An AI algorithm-based building construction material transportation path optimization and scheduling method
Patent Information
- Application Number
- CN202610970206.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-01
- Publication Date
- 2026-09-15
Smart Images
Figure CN122759518A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of construction material transportation route optimization and scheduling technology, specifically to a construction material transportation route optimization and scheduling method based on AI algorithms. Background Technology
[0002] In the construction process, the timely and efficient transportation of construction materials is crucial for ensuring construction progress and controlling costs. Construction sites are typically complex in layout, with dispersed work areas, numerous material demand points, and dynamically changing material requirements at different times. Existing transportation route scheduling largely relies on the experience of dispatchers, making it difficult to adjust transportation plans in a timely manner based on real-time road conditions and changes in construction progress. This easily leads to problems such as duplicate transportation routes, excessive vehicle waiting times, and delays in construction due to untimely material delivery, while also increasing transportation energy consumption and costs. Most existing route scheduling algorithms are designed for general logistics transportation scenarios and are not adapted to the characteristics of narrow roads, variable road occupancy for temporary operations, and significant differences in the priority of material transportation needs within construction sites, thus failing to meet the actual needs of construction material transportation scheduling.
[0003] To address this, a method for optimizing and scheduling the transportation routes of building materials based on AI algorithms is proposed. Summary of the Invention
[0004] The purpose of this invention is to provide a method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms. By establishing a transportation route scheduling scheme map, the method adjusts the scheduling scheme to improve the adaptability and accuracy of transportation scenarios.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] A method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms includes:
[0007] Obtain historical scheduling records of construction material transportation information, and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least the material type, material weight, supply location, and required delivery time.
[0008] Optionally, semantic feature information of the construction material transportation information is extracted based on a semantic recognition model, and multiple semantic clusters are obtained by clustering the semantic feature information of the construction material transportation information based on a clustering model.
[0009] Obtain the cluster radius of multiple semantic clusters, sort the semantic clusters according to the size of the cluster radius to obtain sorted semantic clusters; obtain the minimum distance between the cluster center and the semantic feature information of each semantic cluster according to the Euclidean distance algorithm, and use the semantic feature information of the minimum distance as the semantic feature label of each semantic cluster.
[0010] The ratio of the radius of each semantic cluster to the sum of the radii of all clusters is used as the transportation association weight. The transportation association weight is weighted according to the sorting order as the weighting value to obtain the initial association weight of each semantic cluster.
[0011] The semantic feature labels of each semantic cluster are used as transportation directory terms, and the initial association weights are associated with each transportation directory term to obtain the transportation directory. This method can assign reasonable association weights to different types of transportation needs by extracting features and clustering historical transportation information. When generating the transportation directory, it retains the core features of different transportation tasks and can distinguish the priority and impact of different needs through weights. This provides accurate and clear basic data for subsequent route scheduling, avoids irrelevant transportation information from interfering with the scheduling results, and effectively improves the accuracy and efficiency of subsequent route planning and scheduling.
[0012] Based on the historical scheduling records, obtain the historical status information of all transport vehicles. Based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory to obtain a preliminary association index. Obtain the path scheme data of the historical scheduling records. Based on the path scheme data, optimize the association weight of the preliminary association index to obtain a path optimization map.
[0013] Optionally, historical status information of all dispatched transport vehicles can be extracted based on historical dispatch records. The historical status information includes at least vehicle type, approved load capacity, loaded weight, and vehicle usage time.
[0014] Based on the semantic recognition model, the historical state semantic features of all transport vehicles and the transport vehicle semantic features of the construction material transport information are extracted. The semantic similarity between the historical state semantic features and the transport vehicle semantic features is calculated based on the cosine similarity algorithm. The historical state information with a semantic similarity greater than the vehicle similarity threshold is associated with the initial association weight of the transport directory to obtain the preliminary association index.
[0015] The route plan data is obtained, which includes at least the transportation distance, actual fuel consumption data, delivery delay rate and transportation cost. The route plan data is then used to perform preliminary path association based on the initial path weight and the preliminary association index to obtain an initial path map.
[0016] The path optimization weights are calculated by optimizing the path scheme data using the path weight optimization formula. These optimized weights are then multiplied by the initial path weights of the initial path map to obtain the optimized path map. This method leverages historical scheduling data to accurately optimize path association weights. Compared to traditional path planning methods that rely on experience-based judgment, it effectively reduces overall transportation costs and delivery delays. Furthermore, it can match the actual status of different transport vehicles to complete targeted path allocation, improving vehicle utilization. This approach is suitable for the variable material transportation needs and dispersed transportation nodes in construction scenarios, providing data-driven optimization support for construction material transportation scheduling and ensuring a balance between material transportation efficiency and economy.
[0017] Optionally, the expression for the path weight optimization formula is as follows:
[0018] ;
[0019] in, Optimize the weights for the path. To optimize the algorithm, Where FC represents the transport distance and FC represents the actual fuel consumption data. Let C be the delivery timeout rate, C be the transportation cost, min be the minimum value, and exp be an exponential function with base e, min be the minimum value. and The weighting coefficients are used to calculate the optimal route. This method incorporates weighting coefficients to combine multiple parameters such as transportation distance, actual fuel consumption, delivery delay rate, and transportation cost. This comprehensive approach considers various key indicators affecting route selection, avoiding the bias caused by single-indicator decisions. The resulting route optimization weights better align with the actual needs of building material transportation, balancing timeliness and economy, and further improving the accuracy of route planning.
[0020] Obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the path optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the path optimization map based on the anomaly scheduling matching results to obtain a transportation path scheduling scheme map.
[0021] Optionally, construction anomaly scheduling semantic information is extracted from the historical scheduling records based on a semantic recognition model. The construction anomaly scheduling semantic information includes the cause of the anomaly, the time of the anomaly delay, and the location of the anomaly. Based on the cosine similarity algorithm, the historical status information and route plan data of all transport vehicles in the route optimization map are matched for anomaly information similarity through the cause of the anomaly and the location of the anomaly to obtain a similarity matching result. The similarity matching result greater than the anomaly threshold is selected as the anomaly weight. The construction anomaly scheduling semantic information is associated with the route optimization map according to the anomaly weight to obtain a preliminary transport route scheduling plan map.
[0022] The path optimization weights of the preliminary transportation route scheduling scheme map are dynamically corrected based on the anomaly dynamic correction function and similarity matching results to obtain the transportation route scheduling scheme map. The expression of the anomaly dynamic correction function is as follows:
[0023] ;
[0024] in, The path optimization weights are dynamically adjusted for abnormalities, where max is the maximum value and Cos is the cosine similarity algorithm. The cause of the aforementioned abnormality, This provides historical status information for all transport vehicles. For locations of abnormal routes, PA represents path plan data. Let be a logarithmic function with base e, t be the abnormal delay time, and N be the total number of abnormal delay time records in historical scheduling. This is the record of the i-th abnormal delay. This method can automatically identify abnormal characteristics based on historical scheduling data and dynamically correct the path optimization map, avoiding the problem that static path planning cannot adapt to sudden changes in the construction scenario. Simultaneously, by adjusting the path optimization weights through abnormal weight allocation, the corrected path plan better reflects the impact of abnormal events on transportation in actual construction, improving the rationality and feasibility of the final transportation path scheduling plan, reducing transportation delays caused by abnormal events, and improving the scheduling efficiency of construction material transportation.
[0025] The transportation routes for building materials are optimized based on the aforementioned transportation route scheduling scheme map.
[0026] This application also provides a construction material transportation route optimization and scheduling system based on AI algorithms, including:
[0027] The transportation catalog module is used to obtain construction material transportation information from historical scheduling records and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least material type, material weight, supply location, and required delivery time.
[0028] The route optimization graph module is used to obtain the historical status information of all transport vehicles based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory based on the historical scheduling records to obtain a preliminary association index, obtain the route plan data of the historical scheduling records, and optimize the association weight of the preliminary association index based on the route plan data to obtain the route optimization graph.
[0029] The transportation route scheduling scheme map module is used to obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the route optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the route optimization map based on the anomaly scheduling matching results to obtain the transportation route scheduling scheme map.
[0030] The transportation route optimization module optimizes the transportation routes of building materials based on the transportation route scheduling scheme map.
[0031] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0032] 1. This application generates a weighted transportation catalog by semantic clustering of historical transportation information, then constructs a path optimization graph by combining historical status of transportation vehicles and route plan data, and finally introduces historical abnormal scheduling information to dynamically correct the graph weights. The resulting transportation route scheduling scheme can fully adapt to the variable abnormal situations of material transportation in construction scenarios. Compared with the traditional method of planning routes based solely on distance cost, it effectively reduces the probability of delivery delays, reduces overall transportation costs, and improves the response efficiency and robustness of material transportation scheduling. Attached Figure Description
[0033] Figure 1 This is a flowchart of the method of the present invention.
[0034] Figure 2 This is a system flowchart of the present invention. Detailed Implementation
[0035] 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. 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.
[0036] This invention provides a method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms. The technical solution is as follows:
[0037] Example 1:
[0038] A method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms, referring to Figure 1 As shown, it includes:
[0039] Obtain historical scheduling records of construction material transportation information, and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least the material type, material weight, supply location, and required delivery time.
[0040] Optionally, semantic feature information of the construction material transportation information is extracted based on a semantic recognition model (such as a pre-trained BERT model), and multiple semantic clusters are obtained by clustering the semantic feature information of the construction material transportation information based on a clustering model (such as a DBSCAN model).
[0041] The cluster radius of multiple semantic clusters is obtained, and the semantic clusters are sorted according to the size of the cluster radius to obtain sorted semantic clusters. The minimum distance between the cluster center and the semantic feature information of each semantic cluster is obtained according to the Euclidean distance algorithm, and the semantic feature information with the minimum distance is used as the semantic feature label of each semantic cluster. The specific calculation process is as follows: first, the Euclidean distance between the coordinates of each semantic feature vector and the coordinates of the current cluster center is calculated; then, the distances between all semantic feature vectors in the semantic cluster and the cluster center are counted, and the semantic feature vector with the smallest distance value is selected, and its corresponding semantic content is determined as the semantic feature label of the semantic cluster.
[0042] The ratio of the radius of each semantic cluster to the sum of the radii of all clusters is used as the transportation association weight. This transportation association weight is then weighted according to the sorting order to obtain the initial association weight for each semantic cluster. Specifically, a median is selected based on the sorting ratio and set as the standard weighting value. For example, the standard weighting value is 1. The weighting value of semantic cluster radii larger than this median increases, while the standard weighting value of semantic cluster radii smaller than this median decreases. The specific increase and decrease values can be set based on expert experience; in one embodiment, they are increased by a fixed value. For example, 1.1, 1.2, and 1.3, etc. In another implementation, the weight can be increased according to the multiple of the previous radius and the next radius. For example, the weight of the previous radius is 1.0, and the next radius is 1.25 times the previous radius, so the weight of the next radius is 1.25. This step can extract the frequently transported building materials and carry out more refined scheduling design for these frequently transported building materials. On the one hand, it can reduce the hardware cost of separate transportation scheduling design in various scenarios. On the other hand, it can also unify the design of transportation scheduling problem of frequently transported materials.
[0043] The semantic feature labels of each semantic cluster are used as transportation directory terms, and the initial association weights are associated with each transportation directory term to obtain the transportation directory. This method can assign reasonable association weights to different types of transportation needs by extracting features and clustering historical transportation information. When generating the transportation directory, it retains the core features of different transportation tasks and can distinguish the priority and impact of different needs through weights. This provides accurate and clear basic data for subsequent route scheduling, avoids irrelevant transportation information from interfering with the scheduling results, and effectively improves the accuracy and efficiency of subsequent route planning and scheduling.
[0044] Based on the historical scheduling records, obtain the historical status information of all transport vehicles. Based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory to obtain a preliminary association index. Obtain the path scheme data of the historical scheduling records. Based on the path scheme data, optimize the association weight of the preliminary association index to obtain a path optimization map.
[0045] Optionally, historical status information of all dispatched transport vehicles can be extracted based on historical dispatch records. The historical status information includes at least vehicle type, approved load capacity, loaded weight, and vehicle usage time.
[0046] Based on a semantic recognition model (such as a pre-trained BERT model), historical state semantic features of all transport vehicles and transport vehicle semantic features of the construction material transport information are extracted. The semantic similarity between the historical state semantic features and the transport vehicle semantic features is calculated using a cosine similarity algorithm. Historical state information with semantic similarity greater than the vehicle similarity threshold is associated with the initial association weight of the transport directory to obtain the preliminary association index. The default similarity threshold is 0.8, but it can be set based on expert experience.
[0047] The route plan data is obtained, which includes at least the transportation distance, actual fuel consumption data, delivery delay rate and transportation cost. The route plan data is then used to perform preliminary path association based on the initial path weight and the preliminary association index to obtain an initial path map.
[0048] The path optimization weights are calculated by optimizing the path scheme data using the path weight optimization formula. These optimized weights are then multiplied by the initial path weights of the initial path map to obtain the optimized path map. This method leverages historical scheduling data to accurately optimize path association weights. Compared to traditional path planning methods that rely on experience-based judgment, it effectively reduces overall transportation costs and delivery delays. Furthermore, it can match the actual status of different transport vehicles to complete targeted path allocation, improving vehicle utilization. This approach is suitable for the variable material transportation needs and dispersed transportation nodes in construction scenarios, providing data-driven optimization support for construction material transportation scheduling and ensuring a balance between material transportation efficiency and economy.
[0049] Optionally, the expression for the path weight optimization formula is as follows:
[0050] ;
[0051] in, Optimize the weights for the path. To optimize the algorithm, Where FC represents the transport distance and FC represents the actual fuel consumption data. Let C be the delivery timeout rate, C be the transportation cost, min be the minimum value, and exp be an exponential function with base e, min be the minimum value. and The weighting coefficients are used to calculate the optimal route. This method incorporates weighting coefficients to combine multiple parameters such as transportation distance, actual fuel consumption, delivery delay rate, and transportation cost. This comprehensive approach considers various key indicators affecting route selection, avoiding the bias caused by single-indicator decisions. The resulting route optimization weights better align with the actual needs of building material transportation, balancing timeliness and economy, and further improving the accuracy of route planning.
[0052] To optimize the algorithm, genetic algorithms, particle swarm optimization, etc., can be used. Taking particle swarm optimization as an example, the optimization objectives are to minimize the total transportation distance, the delivery timeout rate, the transportation cost, and the fuel consumption. The process of particle swarm optimization is as follows:
[0053] The first step is to initialize the particle swarm, setting the particle swarm size, maximum number of iterations, learning factor, and inertia weight parameters, and randomly initializing the position and velocity of each particle, where the position of the particle corresponds to the candidate transport path encoding.
[0054] The second step is to calculate the fitness value of each particle according to the aforementioned path optimization weight formula, and record the individual optimal position of each particle and the global optimal position of the entire particle swarm.
[0055] The third step is to iteratively update the velocity and position of each particle, recalculate the fitness value of the updated particles, and update the individual optimal position and the global optimal position respectively.
[0056] The fourth step is to determine whether the maximum number of iterations has been reached or whether the global optimal fitness value meets the preset convergence threshold. If so, the iteration stops, and the transportation path corresponding to the global optimal position, i.e., the optimized transportation path for building materials, is output. Otherwise, the update operation in the third step is repeated until the termination condition is met. The convergence threshold is set based on the experience of relevant technical personnel.
[0057] Obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the path optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the path optimization map based on the anomaly scheduling matching results to obtain a transportation path scheduling scheme map.
[0058] Optionally, construction anomaly scheduling semantic information is extracted from the historical scheduling records based on a semantic recognition model (e.g., a pre-trained BERT model). This semantic information includes the cause of the anomaly, the time of the anomaly delay, and the location of the anomaly. A cosine similarity algorithm is used to perform anomaly information similarity matching on the historical status information and route plan data of all transport vehicles in the route optimization map, based on the cause of the anomaly and the location of the anomaly. Similarity matching results greater than an anomaly threshold are selected as anomaly weights. The default anomaly threshold is 0.9, but it can also be set based on expert experience. This step primarily aims to reduce human error, achieve fully automated calculation, and facilitate subsequent data appending calculations. The construction anomaly scheduling semantic information is then associated with the route optimization map based on the anomaly weights to obtain a preliminary transport route scheduling plan map.
[0059] The path optimization weights of the preliminary transportation route scheduling scheme map are dynamically corrected based on the anomaly dynamic correction function and similarity matching results to obtain the transportation route scheduling scheme map. The expression of the anomaly dynamic correction function is as follows:
[0060] ;
[0061] in, The path optimization weights are dynamically adjusted for abnormalities, where max is the maximum value and Cos is the cosine similarity algorithm. The cause of the aforementioned abnormality, This provides historical status information for all transport vehicles. For locations of abnormal routes, PA represents path plan data. Let be a logarithmic function with base e, t be the abnormal delay time, and N be the total number of abnormal delay time records in historical scheduling. This is the record of the i-th abnormal delay. This method can automatically identify abnormal characteristics based on historical scheduling data and dynamically correct the path optimization map, avoiding the problem that static path planning cannot adapt to sudden changes in the construction scenario. Simultaneously, by adjusting the path optimization weights through abnormal weight allocation, the corrected path plan better reflects the impact of abnormal events on transportation in actual construction, improving the rationality and feasibility of the final transportation path scheduling plan, reducing transportation delays caused by abnormal events, and improving the scheduling efficiency of construction material transportation.
[0062] The transportation routes for building materials are optimized based on the aforementioned transportation route scheduling scheme map. In one embodiment, the transportation vehicles to be transported and the target building materials to be transported are input into the transportation route scheduling scheme map. The transportation directories of the transportation route scheduling scheme map are matched using a cosine similarity algorithm. The maps are then traversed according to the connection weights of the matched transportation directories, sorted from largest to smallest weight, and the top-ranked transportation route scheduling scheme maps are output. These are then combined with an agent to generate multiple schemes, which are then selected by experts.
[0063] Example 2:
[0064] This application also provides a construction material transportation route optimization and scheduling system based on AI algorithms, based on the method of Embodiment 1, referencing... Figure 2 As shown, it includes:
[0065] The transportation catalog module is used to obtain construction material transportation information from historical scheduling records and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least material type, material weight, supply location, and required delivery time.
[0066] The route optimization graph module is used to obtain the historical status information of all transport vehicles based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory based on the historical scheduling records to obtain a preliminary association index, obtain the route plan data of the historical scheduling records, and optimize the association weight of the preliminary association index based on the route plan data to obtain the route optimization graph.
[0067] The transportation route scheduling scheme map module is used to obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the route optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the route optimization map based on the anomaly scheduling matching results to obtain the transportation route scheduling scheme map.
[0068] The transportation route optimization module optimizes the transportation routes of building materials based on the transportation route scheduling scheme map.
[0069] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms, characterized in that, include: Obtain historical scheduling records of construction material transportation information, and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least the material type, material weight, supply location, and required delivery time. Based on the historical scheduling records, obtain the historical status information of all transport vehicles. Based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory to obtain a preliminary association index. Obtain the path scheme data of the historical scheduling records. Based on the path scheme data, optimize the association weight of the preliminary association index to obtain a path optimization map. Obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the path optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the path optimization map based on the anomaly scheduling matching results to obtain a transportation path scheduling scheme map. The transportation routes for building materials are optimized based on the aforementioned transportation route scheduling scheme map.
2. The method for optimizing and scheduling the transportation routes of building materials based on AI algorithms according to claim 1, characterized in that, Obtain historical scheduling records of construction material transportation information, and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least material type, material weight, supply location, and required delivery time. The semantic feature information of the construction material transportation information is extracted based on the semantic recognition model, and multiple semantic clusters are obtained by clustering the semantic feature information of the construction material transportation information based on the clustering model. Obtain the cluster radius of multiple semantic clusters, sort the semantic clusters according to the size of the cluster radius to obtain sorted semantic clusters; obtain the minimum distance between the cluster center and the semantic feature information of each semantic cluster according to the Euclidean distance algorithm, and use the semantic feature information of the minimum distance as the semantic feature label of each semantic cluster. The ratio of the radius of each semantic cluster to the sum of the radii of all clusters is used as the transportation association weight. The transportation association weight is weighted according to the sorting order as the weighting value to obtain the initial association weight of each semantic cluster. The transportation directory is obtained by using the semantic feature labels of each semantic cluster as transportation directory terms and associating the initial association weights with each transportation directory term.
3. The method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms according to claim 2, characterized in that, Based on the historical scheduling records, obtain the historical status information of all transport vehicles. Then, associate the historical status information with the preliminary weights of the transport directory based on the historical scheduling records to obtain a preliminary association index. Obtain the path scheme data from the historical scheduling records. Finally, optimize the association weights of the preliminary association index based on the path scheme data to obtain a path optimization map, including: Based on historical scheduling records, extract historical status information of all scheduled transport vehicles. The historical status information includes at least vehicle type, approved load capacity, loaded weight, and vehicle usage time. Based on the semantic recognition model, the historical state semantic features of all transport vehicles and the transport vehicle semantic features of the construction material transport information are extracted. The semantic similarity between the historical state semantic features and the transport vehicle semantic features is calculated based on the cosine similarity algorithm. The historical state information with a semantic similarity greater than the vehicle similarity threshold is associated with the initial association weight of the transport directory to obtain the preliminary association index. The route plan data is obtained, which includes at least the transportation distance, actual fuel consumption data, delivery delay rate and transportation cost. The route plan data is then used to perform preliminary path association based on the initial path weight and the preliminary association index to obtain an initial path map. The path optimization weight is obtained by optimizing the path scheme data based on the path weight optimization formula. The path optimization weight is then multiplied by the initial path weight of the initial path map to obtain the path optimization map.
4. The method for optimizing and scheduling the transportation routes of building materials based on AI algorithms according to claim 3, characterized in that, The path optimization weights are obtained by optimizing the path scheme data based on the path weight optimization formula, and the expression of the path weight optimization formula is as follows: ； in, Optimize the weights for the path. To optimize the algorithm, Where FC represents the transport distance and FC represents the actual fuel consumption data. Let C be the delivery timeout rate, C be the transportation cost, min be the minimum value, and exp be an exponential function with base e, min be the minimum value. and These are the weighting coefficients.
5. The method for optimizing and scheduling the transportation routes of building construction materials based on AI algorithms according to claim 3, characterized in that, Obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the route optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the route optimization map based on the anomaly scheduling matching results to obtain a transportation route scheduling scheme map, including: The construction anomaly scheduling semantic information is extracted from the historical scheduling records based on the semantic recognition model. The construction anomaly scheduling semantic information includes the cause of the anomaly, the time of the anomaly delay, and the location of the anomaly. Based on the cosine similarity algorithm, the anomaly information similarity is matched with the historical status information and route plan data of all transport vehicles in the route optimization map through the cause of the anomaly and the location of the anomaly. The similarity matching results with an anomaly threshold are selected as the anomaly weights. The construction anomaly scheduling semantic information is associated with the route optimization map according to the anomaly weights to obtain a preliminary transport route scheduling plan map. The path optimization weights of the preliminary transportation route scheduling scheme map are dynamically corrected based on the anomaly dynamic correction function and similarity matching results to obtain the transportation route scheduling scheme map. The expression of the anomaly dynamic correction function is as follows: ； in, The path optimization weights are dynamically adjusted for abnormalities, where max is the maximum value and Cos is the cosine similarity algorithm. The cause of the aforementioned abnormality, This provides historical status information for all transport vehicles. For locations of abnormal routes, PA represents path plan data. Let be a logarithmic function with base e, t be the abnormal delay time, and N be the total number of abnormal delay time records in historical scheduling. This is the i-th record of abnormal time delay.
6. A construction material transportation route optimization and scheduling system based on AI algorithms, characterized in that, include: The transportation catalog module is used to obtain construction material transportation information from historical scheduling records and generate a transportation catalog based on the key data features of the construction material transportation information. The construction material transportation information includes at least material type, material weight, supply location, and required delivery time. The route optimization graph module is used to obtain the historical status information of all transport vehicles based on the historical scheduling records, associate the historical status information with the preliminary weight of the transport directory based on the historical scheduling records to obtain a preliminary association index, obtain the route plan data of the historical scheduling records, and optimize the association weight of the preliminary association index based on the route plan data to obtain the route optimization graph. The transportation route scheduling scheme map module is used to obtain construction anomaly scheduling information from historical scheduling records, extract the anomaly features of the construction anomaly scheduling information and perform anomaly scheduling matching with the route optimization map to obtain anomaly scheduling matching results, and perform anomaly dynamic correction on the route optimization map based on the anomaly scheduling matching results to obtain the transportation route scheduling scheme map. The transportation route optimization module optimizes the transportation routes of building materials based on the transportation route scheduling scheme map.