] Establish a standard construction sequence, as shown in Table 2:

[0073] Table 2: Rule mapping scoring table

[0074] Component combination Standard order Match Column→Beam yes 1.0 Beam → Plate yes 1.0 Plate→Beam no 0.0 Wall → Window yes 0.8

[0075] Data source:

[0076] Project construction process specification diagram (for example: lifting sequence standard);

[0077] Industry knowledge maps and standard documents (such as "Construction Manual" and "Concrete Structure Construction Specifications").

[0078] Construction process matching is a key indicator of whether the construction sequence between components complies with existing industry construction specifications, technical standards, or enterprise process maps. Essentially, this approach structures and computates empirical engineering knowledge data and embeds it into AI simulation models, ensuring the resulting construction sequence diagram is not only logically sound but also realistically operational and meaningful for engineering practice.

[0079] In many cases, structural dependencies and spatial constraints may allow for flexible parallel or sequential installation of components, but a truly rational installation process depends on technical details, construction safety regulations, and material installation methods. For example, the order of attaching insulation panels must follow the principle of "bottom-up, edge first, center second"; elevator shaft formwork and rebar must be completed before internal electromechanical pre-embedded components can be installed; pipeline installation must begin with the main trunk and then the branch, gravity first, then pressure, and large diameters first, then small diameters. This industry knowledge cannot be directly derived from geometric or structural models, but it can be modeled through rule matching or knowledge graph mapping through construction process matching to achieve knowledge compliance review of construction behavior. AI engines rely on this parameter to automatically avoid "process errors" in the generated construction process, such as unreasonable installation sequences and violations of standard operating procedures. It can also be used for cross-project transfer learning, helping AI models automatically generate more practical process sequences based on previous project experience.

[0080] Construction sequence diagram direction (example):

[0081]

[0082] In the above construction sequence direction, nodes represent components and arrows represent the construction dependency relationship.

[0083] Based on the construction sequence diagram and the geometric information of the components, the scenario parameters for construction simulation are automatically generated, such as path planning, component entry time window, equipment layout area, stacking location, etc., as shown in Table 2:

[0084] Table 3: Simulation scenario parameters (schematic)

[0085] Component number Lifting equipment type Lifting starting point Lifting end point Hoisting time window Stacking location Estimated installation time C001 Tower Crane A (5,5,0) (12.5,8.2,4.0) 08:00-10:00 (4,4,0) 25 minutes C002 Tower Crane A (5,5,0) (12.5,8.2,0.0) 07:30-08:30 (4,3,0) 15 minutes

[0086] The data in Table 3 are derived from the combination of path calculation (based on A* or Dijkstra algorithm) and component geometry information with site layout rules, and are automatically generated as a scheduling basis for simulation.

[0087] The completed construction element database, construction sequence diagram, and simulation scenario parameters are sent to the simulation review module through the system's internal pipeline for subsequent dynamic simulation. The pseudo code example of module data organization and transmission is as follows:

[0088]

[0089]

[0090] This pseudocode shows how data is organized within the module and sent to the downstream simulation review module. The three types of core data provide complete support for subsequent simulation logic.

[0091] Through these steps, the initialization module not only achieves fully automated conversion from drawings and BIM models to data structures and simulation scenarios, but also uses AI to calculate construction sequence logic and dynamic parameters, avoiding misjudgments or omissions caused by human intervention. This provides accurate, structured, and evolvable input conditions for the simulation review module. This process lays a solid foundation for subsequent violation identification and path tracing, ensuring high-precision and real-time review results.

[0092] The atlas construction module automatically loads the matching construction process specifications and process standards to construct a specification atlas, converts the specification atlas into a standardized atlas format, and sends the specification atlas to the simulation review module.

[0093] The graph construction module automatically loads construction process specifications and process standard manuals appropriate for the project type and phase, performs semantic analysis, and extracts key elements from the standards (such as process name, construction conditions, equipment requirements, and space restrictions) and the logical relationships between them. This process utilizes language models such as BERT or RoBERTa combined with dependency syntax trees to extract entities and relationships.

[0094] The extracted results are converted into nodes and edges in a graph structure. Nodes represent construction entities (such as components, equipment, and personnel actions) or abstract specifications, while edges represent process logic, spatial constraints, sequencing, or dependency rules between entities. The graph structure is scalable and supports multi-dimensional nesting.

[0095] The graph rule can be abstracted into a triple expression: R = (E i ,R ij ,E j ), where: E i ,E j Represents two construction entity nodes (such as components, processes, operating conditions, etc.), R ij It represents the process relationship between entities, which can be "front", "protruding in space", "needing edge protection", "equipment sharing", etc. R represents a regular relationship, which can form an edge in the graph and is used to standardize simulation behavior.

[0096] In order to connect with the simulation review module, the constructed graph will be uniformly converted into a standard graph format (such as RDF / OWL representation or Neo4jJSON graph structure) and annotated with metadata such as node weight and rule priority.

[0097] The specification graph is ultimately sent to the simulation review module through an interface. The module calls the subgraphs in the graph based on the simulation scenario and performs process compliance comparison and timing logic verification in real time. The following is a simplified pseudocode for graph construction (using the Neo4j graph database):

[0098] frompy2neoimportGraph,Node,Relationship

[0099] #Connect to the graph database

[0100] graph=Graph("bolt: / / localhost:7687",auth=("neo4j","password"))

[0101] #Build a standard entity node

[0102] n1 = Node ("Process", name = "Component Hoisting")

[0103] n2 = Node("Restriction", name = "Lifting Radius < 30m")

[0104] n3 = Node ("Protective Measures", name = "Edge Guardrail")

[0105] #Add process specification edge

[0106] r1=Relationship(n1,"must satisfy",n2)

[0107] r2=Relationship(n1,"must be set",n3)

[0108] #Write to graph database

[0109] graph.create(n1|n2|n3|r1|r2)

[0110] The above code indicates that the component hoisting process in the diagram must meet the lifting radius requirement of less than 30 meters and must be equipped with edge guardrails. This format can be directly read by the AI ​​compliance engine and used during simulation to determine whether the lifting path is legal and whether protective measures are in place.

[0111] The graph format diagram (standard graph node / edge attributes) is shown in the following table:

[0112] Table 4: Standard graph nodes

[0113] Node ID type name Property Description N001 Process Component lifting Corresponding construction task nodes N002 Restrictions Lifting radius <30m Space and equipment constraints N003 Protective measures Edge guardrail Safety measures and regulatory requirements

[0114] Table 5: Edge attributes

[0115] Edge ID Starting point End node Relationship Type Weight R001 N001 N002 Must meet 0.9 R002 N001 N003 Required 0.95

[0116] In the standard graph formats of Tables 4 and 5, the edge relationship type will be used to dynamically trigger rule determination in the simulation; the node attributes will be dynamically bound to the construction element database to achieve accurate mapping and comparison.

[0117] The graph construction module implements a fully automated conversion process from unstructured specifications to computable graphs through language parsing, graph structure modeling, multi-dimensional rule scoring, and standard format encapsulation. This provides the subsequent simulation review module with a knowledge-driven model with contextual logic, rule weights, and multi-dimensional associations. This enables the system to automatically perform "graph-to-behavior" comparisons and complete dynamic compliance assessments throughout the entire process without human intervention.

[0118] The simulation review module automatically calls the specification map through the simulation engine to perform real-time comparison of component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set, which will be sent to the decision output module.

[0119] As a core component of the AI-driven construction drawing compliance review system, the simulation review module is responsible for comparing the construction simulation process with the construction process specification map in real time, thereby accurately identifying anomalies or violations during the construction process. Its main function is to achieve "automatic supervision" of the construction process by comparing construction behavior simulation with standardized specifications, especially providing real-time review support for spatial layout, construction sequence, safety regulations, and other aspects.

[0120] First, during the simulation initialization phase, the system uses a simulation engine (such as Unity3D or AnyLogic) to drive the dynamic evolution of construction components in three-dimensional space and time based on construction drawings, BIM models, and a construction element database. This generates a component behavior sequence B_{t}, representing the spatial state and behavior of the component at each time step. Each component state is represented by a spatial envelope G_{i}(t).

[0121] During the simulation, the system calls the specification graph generated by the graph construction module in real time to match the behavior of the current component with the rules in the graph. Taking spatial conflict as an example, the system compares it based on the following spatial conflict judgment function: Among them, G i (t) and G j (t) are the geometric space envelopes of components i and j at time t, Vol(*) represents the volume calculation function (such as no more than 5% volume intersection), and δ is the conflict threshold. ij When (t)=1, the system determines that there is currently an illegal spatial conflict, such as overlapping component stacking, interlaced lifting paths, etc.

[0122] Next, the simulation review module further verifies whether the time logic of the construction process satisfies the process dependencies. This process uses the following timing constraint judgment formula:

[0123] Here, Denote the simulation start time of process i. Denote the completion time of process j. When F ij (t) = 0, it indicates that a non-compliant phenomenon of "the process under construction has not been completed yet, and the subsequent process has already started" is detected. For example, hoisting the upper component before the concrete has hardened, which is a typical "construction sequence violation".

[0124] In the dynamic simulation evolution, the simulation engine advances the behavior simulation of each component based on time. The system automatically compares the current state with the atlas rules such as spatial specifications and process sequences at each time step. Once a violation is detected, the behavior node is marked as a non-compliant process, and a violation event is generated. This event set records detailed component ID, timestamp, violation type, involved objects, violation rule number, and severity level. For example, if component L-L3201 has an unacceptable spatial intersection with hoisting equipment D01 at May 8, 2025, 10:13:21, the system will record the following information:

[0125] {

[0126] "component_id": "Beam-L3201",

[0127] "timestamp": "2025-05-08T10:13:2",

[0128] "violation_type": "Spatial conflict",

[0129] "conflict_with": "Hoisting equipment-D01",

[0130] "location": {"x": 125.5, "y": 34.2, "z": 9.8},

[0131] "rule_violated": "Hoisting radius is less than the equipment rotation radius",

[0132] "severity_level": "Medium",

[0133] "related_rule_id": "R203-A"

[0134] }

[0135] Associate the above events with relevant components and time nodes, write them into the violation event set uniformly, and finally send this clustered data structure to the decision output module.

[0136] In this way, the simulation review module achieves a complete closed loop, from component behavior modeling and dynamic call-out of specification diagrams to simulation comparison and violation marking. This significantly addresses the previous pain points of relying on manual comparison of construction drawings and neglecting the relationship between construction timing and spatial evolution. It achieves a "dynamic, real-time, and structured" upgrade for construction drawing review, providing solid support for digital construction management.

[0137] The decision output module automatically traces back and generates the violation causal path based on the time point and spatial location of the violation event set, and then outputs the recommended decision.

[0138] The decision output module is the terminal processing core of the AI-driven construction drawing compliance review system. Its function is to perform intelligent analysis and causal reasoning on the violation event set generated by the simulation review module, automatically identify the causes of violations, and propose actionable optimization suggestions or adjustment strategies. The entire process requires no human intervention and is completely driven by the system's internal algorithms, ensuring rapid response and objective consistency.

[0139] First, the received violation event set is parsed, extracting information such as component number, violation time, spatial coordinates, violation type, and violation rule number. This information is then reconstructed into an event graph with a topological structure. For example, nodes represent components and their process status; edges represent dependencies between components, spatial proximity, or concurrency within the same time period; and each edge is labeled with a causal relationship, such as "construction sequence," "spatial overlap," or "path intersection." This process can be illustrated by the following pseudocode:

[0140] for event in violation_events:

[0141] G.add_node(event.component_id, timestamp=event.timestamp, type=event.violation_type)

[0142] For related in event.related_components:

[0143] G.add_edge(event.component_id,related,relation='conflict')

[0144] This step establishes an event graph, enabling subsequent analysis to infer the causes of violations based on the temporal causal chain.

[0145] Based on the event graph structure, the system calls the time backtracking analysis algorithm to trace the impact path of each violation event and identify its causal path. This process is based on topological sorting and critical path identification methods, and recursively generates the violation causal path according to the time dimension and construction logic. Suppose a component C i When a violation occurs at time t, its causal path can be expressed as:

[0146] P i ={C i-n ,...,C i-1 ,C i},t i-n <… <t i , where: P i Represents component C i The causal path of the violation, C i-n Indicates the upstream process component that first caused the current violation, and the components meet The time dependency of n is represented by n, where n represents the number of upstream components involved in the causal path before the component violates the rule. The system generates a violation score for the causal path through the path scoring function, which is defined as follows:

[0147] Where φ is the violation score of the causal path, α j Represents component C j The influencing factors in the causal path (generated according to the criticality of the process, construction location, etc.), Risk (C j ) represents component C j The risk probability of triggering a violation in historical data. A higher score indicates that the corresponding component in the path is more likely to be the main cause of the violation.

[0148] α j It represents the impact factor of the component in the causal path (generated based on the criticality of the process, construction location, etc.). In this application, the number of process dependencies of the component in the causal path (i.e., the number of all processes that require the use of the component) and the number of component accidents (i.e., the number of accidents caused by similar components in the historical construction process) are obtained. The number of process dependencies and the number of component accidents are normalized, and the sum of the normalized number of process dependencies and the number of component accidents is used to obtain the impact factor. The larger the impact factor, the greater the importance of the component in the current causal path.

[0149] The component historical risk probability is used to characterize the frequency and severity of a component's violation triggering in historical data. It is an important basis for the decision output module to perform path scoring and priority tracing. Its calculation formula is as follows: Parameter definition: Risk(C j ) represents component C j The risk probability of triggering a violation in historical data, Vij Represents component C j The number of occurrences of type i violations in historical cases (such as violation of time sequence, spatial conflict, insufficient safety clearance, etc.), T j Represents component C j The total number of times a component is included in historical case statistics—that is, the total number of times it was constructed or simulated—is used for normalization, quantifying its historical violations as a risk probability indicator. This allows the system to statistically assess a component's propensity for violations. During causal tracing, this probability is multiplied by the impact factor to determine the component's contribution to the violation along the path, enabling the system to identify the primary source of the violation.

[0150] Once the causal path for the violation is established, the system automatically generates a mitigation plan based on the rule entries in the code map and the AI ​​engine's rule mapping logic engine. For example, if the violation is caused by a path intersection, the system will recommend adjusting the construction sequence or changing the construction path; if the violation is caused by spatial overlap, the system will automatically recommend adjusting component stacking points or optimizing the lifting sequence; if the violation is triggered by a code provision (such as insufficient clearance), the system will suggest re-arranging the site layout.

[0151] Ultimately, the system outputs all recommendations to the BIM model platform, highlighting offending nodes and visualizing paths in a 3D view. Each recommendation is tied to a time point, component number, and optimization target, facilitating subsequent system resimulation or automated modification of the construction plan. For example, construction workers can see a red-marked offending component within the BIM interface. Clicking on it will prompt a prompt, "Recommend adjusting the hoisting time to avoid intersecting paths," enhancing intuitiveness and execution efficiency.

[0152] In summary, the decision-making output module realizes a complete chain from data structure analysis, cause-effect tracing, automatic optimization generation to 3D visualization output. It solves the defects of traditional compliance review, such as vague violation positioning, lack of systematic adjustment suggestions and reliance on human experience, and greatly improves the intelligence level and implementation efficiency of construction plans.

[0153] Example 3: Please refer to Figure 3 As shown, this embodiment provides an AI-driven construction drawing compliance review method, which includes the following steps:

[0154] Based on the input construction drawings and BIM models, the system automatically identifies various components and materials, their spatial locations, and their relationships, thereby establishing a preliminary database of construction elements. Based on this, the AI ​​engine utilizes a construction logic reasoning model to automatically generate a construction sequence diagram, including component installation sequence, process dependencies, and critical path annotations. Simultaneously, it generates simulation scenario parameters such as material handling routes, storage yard planning, and equipment lifting areas, laying the data foundation for simulation modeling.

[0155] Based on the project type, construction phase, and region, matching construction process specifications and standards are automatically loaded, including rules for component stacking density, safe spacing, construction clearance, and operating time windows. These rules are then converted into a standardized graph format that can be parsed by the rule engine. This rule graph not only has a structural hierarchy but also establishes semantic mapping relationships with spatial areas and construction equipment in the construction drawings, forming constraint boundaries and verification logic.

[0156] The simulation engine activates, using a timeline to drive the entire component process from arrival to handling, hoisting, and installation. At each moment, the specification diagram is automatically called upon to perform a real-time comparison between component behavior and spatial status. During the simulation, if a component's hoisting path conflicts with other components, the yard location exceeds a safe distance, or the construction sequence violates logical dependencies, the corresponding process node is automatically marked, forming a violation event set.

[0157] Based on the time and spatial location of the violation, the system automatically traces back and generates a causal path, clearly identifying the cause of the violation, such as improper construction sequence, overlapping path planning, or unmet regulatory constraints. The system then outputs recommended changes to the process or layout parameters. This backtracking mechanism not only identifies issues but also assists in optimizing solutions.

[0158] All review results are output in a structured form. The system automatically highlights the violation locations and pastes them back into the original BIM model and construction drawing view. At the same time, it generates a comprehensive review report containing construction node information, source of violation rules, and risk level.

[0159] This approach enables a comprehensive review process, from drawings to dynamic construction logic and then to specification comparison. Data from each stage is closely linked to the next, ensuring the integrity and accuracy of the review logic. The entire process requires no human intervention, relying entirely on the AI ​​engine's modeling, simulation, and comparison capabilities, significantly improving the efficiency and depth of construction drawing compliance reviews.

[0160] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0161] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An AI-driven construction drawing compliance review system, characterized by: It includes initialization module, graph construction module, simulation review module and decision output module; Initialization module: Establishes a construction element database based on the input construction drawings and BIM models, uses the AI ​​engine to automatically generate a construction sequence diagram, and simultaneously generates simulation scene parameters. The construction element database, construction sequence diagram, and simulation scene parameters are sent to the simulation review module; Atlas construction module: automatically loads the matching construction process specifications and process standards to build a specification atlas, converts the specification atlas into a standardized atlas format, and sends the specification atlas to the simulation review module; Simulation review module: The simulation engine automatically calls the specification map to perform real-time comparison of component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set, which will be sent to the decision output module; Decision output module: Based on the time points and spatial locations of the violation event set, it automatically traces back and generates the violation causal path, and then outputs the recommended decision.

2. The AI-driven construction drawing compliance review system according to claim 1, characterized in that: The simulation review module drives the dynamic evolution of construction components in three-dimensional space and time through a simulation engine based on construction drawings, BIM models, and a construction element database, generating a component behavior sequence. The spatial state and behavior of the component at each time step in the sequence are represented by a spatial envelope box. During the simulation, the real-time specification map is used to match the behavior of the current component with the rules in the specification map. The spatial conflict judgment function is used to determine whether there is a spatial conflict between components, and the process timing constraint judgment function is used to determine whether the component construction meets the dependency order.

3. The AI-driven construction drawing compliance review system according to claim 2, characterized in that: The decision output module extracts the component number, violation time point, spatial coordinates, violation type, and violation rule number information, and reconstructs it into an event graph with a topological structure; Based on the event graph structure, a time-backtracking analysis algorithm is used to trace the impact path of each violation event, identify the causal path, and generate a violation score for the causal path using a path scoring function. Once the causal path of the violation is established, an avoidance plan is automatically generated based on the rule entries in the specification map and combined with the rule mapping logic engine of the AI ​​engine.

4. The AI-driven construction drawing compliance review system according to claim 3, characterized in that: The decision output module generates a violation score for the causal path through a path scoring function, and the calculation expression is: Where: φ is the violation score of the violation causal path, α j represents the impact factor of component j in the causal path, Risk(C j ) represents component C j The risk probability of triggering a violation in historical data, n represents the number of upstream components involved in the causal path before the component violates the rule.

5. The AI-driven construction drawing compliance review system according to claim 4 is characterized by: The decision output module calls the time backtracking analysis algorithm to trace back the impact path of each violation event and identify its causal path. Based on the topological sorting and path identification method, the violation causal path is recursively generated according to the time dimension and construction logic. Suppose a certain component C i At time t, a violation occurs and the causal path is expressed as: P i ={C i-n ,…,C i-1 ,C i },t i-n <… <t i , where: P i Represents component C i The causal path of the violation, C i-n Indicates the upstream process component that first caused the current violation, and the components meet The time dependency of n is the number of upstream components involved in the causal path before the component violates the rule. represents the simulation start time of process i, represents the completion time of process j.

6. The AI-driven construction drawing compliance review system according to claim 5, characterized in that: During the dynamic simulation evolution, the simulation review module advances the behavior simulation of each component based on time, and automatically compares the current state of the behavior node with the spatial specifications and process sequence diagram rules at each time step. If a violation is detected, the behavior node is marked as a violation process and a violation event is generated. The violation event records the component ID, timestamp, violation type, involved object, violation rule number and severity level.

7. The AI-driven construction drawing compliance review system according to claim 6, characterized in that: The graph construction module loads the construction process specification text and process standard manual for the project type and stage, performs semantic analysis, and extracts elements in the standard, including process name, construction conditions, equipment requirements, and space restrictions; The extraction results are converted into nodes and edges in a graph structure. Nodes represent construction entities, and edges represent the process logic, spatial constraints, sequence, or dependency rules between entities.

8. The AI-driven construction drawing compliance review system according to claim 7 is characterized by: The initialization module performs semantic recognition and object analysis on the input two-dimensional construction drawings and three-dimensional BIM models, including component types, sizes, materials, positions and connection relationships; The AI ​​engine automatically constructs a construction sequence diagram based on the structural dependencies between components and the process annotations in the drawings. In the construction sequence diagram, nodes represent components and arrows represent the construction sequence dependencies. Based on the construction sequence diagram and the geometric information of the components, the scenario parameters for construction simulation are automatically generated. The scenario parameters include path planning, component entry time window, equipment layout area, and stacking location.

9. The AI-driven construction drawing compliance review system according to claim 8, characterized in that: The operation of the AI ​​engine includes the following steps: The structural dependency assignment, spatial overlap and construction process matching between components are obtained, and the structural dependency assignment, spatial overlap and construction process matching are normalized so that the value range of the structural dependency assignment, spatial overlap and construction process matching is mapped to [0,1]. The normalized value of the structural dependency assignment, the normalized value of the spatial overlap and the normalized value of the construction process matching are obtained, and the normalized value of the structural dependency assignment, the normalized value of the spatial overlap and the normalized value of the construction process matching are summed to obtain the dependency weight between components.

10. An AI-driven construction drawing compliance review method, implemented by the review system according to any one of claims 1 to 9, characterized in that: The review method includes the following steps: The review system establishes a construction element database based on the input construction drawings and BIM models, automatically generates construction sequence diagrams using the AI ​​engine, and simultaneously generates simulation scenario parameters; Automatically load the matching construction process specifications and process standards to build a specification map, and convert the specification map into a standardized map format; The simulation engine automatically calls the specification map to perform real-time comparison between component behavior and spatial status. If there are any construction anomalies during the simulation process, the corresponding process nodes will be automatically marked to form a violation event set. Based on the time points and spatial locations of the violation event set, the violation causal path is automatically traced back and the recommended decision is output.

Images