Building component and transportation path generation method based on bim and ai algorithm

Abstract

The application provides a building component and transportation path generation method based on BIM and AI algorithms, relates to the technical field of big data processing, and comprises the following steps: S1, acquiring a BIM model of a current construction scene to form a building component full information model; S2, performing space unit division and installation main channel identification on a current floor plane according to the BIM model, analyzing the spatial dependency relationship between each building component, and generating an installation main sequence according to the spatial dependency relationship; S3, taking the installation main sequence as input, combining the physical properties of each building component in the building component full information model and the constraint conditions of available transportation vehicles in the current construction scene, and generating an initial transportation path; and S4, performing transportation efficiency evaluation on the transportation path, calculating the path coordination degree between each building component when the evaluation result does not satisfy a preset condition, clustering the building components according to the path coordination degree, adjusting the space unit division based on the clustering result, and regenerating the installation main sequence and the transportation path based on the adjusted space unit division.

Description

Technical Field

[0001] This invention relates to the field of big data processing technology, specifically to a method for generating building components and transportation routes based on BIM and AI algorithms. Background Technology

[0002] In prefabricated building construction, building components go through multiple stages from the factory to the installation location, including loading, transportation, stacking, and hoisting. Each stage requires planning corresponding transportation and hoisting routes for the components. Current technology typically involves construction technicians performing spatial analysis based on BIM models and manually determining the installation sequence of each component according to empirical rules such as inside-out, and then arranging transportation batches and hoisting routes accordingly.

[0003] This human-based judgment model has the following significant limitations: While the spatial unit division of a building is fixed according to functional areas during the design phase, the number of components in each functional area often does not match the capacity of transport vehicles. This results in components loaded in one vehicle being scattered across multiple functional areas, or components from one functional area needing to be split across multiple vehicles for transport. This fragmentation directly causes cross-congestion in the on-site path network, with tower crane operation trajectories repeatedly shuttling back and forth, leading to low path utilization. Furthermore, existing path planning methods treat spatial unit division as a fixed input parameter, only able to find a relatively optimal path solution under these given conditions. When the spatial division itself is unreasonable, preventing path optimization, existing methods cannot identify and correct the defects at the input end, and can only passively accept suboptimal results. Summary of the Invention

[0004] The purpose of this invention is to provide a method for generating building components and transportation routes based on BIM and AI algorithms. The method initializes the transportation routes with the physical properties of the building components and the constraints of available transportation vehicles, establishes a transportation efficiency evaluation mechanism, and dynamically updates the transportation routes.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following solution: A method for generating building components and transportation routes based on BIM and AI algorithms includes the following steps: S1. Obtain the BIM model of the current construction scene, extract the geometric coordinates and spatial connection relationships of the building components, and bind them with the corresponding physical attributes to form a full information model of the building components. S2. Based on the BIM model, divide the current floor plan into spatial units and identify the main installation passages, analyze the spatial dependencies between various building components, and generate the main installation sequence accordingly. S3. Taking the main installation sequence as input, and combining the physical properties of each building component in the full information model of building components with the constraints of available transport vehicles in the current construction scenario, generate the initial transport path and the corresponding hoisting path for each building component; where the transport path refers to the set of building components loaded on each transport vehicle and the loading order of the building components in the transport vehicle. S4. Evaluate the transportation efficiency of the transportation route. When the evaluation results do not meet the preset conditions, calculate the path coordination degree between each building component and cluster the building components accordingly. Adjust the spatial unit division based on the clustering results. Based on the adjusted spatial unit division, regenerate the main installation sequence and transportation route. Among them, path coordination degree is determined based on the spatial overlap between the hoisting paths of each building component.

[0006] Further physical attributes include: outer dimensions, weight, three-dimensional coordinates of the center of gravity, permissible transport posture, maximum number of stacked layers, and lifting point location.

[0007] Furthermore, the process of dividing the space units is as follows: based on the BIM model, the boundary information of the walls and columns in the current floor plan is obtained, the area enclosed by the walls and / or columns is automatically identified, and functional attributes are assigned to each area in combination with IFC spatial semantic tags to form multiple space units; the process of identifying the main installation passage is as follows: based on the BIM model, the relative relationship between the boundary openings of each space unit and the position of the tower crane is obtained, the hoisting entrance is automatically identified, and the main installation passage is constructed accordingly.

[0008] Furthermore, the process of generating the main installation sequence is as follows: using the shortest topological step from the hoisting entrance to each building component as the spatial depth value, determining the installation sequence constraints based on the relationship between the spatial depth values ​​of each building component, and constructing a directed acyclic graph; then performing reverse topological sorting to generate the main installation sequence. The installation sequence is based on the spatial depth value from largest to smallest. Among building components with the same spatial depth value, those located on both sides of the main installation channel are arranged first, and those with the smallest spatial depth value are arranged last. In the directed acyclic graph, the nodes are building components, and the directed edges point from building components with larger spatial depth values ​​to building components with smaller spatial depth values.

[0009] Furthermore, the constraints on available transport vehicles include the number of available transport vehicles, the maximum rated load of each available transport vehicle, and the available volume inside its cargo compartment; while the loading sequence of building components inside the transport vehicle is the reverse of the arrangement sequence of the corresponding building components in the main installation sequence.

[0010] Furthermore, the process of generating the initial transportation route is as follows: based on the main installation sequence, according to the physical properties of the building components and the constraints of available transport vehicles, each building component is divided into multiple transportation batches, and the loading sequence of the building components within each transportation batch is determined to obtain the initial transportation route.

[0011] Furthermore, the process of evaluating the transportation efficiency of the transportation route is as follows: assessing the number of different spatial units to which the building components belong in each available transportation vehicle, and simulating the density of intersections of all hoisting routes; the preset condition is that the number of different spatial units to which the building components belong in each available transportation vehicle does not exceed a first threshold, and the density of intersections of the route segments does not exceed a second threshold.

[0012] Furthermore, the process of calculating the path coordination degree between various building components is as follows: the spatial overlap between the hoisting paths of various building components is calculated by weighting the AI ​​model; the AI ​​model uses the actual execution data of the hoisting paths of building components in historical construction projects as training samples to learn the weight parameters of various factors affecting the path coordination degree.

[0013] Furthermore, the process of adjusting the spatial unit division based on the clustering results is as follows: obtain the clustering results that group building components with high path coordination into the same component group. If a component group in the clustering results contains building components of two spatially adjacent spatial units, and the two spatial units each have a hoisting entrance and are connected to the same main installation channel, then the two spatial units are merged into one spatial unit. If the building components of a spatial unit in the clustering results are split into two component groups, then the spatial unit is split into two spatial units.

[0014] The beneficial effects of this invention are: This invention provides a method for generating building components and transportation routes based on BIM and AI algorithms. Compared with existing route planning methods, it proposes a dynamic update of transportation routes: First, the geometric information of the components is extracted from the BIM model and bound to physical attributes to form a full-information model. Simultaneously, the floor plan is divided into spatial units and the main installation channel is identified. Based on the spatial dependencies between components, the main installation sequence is generated. Then, using the main installation sequence as input, combined with the physical attributes of the components and the constraints of available transport vehicles, an initial transportation route and the hoisting route for each component are generated. Subsequently, the transportation route is evaluated for efficiency. Evaluation indicators include the number of different spatial units involved in the components loaded by each transport vehicle, and the density of intersections of the route segments after simulating all hoisting routes. When the evaluation results do not meet the preset conditions, the spatial overlap between the hoisting routes of each component is calculated as the route synergy, and the components are re-clustered accordingly. Based on the clustering results, the spatial unit division is adjusted, and then the main installation sequence and transportation route are regenerated based on the adjusted spatial division.

[0015] As can be seen, this method establishes a feedback link between transportation efficiency assessment and spatial division adjustment, enabling transportation route generation to move beyond passive, one-way calculations based on predetermined spatial divisions. Instead, it proactively identifies and adjusts unreasonable spatial groupings based on quantitative feedback from transportation efficiency. After adjustment, component groupings tend to correspond one-to-one with transportation batches, with components loaded on each transport vehicle concentrated within the same spatial area, fundamentally reducing path intersections and secondary handling caused by dispersed component loading. Simultaneously, the generation of the main installation sequence is automatically completed based on spatial depth values ​​and inverse topological sorting, eliminating reliance on manual experience and improving the efficiency and consistency of construction organization design. The component full-information model unifies physical attributes such as external dimensions, weight, and permissible transport posture with geometric information, providing a precise data foundation for transportation route generation and efficiency assessment, ensuring the accuracy and executability of the calculation results. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the method in Embodiment 1 of the present invention.

[0017] Figure 2 This is a schematic diagram of the BIM model in Embodiment 1 of the present invention.

[0018] Figure 3 This is a schematic diagram of the spatial units obtained in Embodiment 1 of the present invention.

[0019] Figure 4 This is a first schematic diagram of the installation between the main channel and the building components in Embodiment 1 of the present invention.

[0020] Figure 5 This is a second schematic diagram of the installation between the main channel and the building components in Embodiment 1 of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0023] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0024] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0025] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0026] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0027] The present invention will now be described in detail with reference to the accompanying drawings and embodiments: Example 1 In this embodiment, a method for generating building components and transportation routes based on BIM and AI algorithms is proposed. The method is as follows: Figure 1 As shown, it includes the following steps: S1. Obtain the BIM model of the current construction scene, extract the geometric coordinates and spatial connection relationships from it, and bind them with the corresponding physical attributes to form a full information model of building components. S2. Based on the BIM model, divide the current floor plan into spatial units and identify the main installation passages, analyze the spatial dependencies between various building components, and generate the main installation sequence accordingly. S3. Taking the main installation sequence as input, and combining the physical properties of each building component in the full information model of building components with the constraints of available transport vehicles in the current construction scenario, generate the initial transport path and the corresponding hoisting path for each building component; where the transport path refers to the set of building components loaded on each transport vehicle and the loading order of the building components in the transport vehicle. S4. Evaluate the transportation efficiency of the transportation route. When the evaluation results do not meet the preset conditions, calculate the path coordination degree between each building component and cluster the building components accordingly. Adjust the spatial unit division based on the clustering results. Based on the adjusted spatial unit division, regenerate the main installation sequence and transportation route. Among them, path coordination degree is determined based on the spatial overlap between the hoisting paths of each building component.

[0028] Based on the above principles, the present invention will be further explained and described as follows: In this embodiment, it is assumed that a tower crane is set up on the south side of the current construction site, and four flatbed trucks are available for transportation. Each truck has a maximum load capacity of 15 tons and the internal dimensions of the truck bed are 10.0 meters long, 2.5 meters wide, and 3.0 meters high.

[0029] In step S1, the BIM model of the current construction scene is obtained, and the geometric coordinates and spatial connection relationships of 10 building components are extracted. Specifically, the 10 building components are: Bedroom 1 exterior wall W1 (6 tons), Bedroom 1 interior wall W2 (3 tons), Bedroom 2 exterior wall W3 (5 tons), Bedroom 2 interior wall W4 (3 tons), Living room north wall W5 (5.5 tons), Living room south wall W6 (5.5 tons), corridor partition wall W7 (2.5 tons), Bedroom 1 floor slab S1 (2 tons), Bedroom 2 floor slab S2 (2 tons), and Living room floor slab S3 (3.5 tons), with a total weight of 38 tons. Taking Bedroom 1 exterior wall W1 as an example, its design dimensions can be read from the BIM model as 3 meters long, 2.8 meters high, and 0.2 meters thick, with installation coordinates of (2, 8, 6). It is connected to the corridor through the south doorway, and its corresponding physical attributes are bound to form a full information model of the building component.

[0030] The physical attributes include: outer dimensions, weight, three-dimensional coordinates of the center of gravity, permissible transport posture, maximum number of stacking layers, and location of lifting points. Specifically, taking the exterior wall W1 of bedroom 1 as an example, the physical attributes include: outer dimensions of 3.2m × 2.85m × 0.28m (including 4 top hanging nails and pre-embedded junction boxes), measured weight of 6 tons, three-dimensional coordinates of the center of gravity (1.6, 1.4, 0.14), permissible vertical transport with an inclination angle not exceeding 5 degrees, not allowed to be stacked, 4 lifting points located at the top four corners, and a permissible load of 2.5 tons per point.

[0031] Preferably, in step S2, the spatial unit division process is as follows: Figure 2 As shown, the boundary information of walls and columns within the current floor plan can be obtained in the BIM model, the area enclosed by walls and / or columns can be automatically identified, and functional attributes can be assigned to each area in combination with IFC spatial semantic tags to form multiple spatial units; the process of identifying the main installation passage is as follows: the relative relationship between the boundary opening of each spatial unit and the position of the tower crane is obtained according to the BIM model, the hoisting entrance is automatically identified, and the main installation passage is constructed accordingly.

[0032] Specifically, the process of dividing the space into units is as follows: Based on the BIM model, the boundary information of the walls and columns within the current floor plan is obtained; the areas enclosed by the walls and columns are automatically identified, resulting in four areas; for example... Figure 3As shown, combined with IFC spatial semantic tags, functional attributes are assigned to each area: Bedroom 1 (12 square meters), Bedroom 2 (12 square meters), Living Room (24 square meters), and Corridor (8 square meters), forming four spatial units. The process of identifying the main installation passage is as follows: The tower crane is located on the south side of the building. Based on the BIM model, the relative relationship between the boundary openings of each spatial unit and the tower crane position is obtained, and the hoisting entrance is automatically identified: The doorway on the south side of Bedroom 1 is 1.2 meters wide, facing the tower crane direction, and has no obstructions during penetration detection, thus being marked as the hoisting entrance of Bedroom 1; Here, taking Bedroom 1 as an example, a component topology reachability graph is constructed starting from the hoisting entrance. After merging high-frequency path segments, the main installation passage is formed, with a width of 1.8 meters, extending from the hoisting entrance to the installation position of the deepest part of the Bedroom 1 exterior wall W1, and the main installation passage is constructed accordingly, as shown below. Figure 4 and Figure 5 As shown. The tower crane direction is in... Figure 4 and Figure 5 The diagram uses a triangular marker.

[0033] It should be noted that in actual construction scenarios, there may be multiple spatial units with hoisting entrances on the same floor, corresponding to multiple candidate main installation channels. To avoid manual selection leading to deviations from the optimal initial solution, this invention automatically determines the main installation channel through the following strategy: (1) First, calculate the number of building components served by each candidate channel (i.e., the total number of components that can be reached topologically from the entrance of the channel), and select the one with the largest number as the main installation channel; (2) If the number is the same, compare the total weight of the components served by each candidate channel and select the one with the largest weight. The reason is that the heavy-load area has higher requirements for the accuracy of path planning. (3) If they are still in parallel, select the candidate channel with the smallest spatial depth value variance so that the channel covers the components with the most balanced depth and avoids path planning mismatch caused by local excessive depth or shallowness.

[0034] In this embodiment, the tower crane is located on the south side of the building, and both bedrooms 1 and 2 face the crane's hoisting entrance, forming two candidate main installation channels. To avoid arbitrary manual designation, this invention automatically determines the main installation channels according to the above strategy: the building components served by the candidate main installation channel of bedroom 1 are W1, W2, and S1, with a total weight of 6+3+2=11 tons; the building components served by the candidate main installation channel of bedroom 2 are W3, W4, and S2, with a total weight of 5+3+2=10 tons. Since 11 tons > 10 tons, the system automatically selects the candidate main installation channel of bedroom 1 as the main installation channel, as follows. Figure 4 and Figure 5 As shown, no manual intervention is required throughout the entire process. This also demonstrates the automated decision-making capability of this method in the initial channel selection stage.

[0035] Preferably, in step S2, the process of generating the main installation sequence is as follows: using the shortest topological step from the hoisting entrance to each building component as the spatial depth value, determining the installation sequence constraints based on the relationship between the spatial depth values ​​of each building component, and constructing a directed acyclic graph; performing reverse topological sorting to generate the main installation sequence; The installation sequence is based on the spatial depth value from largest to smallest. Among building components with the same spatial depth value, those located on both sides of the main installation channel are arranged first, and those with the smallest spatial depth value are arranged last. In the directed acyclic graph, the nodes are building components, and the directed edges point from building components with larger spatial depth values ​​to building components with smaller spatial depth values.

[0036] Specifically, the installation sequence follows an inside-out logic, starting from the deepest part of each spatial unit, furthest from the hoisting entrance, and gradually progressing towards the entrance. It should be noted that, taking bedroom 1 as an example, W1 refers to the wall panel inside bedroom 1 furthest from the hoisting entrance; W2 refers to the wall panel located between W1 and the hoisting entrance, relatively closer to the entrance. For each spatial unit, the wall panel at the deepest part of the space is called the exterior wall, and the wall panel closer to the entrance is called the interior wall.

[0037] Based on the above principles, the process of generating the main installation sequence is as follows: Figure 4 and Figure 5 For example, the shortest topological step from the hoisting entrance to each building component is used as the spatial depth value. W1 needs to cross a virtual inner wall dividing line, which is counted as 2 steps; W2 is directly reachable, which is counted as 1 step; and S1 is located above the entrance, which is counted as 0 steps. The installation order constraints are determined based on the relationship between the spatial depth values ​​of each building component to construct a directed acyclic graph (DAG). Inverse topological sorting is performed to generate the main installation order. In the DAG, the nodes are building components, and the directed edges point from building components with larger spatial depth values ​​to building components with smaller spatial depth values, such as W1 (2 steps) → W2 (1 step), W1 → S1 (0 steps), and W2 → S1. A depth-first search is used to backtrack from S1. During the backtracking phase, nodes are pushed onto a stack, and the order of popping nodes from the stack is the main installation order of bedroom 1: W1 → W2 → S1.

[0038] Similarly, in bedroom 2, W3 is the outer wall panel furthest from the hoisting entrance, and W4 is the inner wall panel closest to the entrance, so the main installation sequence is W3→W4→S2. In the living room, W5 is the north wall (furthest from the entrance), and W6 is the south wall (closest to the entrance), so the main installation sequence is W5→W6→S3. In the corridor, W7 forms a separate spatial unit, and the main installation sequence is W7. Combining these, the main installation sequence for the entire floor is: W1→W2→W3→W4→W7→W5→W6→S1→S2→S3.

[0039] Preferably, the constraints on available transport vehicles include the number of available transport vehicles, the maximum rated load of each available transport vehicle, and the available volume inside its cargo compartment; and the loading sequence of building components inside the transport vehicle is the reverse of the arrangement sequence of the corresponding building components in the main installation sequence.

[0040] Physical attributes are used for loading planning and transportation route generation. Specifically, based on the weight and outer dimensions of each building component, combined with the rated load capacity and cargo volume of the transport vehicles, it determines how to allocate components to different vehicles (i.e., divide them into transport batches) and the loading order of components within the cargo compartment (usually the reverse of the installation order). Furthermore, the allowed transport posture and maximum stacking layer are used to determine whether components can be laid flat, upright, or stacked to optimize cargo compartment space utilization; the three-dimensional coordinates of the center of gravity and the location of the lifting points are used for hoisting route planning to ensure stability during lifting and transportation.

[0041] Preferably, the process of generating the initial transportation route is as follows: based on the main installation sequence, according to the physical properties of the building components and the constraints of available transportation vehicles, each building component is divided into multiple transportation batches, and the loading sequence of the building components within each transportation batch is determined to obtain the initial transportation route.

[0042] Specifically, in this embodiment, four flatbed trucks are available for transport, each with a maximum load capacity of 15 tons and an internal volume of approximately 75 cubic meters. The process of generating the initial transport route is as follows: based on the main installation sequence, according to the physical properties of the building components and the constraints of the available transport vehicles, each building component is divided into multiple transport batches, and the loading order of the building components within each transport batch is determined to obtain the initial transport route. The specific division results are as follows: vehicle 1 loads W2 and W1 (9 tons), vehicle 2 loads W4 and W3 (8 tons), vehicle 3 loads W6, W5, and W7 (13.5 tons), and vehicle 4 loads S3, S2, and S1 (7.5 tons). The loading order of the building components in the transport vehicles is the reverse of the arrangement order of the corresponding building components in the main installation sequence. Taking vehicle 1 as an example, W2 is loaded first and placed deep in the truck bed, while W1 is loaded later and placed at the rear of the truck bed. When unloading, W1 is unloaded first and W2 is unloaded last, which is the reverse of the installation order W1→W2 of this batch. At the same time, the hoisting path for each building component is generated from the unloading point to the installation position.

[0043] Preferably, the process of evaluating the transportation efficiency of the transportation route is as follows: evaluating the number of different spatial units to which the building components belong in each available transportation vehicle, and simulating the density of intersections of all hoisting routes; the preset condition is that the number of different spatial units to which the building components belong in each available transportation vehicle does not exceed a first threshold, and the density of intersections of the route segments does not exceed a second threshold.

[0044] In step S4, the process of evaluating the transportation efficiency of the transportation routes involves assessing the number of different spatial units to which the building components belong in each available transportation vehicle, and simulating the density of intersections of all hoisting routes. The assessment shows that vehicle 1 carries W1 and W2, both belonging to bedroom 1, involving one spatial unit; vehicle 2 carries W3 and W4, both belonging to bedroom 2, involving one spatial unit; vehicle 3 carries W6 (living room), W5 (living room), and W7 (corridor), involving two spatial units (living room and corridor); and vehicle 4 carries S3 (living room), S2 (bedroom 2), and S1 (bedroom 1), involving three spatial units. Simultaneously, after simulating all hoisting routes in the digital twin environment, the hoisting route for W7 in vehicle 3 enters from the corridor entrance, while the hoisting routes for W5 and W6 enter from the living room entrance. These two routes intersect at the junction of the corridor and the living room, resulting in an excessive density of intersections. The preset condition is that the number of different spatial units to which the building components belong in each available transportation vehicle does not exceed a first threshold (set to 1 in this embodiment), and the density of intersections of the routes does not exceed a second threshold. The number of spatial units involved in vehicles 3 and 4 both exceeded the first threshold, and the density of intersections of the path segments exceeded the second threshold. Therefore, the evaluation results did not meet the preset conditions.

[0045] In summary, this invention incorporates a performance evaluation step, using two indicators—fragmentation degree and path intersection density—to measure the rationality of transportation routes, providing an objective basis for determining whether optimization is necessary.

[0046] When the evaluation results do not meet the preset conditions, the path coordination degree between each building component is calculated, and the building components are clustered accordingly. The process of calculating the path coordination degree between each building component is as follows: the spatial overlap between the hoisting paths of each building component is calculated by weighting the results using an AI model. This AI model uses the actual execution data of the hoisting paths of building components in historical construction projects as training samples to learn the weight parameters of each factor affecting the path coordination degree. Taking the path coordination degree calculation of corridor partition wall W7 and other components as an example: the overlap of the hoisting path node sequence between W7 and bedroom 1 interior wall W2 is 85%, and the path coordination degree score after weighted calculation by the AI ​​model is 0.88; the overlap of the hoisting path node sequence between W7 and living room south wall W6 is 20%, and the path coordination degree after weighted calculation by the AI ​​model is 0.32. Building components with high path coherence are grouped into the same component group. The clustering results are as follows: W1, W2, W7, and S1 are grouped into the first component group (the path coherence is higher than 0.8), W3, W4, and S2 are grouped into the second component group, and W5, W6, and S3 are grouped into the third component group.

[0047] The introduction of path coordination is a key element enabling dynamic optimization in this method. It does not rely on empirical labels such as functional zoning, but rather measures the transport compatibility between components based on the actual spatial overlap of their hoisting paths. In this way, even if two components belong to different functional areas in their design, as long as their hoisting paths overlap significantly, transporting them on the same vehicle is reasonable.

[0048] The spatial unit division is adjusted based on clustering results. The process of adjusting the spatial unit division based on clustering results is as follows: obtain the clustering results that group building components with high path coordination into the same component group. If a component group in the clustering results contains building components of two spatially adjacent spatial units, and the two spatial units each have a hoisting entrance and are connected to the same main installation passage, then the two spatial units are merged into one spatial unit. The first component group contains the spatial units of bedroom 1 (W1, W2, S1) and the spatial unit of the corridor (W7). The two spatial units are spatially adjacent, each has a hoisting entrance and is connected to the same corridor, which meets the merging condition. Therefore, the two spatial units are merged into one spatial unit to form a collaborative transportation area.

[0049] Based on the adjusted spatial unit division, the main installation sequence and transportation path are regenerated. Returning to step S2, the main installation sequence is regenerated as follows: within the collaborative transportation area, W1→W2→W7→S1; within bedroom 2, W3→W4→S2; within the living room, W5→W6→S3. Step S3 is executed again to generate the transportation path: Vehicle A loads W1, W2, W7, and S1 (13.5 tons); Vehicle B loads W3, W4, and S2 (10 tons); and Vehicle C loads W5, W6, and S3 (14.5 tons). The total load of the three vehicles is 37.0 tons, equal to the total number of components. Vehicle D is not used. A transportation efficiency assessment is performed again. The number of spatial units involved in the components loaded by each transport vehicle is 1, and the intersection density of the path segments is reduced to below the second threshold. The assessment results meet the preset conditions, and the final transportation path is output.

[0050] The final results show that the initial transportation route used all 4 vehicles, with an average loading rate of about 61.5%, and there were fragmentation and path intersection problems; while the transportation route dynamically updated by this method only used 3 vehicles, and the components loaded by each vehicle were concentrated in the same spatial unit, which improved the vehicle loading rate.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for generating building components and transportation routes based on BIM and AI algorithms, characterized in that, The method includes the following steps: S1. Obtain the BIM model of the current construction scene, extract the geometric coordinates and spatial connection relationships of the building components, and bind them with the corresponding physical attributes to form a full information model of the building components. S2. Based on the BIM model, divide the current floor plan into spatial units and identify the main installation passages, analyze the spatial dependencies between various building components, and generate the main installation sequence accordingly. S3. Taking the main installation sequence as input, and combining the physical properties of each building component in the full information model of building components with the constraints of available transport vehicles in the current construction scenario, generate the initial transport path and the corresponding hoisting path for each building component; where the transport path refers to the set of building components loaded on each transport vehicle and the loading order of the building components in the transport vehicle. S4. Evaluate the transportation efficiency of the transportation route. When the evaluation results do not meet the preset conditions, calculate the path coordination degree between each building component and cluster the building components accordingly. Adjust the spatial unit division based on the clustering results. Based on the adjusted spatial unit division, regenerate the main installation sequence and transportation route. The process of evaluating the transportation efficiency of the transportation route is as follows: assess the number of different spatial units to which the building components belong in each available transportation vehicle, and simulate the density of intersections of all hoisting routes; the preset condition is that the number of different spatial units to which the building components belong in each available transportation vehicle does not exceed a first threshold, and the density of intersections of the route segments does not exceed a second threshold. Among them, path coordination degree is determined based on the spatial overlap between the hoisting paths of each building component.

2. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 1, characterized in that, Physical properties include: outer dimensions, weight, three-dimensional coordinates of the center of gravity, permissible transport posture, maximum number of stacked layers, and lifting point location.

3. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 1, characterized in that, The process of dividing space units is as follows: Based on the BIM model, the boundary information of the walls and columns in the current floor plan is obtained, the area enclosed by the walls and / or columns is automatically identified, and functional attributes are assigned to each area in combination with IFC spatial semantic tags to form multiple space units; The process of identifying the main installation passage is as follows: Based on the BIM model, the relative relationship between the boundary openings of each space unit and the position of the tower crane is obtained, the hoisting entrance is automatically identified, and the main installation passage is constructed accordingly.

4. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 3, characterized in that, The process of generating the main installation sequence is as follows: the shortest topological step from the hoisting entrance to each building component is used as the spatial depth value; the installation sequence constraints are determined according to the relationship between the spatial depth values ​​of each building component to construct a directed acyclic graph; and the reverse topological sorting is performed to generate the main installation sequence. The installation sequence is based on the spatial depth value from largest to smallest. Among building components with the same spatial depth value, those located on both sides of the main installation channel are arranged first, and those with the smallest spatial depth value are arranged last. In the directed acyclic graph, the nodes are building components, and the directed edges point from building components with larger spatial depth values ​​to building components with smaller spatial depth values.

5. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 1, characterized in that, The constraints on available transport vehicles include the number of available transport vehicles, the maximum rated load of each available transport vehicle, and the available volume inside its cargo compartment; while the loading sequence of building components inside the transport vehicle is the reverse of the arrangement sequence of the corresponding building components in the main installation sequence.

6. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 1, characterized in that, The process of generating the initial transportation route is as follows: based on the main installation sequence, according to the physical properties of the building components and the constraints of available transport vehicles, each building component is divided into multiple transportation batches, and the loading order of the building components within each transportation batch is determined to obtain the initial transportation route.

7. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 1, characterized in that, The process of calculating the path coordination degree between various building components is as follows: the spatial overlap between the hoisting paths of various building components is calculated by weighting the AI ​​model; the AI ​​model uses the actual execution data of the hoisting paths of building components in historical construction projects as training samples to learn the weight parameters of various factors affecting the path coordination degree.

8. The method for generating building components and transportation routes based on BIM and AI algorithms according to claim 3, characterized in that, The process of adjusting the spatial unit division based on the clustering results is as follows: obtain the clustering results that group building components with high path coherence into the same component group. If a component group in the clustering results contains building components of two spatially adjacent spatial units, and the two spatial units each have a hoisting entrance and are connected to the same main installation channel, then the two spatial units are merged into one spatial unit. If the building components of a spatial unit are split into two component groups in the clustering results, then the spatial unit is split into two spatial units.

Images