Logistics cargo transportation scheduling method and system based on artificial intelligence

Abstract

The invention relates to the technical field of logistics management, in particular to a logistics cargo transportation scheduling method and system based on artificial intelligence. The method comprises the following steps: obtaining logistics node data; performing multi-dimensional logistics data acquisition on the logistics node data to generate multi-dimensional logistics data; a logistics monitoring network is constructed according to the logistics node data as nodes and the multi-dimensional logistics data, and the logistics monitoring network is generated and comprises a storage node, a distribution center node and a transportation tool node; performing data preprocessing on the multi-dimensional logistics data through a logistics monitoring network to generate standard multi-dimensional logistics monitoring data; and performing adjacent logistics storage node position search on the standard multi-dimensional logistics monitoring data through the distribution center node to generate adjacent storage node position information data. According to the invention, through multi-dimensional data acquisition, dynamic path optimization, delay prediction and a full-link feedback mechanism, the efficiency and reliability of logistics cargo scheduling are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a logistics cargo transportation scheduling method and system based on artificial intelligence. Background Art

[0002] Early scheduling methods mainly relied on manual experience and lacked systematicity and efficiency. With the development of information technology, computer-based scheduling systems began to emerge, using simple algorithms for cargo distribution and route planning, but due to the limitation of computing power, the application effect of these systems was limited. Entering the 21st century, the development of the Internet of Things (IoT) enables logistics management to monitor the transportation process in real time. Through sensors and GPS technology, logistics companies can obtain the location information of goods in real time, thereby optimizing scheduling decisions. However, the surge in data volume has also brought new challenges, and traditional algorithms are difficult to handle complex scheduling problems. With the maturity of big data technology, logistics scheduling systems have begun to introduce data analysis and machine learning technologies. By analyzing historical data, AI can identify transportation patterns and demand fluctuations, thereby predicting the best transportation plan. In addition, the application of deep learning algorithms enables the system to make more accurate route planning and scheduling decisions. However, the existing technologies are often static for route planning, and it is difficult to adjust them in real time according to the dynamic data of logistics nodes. At the same time, they are relatively lagging in predicting transportation delays and responding to abnormal situations, which leads to low efficiency and reliability of logistics cargo scheduling. Summary of the invention

[0003] Based on this, it is necessary to provide an artificial intelligence-based logistics cargo transportation scheduling method and system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a logistics cargo transportation scheduling method based on artificial intelligence is provided, the method comprising the following steps:

[0005] Step S1: Obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data;

[0006] Step S2: Search the positions of adjacent logistics storage nodes for standard multi-dimensional logistics monitoring data through the distribution center node to generate adjacent storage node position information data; perform initial transportation path planning for logistics cargo transportation destination data through adjacent storage node position information data to generate initial logistics cargo transportation planning path data; perform storage scheduling retention analysis on standard multi-dimensional logistics monitoring data through storage nodes to generate storage scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data to generate logistics cargo transportation planning optimization path data;

[0007] Step S3: prioritize the logistics order data for cargo transportation according to the optimized path data of the logistics cargo transportation planning, and generate cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate a normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multidimensional logistics monitoring data based on the transportation tool node, and generate abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation according to the abnormal delay prediction data, and generate multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate multimodal transportation scheduling simulation data of cargo transportation;

[0008] Step S4: Perform a dispatching capacity analysis on the multimodal dispatching simulation data for freight transport to generate dispatching capacity analysis data; construct a joint dispatching plan for the multimodal dispatching simulation data for freight transport through the dispatching capacity analysis data to generate joint dispatching plan data; perform full-link transport feedback on the logistics monitoring network based on the joint dispatching plan data and the logistics transport anomaly detection data to generate full-link transport feedback data to execute logistics freight transport dispatching operations.

[0009] The present invention ensures the comprehensiveness and diversity of information by acquiring logistics node data and performing multi-dimensional data collection, so that subsequent analysis can be based on a richer data foundation. By preprocessing multi-dimensional logistics data, standard multi-dimensional logistics monitoring data is generated, which provides high-quality data support for subsequent analysis and decision-making of the logistics system and reduces the decision-making risk caused by data inconsistency. The distribution center node searches for the location of adjacent logistics storage nodes to ensure the rationality and effectiveness of path planning. Such an operation not only improves transportation efficiency but also reduces transportation costs. The initial path is screened based on the storage scheduling retention data to ensure the optimization of the transportation path and minimize transportation time and resource consumption. The logistics order data is prioritized for cargo transportation, so that resource allocation is more reasonable, and the transportation needs of high-priority orders are met first, thereby improving customer satisfaction. By analyzing the priority sorting data, a normal cargo transportation scheduling strategy is generated, which further ensures the efficiency and adaptability of transportation. Based on the abnormal detection of standard multi-dimensional logistics monitoring data, the system's prediction ability for potential transportation delays is improved, transportation problems are identified in advance, and managers are helped to adjust the scheduling plan in time to reduce the negative impact of delays. Through the multimodal transport scheduling of cargo transportation based on abnormal delay prediction data, the use of transportation resources is optimized, the flexible switching of transportation modes is realized, and different transportation needs are adapted. The generation and analysis of multimodal transport scheduling simulation data enables the scheduling scheme to predict and adjust the effect before implementation, reducing the risk in actual operation. Through the analysis of scheduling simulation data, the utilization rate of transportation resources can be evaluated and optimized in real time, the rationality of capacity allocation can be ensured, and the efficiency of logistics operation can be improved. Combining the joint scheduling scheme data with the logistics transportation anomaly detection data, the full-link transportation feedback data is generated, so that the information of each link of logistics transportation is transparent, which is convenient for managers to adjust strategies in time and improve the overall logistics efficiency. The method provides effective data support for decision makers by constructing a systematic logistics monitoring network and an intelligent data processing mechanism, making the decision-making process more scientific and accurate. In the ever-changing market environment, the scheduling method based on artificial intelligence can quickly adapt to new situations, ensure the efficient operation of the logistics system, and have strong adaptability. Therefore, the present invention improves the efficiency and reliability of logistics cargo scheduling through multidimensional data acquisition, dynamic path optimization, delay prediction and full-link feedback mechanism.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Obtain logistics node data;

[0012] Step S12: Using edge computing network technology to deploy sensor equipment on logistics node data to obtain logistics node sensor equipment deployment data; performing multi-dimensional logistics data collection on the logistics node sensor equipment deployment data to generate multi-dimensional logistics data;

[0013] Step S13: Use logistics node data as nodes and multi-dimensional logistics data as edges to construct a logistics monitoring network, generating a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes, and transportation tool nodes;

[0014] Step S14: preprocessing the multidimensional logistics data through the logistics monitoring network to generate standard multidimensional logistics monitoring data, wherein the data preprocessing includes data cleaning, data denoising, data normalization and data standardization.

[0015] The present invention realizes real-time monitoring of each link of logistics by constructing a logistics monitoring network including storage nodes, distribution center nodes and transportation tool nodes, which enables managers to timely discover and handle anomalies in the logistics process, optimize logistics paths and improve transportation efficiency. Multidimensional data collection of logistics nodes is performed using sensing equipment, and the accuracy and consistency of logistics data are ensured through data preprocessing steps (such as cleaning, denoising, normalization and standardization), providing reliable basic data for subsequent analysis and decision-making. The application of edge computing network technology enables sensing equipment to perform calculations and processing close to the data source, reduces the delay of data transmission, and improves the system response speed, which helps to achieve a more intelligent and real-time logistics monitoring and management system. The logistics monitoring network can adapt to logistics scenarios of different scales and types through the flexible configuration of nodes and edges. Whether it is a large-scale storage network or a multi-node transportation and distribution chain, it can operate effectively under this network architecture, enhancing the scalability and flexibility of the system. Through standardized multidimensional logistics monitoring data, not only the data quality is improved, but also a consistent data basis can be provided for the collaborative operation of multiple systems, promoting seamless docking and collaboration between different logistics management systems or platforms.

[0016] Preferably, step S2 comprises the following steps:

[0017] Step S21: extracting logistics orders from standard multi-dimensional logistics monitoring data through the distribution center node to obtain logistics order data; performing destination analysis on the logistics order data to generate logistics cargo transportation destination data;

[0018] Step S22: searching for the location of adjacent storage nodes for the logistics cargo transportation destination data to generate adjacent storage node location information data; performing initial transportation path planning for the logistics cargo transportation destination data using the adjacent storage node location information data to generate initial logistics cargo transportation planning path data;

[0019] Step S23: extracting storage space from the standard multi-dimensional logistics monitoring data through the storage node to obtain storage space data; performing available space analysis on the storage space data to generate available storage space data;

[0020] Step S24: Perform a warehouse scheduling retention analysis on the warehouse storage space available data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

[0021] The present invention extracts logistics order data and performs destination analysis to ensure that the destination of each batch of goods is accurately located, which makes the logistics distribution process more efficient and reduces the delay or loss of goods caused by the wrong destination. Based on the location information of adjacent storage nodes, the initial transportation path planning can effectively reduce the transportation distance and time and optimize the logistics cost. In addition, the transportation path is further optimized through subsequent path screening to ensure that the goods are distributed on the optimal route and improve the overall logistics efficiency. By extracting storage space data and analyzing available space, storage nodes can grasp the current available storage resources in real time to avoid insufficient or wasteful storage space, which is conducive to the reasonable allocation and efficient use of storage resources. Storage scheduling retention analysis can effectively avoid excessive backlog or ineffective mobilization of goods through dynamic management of storage space, improve the operating efficiency of storage nodes, and this scheduling optimization further affects the screening of cargo transportation paths to ensure close cooperation between storage and transportation. Step S2 realizes the optimization of the entire process from order generation to transportation and distribution through the collaborative work of multiple links such as order extraction, destination analysis, storage scheduling, and path optimization, improves the overall efficiency of the logistics system, reduces costs, and improves service quality. By optimizing the screening of transportation routes and combining them with the effective scheduling of warehousing resources, unnecessary intermediate links and resource waste are greatly reduced, thereby shortening transportation time and reducing logistics costs.

[0022] Preferably, step S24 includes the following steps:

[0023] Step S241: Performing space availability time series trend analysis on the warehouse storage space availability data to generate space availability time series trend data; performing curve conversion on the space availability time series trend data to generate a space availability time series trend curve;

[0024] Step S242: Calculate the curvature of the space available time series trend curve to obtain storage space available curvature data; distinguish the positive and negative values ​​of the storage space available curvature data, and when the storage space available curvature data is positive, mark the corresponding storage node as congested based on the storage space available curvature data to generate congested storage node data;

[0025] Step S243: when the storage space available curvature data is a negative value, the corresponding storage node is marked as idle based on the storage space available curvature data to generate idle storage node data; the initial logistics cargo transportation planning path data is screened for the first time according to the congested storage node data and the idle storage node data, and the congested storage node data is removed, thereby generating a logistics cargo transportation planning screening path;

[0026] Step S244: Perform a warehouse scheduling retention analysis on the available data of the warehouse storage space according to the idle warehouse node data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

[0027] The present invention can predict the change trend of storage space in advance by analyzing the available data of storage space in a time series trend, and distinguish the positive and negative values ​​according to the curvature data of the storage space, mark the crowded state or idle state of the storage node, so as to carry out resource allocation and scheduling in a targeted manner and maximize the utilization rate of storage resources. The generation of the available time series trend curve of space and the curvature calculation enable the system to monitor the state of storage nodes in real time. Through the crowded state marking and the idle state marking, the logistics system can instantly understand the usage of each storage node and quickly adjust the storage scheduling plan. Through the first path screening, the crowded storage nodes are eliminated to avoid the delay of goods due to insufficient storage resources or congestion, and ensure that the nodes passed by the transportation path have sufficient storage space when the goods arrive, which effectively reduces the potential delay risk in the transportation process and improves the transportation efficiency. The storage scheduling retention analysis further evaluates the available space of the idle storage nodes, so that the storage scheduling plan can be dynamically adjusted according to the actual situation, ensuring that the load of the storage node is within a reasonable range, and avoiding the waste or overload of storage resources. The spatial availability time series trend analysis and curvature calculation provide key data support for the logistics system, helping the system to make more intelligent decisions in route planning and warehouse scheduling. This data-driven decision-making mechanism effectively improves the overall response speed and accuracy of the logistics chain. Through the optimization and screening of routes and the intelligent scheduling of warehouse nodes, the additional transportation and storage costs caused by warehouse congestion or improper scheduling are avoided, and the efficiency and economy of the entire logistics process are improved.

[0028] Preferably, step S3 comprises the following steps:

[0029] Step S31: extracting logistics cargo features from logistics order data to obtain logistics cargo feature data; evaluating cargo importance from the logistics cargo feature data to generate cargo importance evaluation data; and prioritizing cargo transportation from the logistics order data based on the cargo importance evaluation data and optimizing the path data according to the logistics cargo transportation planning to generate cargo transportation priority ranking data;

[0030] Step S32: confirming the mode of transport according to the cargo transport priority ranking data, and generating a normal cargo transport scheduling strategy; extracting transport data from the standard multi-dimensional logistics monitoring data based on the transport tool node through the normal cargo transport scheduling strategy, and obtaining logistics transport data;

[0031] Step S33: performing transportation anomaly detection on the logistics transportation data to generate logistics transportation anomaly detection data; performing abnormal delay prediction on the logistics order data based on the logistics transportation anomaly detection data to generate abnormal delay prediction data;

[0032] Step S34: Perform freight transport multimodal transport scheduling on the logistics transportation data through the abnormal delay prediction data to generate freight transport multimodal transport scheduling data; perform scheduling simulation on the freight transport multimodal transport scheduling data to generate freight transport multimodal transport scheduling simulation data.

[0033] The present invention can accurately determine the transportation priority of each batch of goods by extracting the characteristics of logistics goods and evaluating their importance, and generate cargo transportation priority sorting data, which helps to reasonably arrange the transportation order, ensure that urgent goods or high-value goods are processed first, and optimize the allocation of logistics resources. The transportation mode is confirmed based on the cargo transportation priority sorting data, making transportation scheduling more flexible. For example, important goods can give priority to faster or safer transportation modes, while ordinary goods can choose lower-cost modes, thereby achieving rational use of resources. Through real-time monitoring and abnormal detection of logistics transportation data, problems occurring during transportation can be quickly discovered, and abnormal delay prediction of logistics orders can be performed based on the detection results, which helps to warn potential delay risks in advance, so that logistics management personnel can take countermeasures in time to reduce the impact of delays. According to the abnormal delay prediction data, multimodal transportation scheduling is performed on logistics transportation to achieve collaborative work between different transportation modes. Multimodal transportation scheduling data enables the system to select the optimal transportation combination according to actual conditions, reducing delays and cost increases caused by the limitation of a single transportation mode. Through the simulation of multimodal transport scheduling, the feasibility and effectiveness of different scheduling strategies can be evaluated in advance, which not only improves the accuracy of logistics scheduling, but also optimizes logistics routes and transportation methods in advance to avoid delays or waste of resources caused by incorrect scheduling strategies. Step S3 optimizes the entire process from cargo priority sorting, transportation method confirmation, anomaly detection to multimodal transport scheduling, forming a complete logistics management closed loop. Data-driven decisions in each link can effectively improve the overall efficiency and response speed of the logistics chain and reduce operating costs.

[0034] Preferably, step S33 includes the following steps:

[0035] Step S331: confirming the transport location coordinates of the logistics transport data to obtain the logistics transport location coordinates; updating the logistics transport location coordinates based on a preset time interval to generate updated logistics transport location coordinates;

[0036] Step S332: constructing coordinate movement trajectories for the logistics transportation location coordinates and the updated logistics transportation location coordinates to generate coordinate update trajectory data; discretizing the coordinate update trajectory data to generate coordinate update trajectory discrete points; calculating the distance between adjacent discrete points of the coordinate update trajectory discrete points to obtain coordinate update discrete point distance data;

[0037] Step S333: comparing the coordinate updated discrete point distance data with a preset discrete point distance threshold, when the coordinate updated discrete point distance data is less than the preset discrete point distance threshold, marking the corresponding coordinate updated trajectory data as abnormal coordinate trajectory data;

[0038] Step S334: perform transportation anomaly detection on the logistics transportation data based on the abnormal coordinate trajectory data to generate logistics transportation anomaly detection data; divide the logistics transportation anomaly detection data into data sets to generate a model training set and a model test set; perform model training on the model training set through a long short-term memory network algorithm to generate an abnormal delay prediction pre-model;

[0039] Step S335: perform model optimization iteration on the abnormal delay prediction pre-model through the model test set to generate an abnormal delay prediction model; import the logistics transportation abnormality detection data and logistics order data into the abnormal delay prediction model to perform abnormal delay prediction and generate abnormal delay prediction data.

[0040] The present invention can track the transportation location of goods in real time by confirming the transportation location coordinates and updating the coordinates within a preset time interval, which helps to ensure that each link in the logistics chain can obtain the latest transportation information in a timely manner, thereby improving the real-time and accuracy of logistics monitoring. By constructing and discretizing the trajectory of the logistics transportation location, the system can accurately calculate the distance between adjacent discrete points and monitor the deviation of the cargo transportation trajectory in real time. When the transportation trajectory is abnormal, the system can immediately mark and generate abnormal coordinate trajectory data to provide data support for subsequent abnormal detection and delay prediction. Based on the comparison between the discrete point distance and the preset threshold, the system can quickly identify abnormal transportation trajectories that cause delays or risks. Marking these abnormal trajectories helps logistics managers to promptly discover potential problems in transportation and avoid delays or damage. Through the long short-term memory network (LSTM) algorithm, the system can model the transportation anomaly detection data and train the abnormal delay prediction model. The LSTM algorithm is good at processing time series data and can accurately predict future delay risks based on the changing trends of historical trajectories and transportation data. By dividing the data set and optimizing the model iteratively, the abnormal delay prediction model can be continuously improved according to actual logistics data to improve the accuracy and reliability of the prediction. Through the optimization and iteration of the model, the system can better cope with the complex and ever-changing logistics and transportation environment. The introduction of the abnormal delay prediction model enables the system to predict the delay problems that occur during transportation in advance, thereby providing logistics managers with sufficient time to take countermeasures. Through accurate delay prediction, the system can optimize logistics scheduling and avoid additional costs and reduced customer satisfaction caused by delays. From transportation trajectory construction, discrete point calculation to LSTM-based delay prediction, the entire step S33 is a data-driven intelligent logistics management process. This process not only improves the accuracy of logistics monitoring, but also greatly enhances the prediction and response capabilities in logistics transportation through machine learning algorithms.

[0041] Preferably, step S34 includes the following steps:

[0042] Step S341: performing an initial transportation mode analysis on the logistics transportation data to generate initial transportation mode data; performing delay segment segmentation on the logistics transportation data using the abnormal delay prediction data to generate delay segment segmentation data;

[0043] Step S342: performing intermodal transport mode matching analysis on the delayed segment data to generate multimodal transport feasibility data, where the intermodal transport modes include land transport, sea transport, air transport and rail transport; performing transport resource allocation on the delayed segment data according to the multimodal transport feasibility data to generate multimodal transport dispatching data for freight transport;

[0044] Step S343: reconstructing the path of the delayed section segmentation data according to the freight transport multimodal transport scheduling data to generate multimodal transport reconstructed path data;

[0045] Step S344: Perform scheduling simulation on the delayed road segmentation data through multimodal transport reconstruction path data to generate multimodal transport scheduling simulation data for freight transportation.

[0046] The present invention can quickly generate initial transportation mode data by analyzing the initial transportation mode of logistics transportation data, and divide the delayed sections in combination with abnormal delay prediction data. This segmented analysis allows the system to accurately identify the sections that cause delays during transportation, which helps logistics management personnel to formulate targeted response strategies. After identifying the delayed sections, the system evaluates the feasibility of multimodal transportation such as land, sea, air, and rail transportation through intermodal mode matching analysis. Based on the multimodal feasibility data, the system can dynamically allocate transportation resources to the delayed sections to ensure that the best combination of transportation modes is selected in the shortest time, thereby improving transportation efficiency. The delayed section segmentation data is reconstructed through the multimodal transportation scheduling data to generate a more efficient multimodal reconstruction path. This step effectively solves the path inefficiency problem caused by the delayed sections during transportation, provides an optimized transportation path for the smooth transportation of goods, and reduces the risks caused by delays. By conducting scheduling simulation on the multimodal transport reconstruction path data, the actual effects of different transport modes and path combinations can be evaluated in advance, and multimodal transport scheduling simulation data can be generated. This simulation process helps the system to discover potential problems in advance and provide a reliable basis for the final transport scheduling decision, further reducing operational risks. Step S34 enables the logistics system to flexibly adjust the transportation plan in a complex transportation environment through the segmentation of delayed sections, intermodal feasibility analysis, path reconstruction and simulation, which not only improves the flexibility of transportation scheduling, but also enhances the ability of logistics management to respond to emergencies. The introduction of multimodal transport and the optimal allocation of resources enable the logistics system to utilize various transportation resources more efficiently, reducing delays and additional costs caused by a single mode of transportation or improper path selection. In addition, the scheduling simulation process can identify potential delay risks in advance, thereby effectively reducing unnecessary losses during transportation.

[0047] Preferably, step S4 comprises the following steps:

[0048] Step S41: extracting the dispatch delivery time from the freight transport multimodal transport dispatch simulation data to generate dispatch delivery time data; performing dispatch capacity analysis on the dispatch delivery time data and abnormal delay prediction data to generate dispatch capacity analysis data;

[0049] Step S42: constructing a joint scheduling scheme for the freight transport multimodal transport scheduling simulation data through the scheduling capacity analysis data to generate joint scheduling scheme data;

[0050] Step S43: Match the joint transportation plan data and the logistics transportation anomaly detection data to the joint transportation plan, and generate the abnormal transportation joint scheduling strategy; perform full-link transportation feedback on the logistics monitoring network according to the normal cargo transportation scheduling strategy and the abnormal transportation joint scheduling strategy, and generate full-link transportation feedback data to execute the logistics cargo transportation scheduling operation.

[0051] The present invention generates dispatch delivery time data by extracting the dispatch delivery time of the multimodal dispatch simulation data of cargo transportation. These data are combined with the abnormal delay prediction data to perform dispatch capacity analysis, ensuring that the system can reasonably plan and allocate capacity according to transportation demand, improving transportation efficiency and reducing resource waste. Based on the dispatch capacity analysis data, the system can construct a joint dispatch scheme for multimodal dispatch. The joint dispatch scheme data not only covers the coordination and cooperation between different modes of transportation, but also can deal with emergencies that occur during transportation, providing a flexible and efficient transportation dispatch solution. By matching the joint dispatch scheme data with the logistics transportation anomaly detection data, an abnormal transportation joint dispatch strategy is generated. This strategy can be coordinated and applied with the conventional cargo transportation dispatch strategy to ensure that both conventional transportation tasks can be handled in the logistics chain and abnormal transportation situations can be quickly responded to, thereby improving the overall resilience of the logistics system. Through the generation of full-link transportation feedback data, the logistics system can monitor the transportation situation of the entire logistics chain in real time and make dynamic adjustments according to normal and abnormal transportation dispatch strategies. This full-link feedback mechanism enhances the transparency and controllability of logistics transportation, ensures the efficient operation of each link, and reduces the risk of transportation delays or interruptions. By combining multi-dimensional data such as multimodal transport scheduling simulation, capacity analysis, abnormal delay prediction and full-link feedback, an intelligent logistics scheduling execution solution is provided. The system can dynamically adjust the scheduling strategy based on real-time data and predictive analysis to ensure the smoothness and punctuality of cargo transportation and improve the overall operational efficiency of the logistics system. Through the construction of scheduling capacity analysis and abnormal joint scheduling strategy, the system can foresee potential delay risks in transportation and take measures in advance to optimize it, thereby minimizing the impact of delays. In addition, reasonable scheduling capacity analysis ensures the full utilization of logistics resources, avoids problems caused by excess or insufficient capacity, and optimizes the resource allocation of the logistics system. Through the construction of joint scheduling solutions and full-link feedback mechanisms, the coordination of various links in logistics has been significantly improved. The coordinated operation between different nodes and transportation modes ensures seamless connection of the entire process from warehousing to transportation and then to distribution, and improves the overall operational efficiency of the entire logistics system. The combination of joint scheduling strategy and abnormal scheduling strategy makes the logistics system more robust when dealing with sudden abnormal situations. Whether it is a delay in transportation, equipment failure or traffic congestion, the system can respond quickly and adjust the transportation plan to ensure the continuity and stability of the logistics chain.

[0052] Preferably, step S41 includes the following steps:

[0053] Step S411: extracting time nodes from the freight transport multimodal transport scheduling simulation data to generate scheduling time node data; calculating the delivery time from the scheduling time node data to generate scheduling delivery time data;

[0054] Step S412: Compare the dispatch delivery time data with the abnormal delay prediction data to generate time difference analysis data; perform capacity dispatch analysis on the time difference analysis data to generate preliminary dispatch capacity analysis data, wherein the capacity dispatch analysis formula is as follows:

[0055]

[0056] In the formula, U represents the capacity utilization rate, D represents the transportation demand, C represents the available capacity, E represents the scheduling efficiency factor, and T d Indicated as the actual delay time, T s It is indicated as the scheduled delivery time;

[0057] Step S413: Optimize the capacity of the preliminary dispatch capacity analysis data to generate dispatch capacity optimization data; perform cluster analysis of actual transportation conditions on the dispatch capacity optimization data to generate dispatch capacity analysis data.

[0058] The present invention generates accurate scheduling delivery time data by extracting the time nodes of scheduling simulation data and calculating the delivery time, which provides a basis for subsequent delay warning and optimization, ensuring that the logistics system can accurately control the time progress of each transportation link, thereby improving the punctuality and reliability of delivery. The system generates time difference analysis data by comparing the scheduling delivery time data with the abnormal delay prediction data. This time difference analysis enables the system to predict the delay problem in advance and conduct capacity scheduling analysis in time, thereby effectively reducing the negative impact caused by the delay. The application of the formula enables the system to scientifically evaluate the capacity utilization rate and ensure that its utility is maximized under the condition of limited transportation resources. The scheduling capacity optimization data is generated by optimizing the preliminary scheduling capacity analysis data, and cluster analysis of the actual transportation situation is performed. This optimization and cluster analysis can identify the capacity demand and performance of different transportation schemes, thereby providing more accurate data support for subsequent scheduling strategies, which greatly improves the system's ability to cope with a changing transportation environment, especially in the face of logistics peaks or abnormal situations, it can dynamically adjust the capacity allocation to ensure the smooth progress of logistics transportation. Through precise capacity scheduling analysis and optimization, the system can perform scheduling optimization in advance when the risk of delay is high, ensuring the reasonable allocation of capacity under limited resources, thereby reducing the possibility of delays. This not only improves the overall transportation efficiency of the logistics system, but also reduces customer complaints and resource waste caused by delays. By combining the scheduling delivery time with abnormal prediction data, the system realizes intelligent capacity management. According to the capacity analysis data generated in real time, the system can dynamically adjust the scheduling plan to ensure that the cargo transportation is completed within the optimal time, which improves the flexibility and intelligence level of the logistics system. Step S41 establishes a full-process control system from time node extraction, time difference analysis to capacity scheduling optimization. The data flow of each link is closely connected, enabling the system to fully control the time and capacity in the transportation process and ensure the stability and efficiency of logistics transportation.

[0059] In this specification, a logistics cargo transportation scheduling system based on artificial intelligence is provided, which is used to execute the above-mentioned logistics cargo transportation scheduling method based on artificial intelligence. The logistics cargo transportation scheduling system based on artificial intelligence includes:

[0060] The logistics network construction module is used to obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data;

[0061] The path planning module is used to search for the positions of adjacent logistics storage nodes on the standard multi-dimensional logistics monitoring data through the distribution center node, and generate the position information data of adjacent storage nodes; to perform initial transportation path planning on the logistics cargo transportation destination data through the position information data of adjacent storage nodes, and generate the initial logistics cargo transportation planning path data; to perform storage scheduling retention analysis on the standard multi-dimensional logistics monitoring data through the storage node, and generate the storage scheduling retention data; to perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data, and generate the logistics cargo transportation planning optimization path data;

[0062] The transport scheduling module is used to prioritize the logistics order data according to the optimized path data of the logistics cargo transportation plan, and generate the cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate the normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multi-dimensional logistics monitoring data based on the transportation tool node, and generate the abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation based on the abnormal delay prediction data, and generate the multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate the multimodal transportation scheduling simulation data of cargo transportation;

[0063] The link feedback module is used to perform scheduling capacity analysis on the multimodal freight transport scheduling simulation data and generate scheduling capacity analysis data; to construct a joint scheduling plan for the multimodal freight transport scheduling simulation data through the scheduling capacity analysis data and generate joint scheduling plan data; to perform full-link transportation feedback on the logistics monitoring network based on the joint scheduling plan data and logistics transportation anomaly detection data, and generate full-link transportation feedback data to execute logistics freight transport scheduling operations.

[0064] The beneficial effect of the present invention is that by fully acquiring logistics node data, it can ensure that the system has a rich information foundation, which is helpful for subsequent analysis and decision-making, generates multi-dimensional logistics data and builds a monitoring network, realizes real-time monitoring of warehousing, distribution centers and transportation tools, and improves logistics transparency. Standardize multi-dimensional logistics monitoring data to make it suitable for subsequent analysis, ensure data quality and consistency, and thus improve the accuracy of decision-making. Through adjacent node search, the system can quickly identify the best storage location, provide a basis for path planning, reduce transportation time and cost, and the generated initial transportation planning path provides a reasonable transportation plan for logistics goods to ensure that the goods are delivered on time. By analyzing warehousing scheduling data, identifying potential scheduling problems, it is helpful to optimize inventory management and resource allocation. Path optimization is performed based on scheduling retention data to ensure that the best transportation route is selected, improve transportation efficiency and customer satisfaction. Prioritize transportation orders so that important orders are processed in a timely manner and optimize resource utilization. The normal cargo transportation scheduling strategy formulated based on priority ensures that transportation is carried out in the most appropriate way and improves response speed. Predict transportation delay risks, take countermeasures in advance, and reduce losses caused by delays. Improve transportation flexibility and efficiency through multimodal transport scheduling, making the connection between different modes of transport smoother. Simulating scheduling can evaluate the feasibility of different plans, optimize actual scheduling decisions, and reduce risks. Analyzing the scheduling capacity can identify potential bottlenecks and waste of resources, thereby achieving more efficient resource allocation. Generate a joint scheduling plan based on capacity analysis, so that multiple transportation links can work in coordination and improve overall efficiency. Through full-link transportation feedback, the logistics process can be monitored in real time, and the strategy can be adjusted in time to ensure smooth and efficient logistics operations, and improve the adaptability and response speed of the overall supply chain. Therefore, the present invention improves the efficiency and reliability of logistics cargo scheduling through multi-dimensional data collection, dynamic path optimization, delay prediction and full-link feedback mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of the steps of a logistics cargo transportation scheduling method based on artificial intelligence;

[0066] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0067] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0068] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0069] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0071] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, and the term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0072] To achieve this, please refer to Figures 1 to 3 , a logistics cargo transportation scheduling method based on artificial intelligence, the method comprising the following steps:

[0073] Step S1: Obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data;

[0074] Step S2: Search the positions of adjacent logistics storage nodes for standard multi-dimensional logistics monitoring data through the distribution center node to generate adjacent storage node position information data; perform initial transportation path planning for logistics cargo transportation destination data through adjacent storage node position information data to generate initial logistics cargo transportation planning path data; perform storage scheduling retention analysis on standard multi-dimensional logistics monitoring data through storage nodes to generate storage scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data to generate logistics cargo transportation planning optimization path data;

[0075] Step S3: prioritize the logistics order data for cargo transportation according to the optimized path data of the logistics cargo transportation planning, and generate cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate a normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multidimensional logistics monitoring data based on the transportation tool node, and generate abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation according to the abnormal delay prediction data, and generate multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate multimodal transportation scheduling simulation data of cargo transportation;

[0076] Step S4: Perform a dispatching capacity analysis on the multimodal dispatching simulation data for freight transport to generate dispatching capacity analysis data; construct a joint dispatching plan for the multimodal dispatching simulation data for freight transport through the dispatching capacity analysis data to generate joint dispatching plan data; perform full-link transport feedback on the logistics monitoring network based on the joint dispatching plan data and the logistics transport anomaly detection data to generate full-link transport feedback data to execute logistics freight transport dispatching operations.

[0077] The present invention ensures the comprehensiveness and diversity of information by acquiring logistics node data and performing multi-dimensional data collection, so that subsequent analysis can be based on a richer data foundation. By preprocessing multi-dimensional logistics data, standard multi-dimensional logistics monitoring data is generated, which provides high-quality data support for subsequent analysis and decision-making of the logistics system and reduces the decision-making risk caused by data inconsistency. The distribution center node searches for the location of adjacent logistics storage nodes to ensure the rationality and effectiveness of path planning. Such an operation not only improves transportation efficiency but also reduces transportation costs. The initial path is screened based on the storage scheduling retention data to ensure the optimization of the transportation path and minimize transportation time and resource consumption. The logistics order data is prioritized for cargo transportation, so that resource allocation is more reasonable, and the transportation needs of high-priority orders are met first, thereby improving customer satisfaction. By analyzing the priority sorting data, a normal cargo transportation scheduling strategy is generated, which further ensures the efficiency and adaptability of transportation. Based on the abnormal detection of standard multi-dimensional logistics monitoring data, the system's prediction ability for potential transportation delays is improved, transportation problems are identified in advance, and managers are helped to adjust the scheduling plan in time to reduce the negative impact of delays. Through the multimodal transport scheduling of cargo transportation based on abnormal delay prediction data, the use of transportation resources is optimized, the flexible switching of transportation modes is realized, and different transportation needs are adapted. The generation and analysis of multimodal transport scheduling simulation data enables the scheduling scheme to predict and adjust the effect before implementation, reducing the risk in actual operation. Through the analysis of scheduling simulation data, the utilization rate of transportation resources can be evaluated and optimized in real time, the rationality of capacity allocation can be ensured, and the efficiency of logistics operation can be improved. Combining the joint scheduling scheme data with the logistics transportation anomaly detection data, the full-link transportation feedback data is generated, so that the information of each link of logistics transportation is transparent, which is convenient for managers to adjust strategies in time and improve the overall logistics efficiency. The method provides effective data support for decision makers by constructing a systematic logistics monitoring network and an intelligent data processing mechanism, making the decision-making process more scientific and accurate. In the ever-changing market environment, the scheduling method based on artificial intelligence can quickly adapt to new situations, ensure the efficient operation of the logistics system, and have strong adaptability. Therefore, the present invention improves the efficiency and reliability of logistics cargo scheduling through multidimensional data acquisition, dynamic path optimization, delay prediction and full-link feedback mechanism.

[0078] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a logistics cargo transportation scheduling method based on artificial intelligence of the present invention. In this example, the logistics cargo transportation scheduling method based on artificial intelligence includes the following steps:

[0079] Step S1: Obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data;

[0080] In the embodiment of the present invention, the logistics nodes to be collected are identified and determined, including storage nodes, distribution center nodes and transportation tool nodes. Sensors (such as RFID tags, GPS devices) and data acquisition systems are used to collect basic information (such as location, status, inventory level, etc.) of each logistics node in real time. The dimensions of multidimensional logistics data are defined, such as time dimension, space dimension, quantity dimension and status dimension. Logistics information from different data sources are integrated to ensure data consistency and integrity. Data cleaning and deduplication technology is used to handle missing values ​​and outliers. The collected logistics data is converted into a multidimensional data format to generate a multidimensional logistics data set. According to the logistics node data, each logistics node is regarded as a node in the network, and the connection relationship between nodes (such as the relationship between the warehouse and the distribution center) is defined. The structure of the logistics monitoring network is designed, including the hierarchical relationship and interaction path of each node, to ensure the smooth transmission of data flow. The network construction tool (such as the graph database) is used to realize the visualization of the logistics monitoring network, and a unique identifier is assigned to each node. The multidimensional logistics data in the logistics monitoring network is cleaned, redundant information is eliminated, and the accuracy and consistency of the data are ensured. The multidimensional data is converted according to a preset standard format for subsequent analysis and processing. For example, the quantities of different units are converted into a unified standard unit. Finally, a standard multi-dimensional logistics monitoring data set is generated, including cleaned and normalized data, to provide support for subsequent data analysis and decision-making.

[0081] Step S2: Search the positions of adjacent logistics storage nodes for standard multi-dimensional logistics monitoring data through the distribution center node to generate adjacent storage node position information data; perform initial transportation path planning for logistics cargo transportation destination data through adjacent storage node position information data to generate initial logistics cargo transportation planning path data; perform storage scheduling retention analysis on standard multi-dimensional logistics monitoring data through storage nodes to generate storage scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data to generate logistics cargo transportation planning optimization path data;

[0082] In an embodiment of the present invention, the location information (such as longitude and latitude, address, etc.) of the storage node is extracted by using standard multi-dimensional logistics monitoring data. The adjacent location of the storage node is searched through the distribution center node, and the storage nodes adjacent to the distribution center are identified by using a geographic information system (GIS) or other spatial analysis tools, and a database containing the location information of the adjacent storage nodes is generated, and the relevant attributes of the adjacent nodes (such as distance, capacity, etc.) are recorded. According to the transportation destination of the logistics goods, the location information and relevant attributes of the destination (such as priority, timeliness, etc.) are extracted. Using a path planning algorithm (such as Dijkstra algorithm, etc.), combined with the location information data of adjacent storage nodes, the initial transportation path of the logistics goods is planned, and the initial logistics goods transportation planning path data is generated, including the storage nodes passed, transportation time, transportation distance and other information. The standard multi-dimensional logistics monitoring data is analyzed to collect the storage node's in and out records, storage cycle, cargo turnover rate and other information. Apply data analysis techniques (such as regression analysis and time series analysis) to analyze the scheduling retention of storage nodes, evaluate the efficiency and retention rate of storage scheduling, generate storage scheduling retention data, and record the retention and scheduling efficiency of each storage node for subsequent analysis. According to the storage scheduling retention data, set the path screening criteria, such as transportation timeliness, cost, risk and other indicators. Use optimization algorithms (such as genetic algorithms, simulated annealing algorithms, etc.) to screen and optimize the initial logistics cargo transportation planning path data, consider various screening criteria, and generate optimized logistics cargo transportation planning path data to ensure the best performance of the final path in various indicators.

[0083] Step S3: prioritize the logistics order data for cargo transportation according to the optimized path data of the logistics cargo transportation planning, and generate cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate a normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multidimensional logistics monitoring data based on the transportation tool node, and generate abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation according to the abnormal delay prediction data, and generate multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate multimodal transportation scheduling simulation data of cargo transportation;

[0084] In an embodiment of the present invention, the relevant information of each logistics order is extracted from the logistics cargo transportation planning optimization path data, including the order quantity, delivery deadline, customer priority, etc. According to the importance of the order, the delivery deadline and the customer demand, the priority sorting standard of cargo transportation is set. A sorting algorithm (such as a weighted scoring method) is applied to calculate each order, generate cargo transportation priority sorting data, and determine the transportation priority of each order. According to the cargo transportation priority sorting data, the standards for confirming the mode of transportation are set, such as cost, timeliness and cargo characteristics (such as fragility and perishability). The adaptability of different modes of transportation (such as land transportation, sea transportation, and air transportation) to each order is analyzed, and timeliness and cost-effectiveness are comprehensively considered. According to the evaluation results, a normal cargo transportation scheduling strategy is generated to clarify the transportation mode and related scheduling details of each order. The standard multidimensional logistics monitoring data is analyzed based on the transportation tool node, and historical transportation data is collected, including transportation time, delay records and abnormal events. Time series analysis or machine learning models (such as regression analysis, LSTM network) are applied to predict abnormal transportation delays, generate abnormal delay prediction data, identify potential transportation delay risks, and evaluate the delay range. Utilize abnormal delay prediction data to conduct a comprehensive analysis of cargo transportation scheduling and identify factors that affect transportation efficiency in multimodal transport. Develop a multimodal cargo transportation scheduling plan, rationally configure different transportation tools (such as trucks, ships, airplanes) and routes to improve overall transportation efficiency, generate multimodal cargo transportation scheduling data, and record the execution plan of each transportation link in detail. Establish a transportation scheduling simulation environment and use computer simulation software (such as AnyLogic, MATLAB, etc.) to conduct virtual testing of transportation scheduling. Execute the generated multimodal cargo transportation scheduling data in the simulation environment, observe the dynamic changes in the transportation process, evaluate the effectiveness of the scheduling strategy, generate multimodal cargo transportation scheduling simulation data, and record key indicators in the simulation process (such as transportation time, cost, delay, etc.) to provide reference for subsequent decision-making.

[0085] Step S4: Perform a dispatching capacity analysis on the multimodal dispatching simulation data for freight transport to generate dispatching capacity analysis data; construct a joint dispatching plan for the multimodal dispatching simulation data for freight transport through the dispatching capacity analysis data to generate joint dispatching plan data; perform full-link transport feedback on the logistics monitoring network based on the joint dispatching plan data and the logistics transport anomaly detection data to generate full-link transport feedback data to execute logistics freight transport dispatching operations.

[0086] In the embodiment of the present invention, relevant indicators are extracted from the multimodal transport dispatch simulation data of freight transport, including the load, transportation time, dispatch frequency and other information of the transport tools. A dispatch capacity evaluation model is established, taking into account the capacity of each transport tool, the dispatch timeliness and the matching degree between it and the cargo demand. The simulated data is subjected to capacity analysis, the impact of different transport modes and dispatch strategies on the overall capacity is evaluated, the insufficient or redundant capacity is identified, the dispatch capacity analysis data is generated, and a visual report of the efficiency of capacity utilization is provided for reference in subsequent decision-making. According to the dispatch capacity analysis data, the construction standard of the joint dispatch scheme is set to ensure the coordination and reasonable allocation of the capacity of each transport tool. The joint dispatch algorithm (such as linear programming, optimal dispatch algorithm, etc.) is used to combine the characteristics of the transport tool and the dispatch requirements to construct a joint dispatch scheme. The preliminarily constructed joint dispatch scheme is verified to ensure that the scheme is feasible in actual dispatch and can maximize the utilization of capacity. If necessary, adjustments are made to generate joint dispatch scheme data to record the dispatch arrangements and plans of each transport tool in detail. The joint dispatch scheme data is integrated with the logistics transportation anomaly detection data to analyze the transportation status and risks of each link. Utilize the logistics monitoring network to monitor all aspects of cargo transportation and ensure real-time update and feedback of information. Establish a feedback mechanism based on real-time monitoring results to promptly feedback abnormal situations to the dispatching system, adjust transportation strategies or dispatching plans, generate full-link transportation feedback data, record feedback information and corresponding adjustment measures, and support subsequent logistics cargo transportation dispatching operations.

[0087] Preferably, step S1 comprises the following steps:

[0088] Step S11: Obtain logistics node data;

[0089] Step S12: Using edge computing network technology to deploy sensor equipment on logistics node data to obtain logistics node sensor equipment deployment data; performing multi-dimensional logistics data collection on the logistics node sensor equipment deployment data to generate multi-dimensional logistics data;

[0090] Step S13: Use logistics node data as nodes and multi-dimensional logistics data as edges to construct a logistics monitoring network, generating a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes, and transportation tool nodes;

[0091] Step S14: preprocessing the multidimensional logistics data through the logistics monitoring network to generate standard multidimensional logistics monitoring data, wherein the data preprocessing includes data cleaning, data denoising, data normalization and data standardization.

[0092] In the embodiment of the present invention, by acquiring logistics node data related to the logistics chain, these node data usually come from the following categories: storage node: including warehouse location, inventory information, storage equipment, etc. Distribution center node: including distribution center location, scheduling information, cargo sorting data, etc. Transport tool node: including location information, status information, route planning, etc. of transport tools such as trucks, airplanes, and ships. These node data can be collected through existing databases, RFID, GPS sensors, inventory management systems, and other channels in the logistics system. The collection of node data is the basis for system construction. Sensors are installed on storage nodes, distribution center nodes, and transport tool nodes, including environmental sensors (such as temperature, humidity), positioning sensors, cargo status monitoring sensors, etc. These sensors perform real-time data collection through edge computing networks. The sensor equipment collects various data in the logistics nodes to form multi-dimensional logistics data, which include but are not limited to: location information (GPS), cargo temperature and humidity information, operating status of transport tools (such as fuel consumption, speed, etc.), and cargo storage conditions of storage nodes (such as inventory, cargo arrangement, etc.). The edge computing network can quickly process these data between the sensor and the central server, and perform preliminary analysis and filtering, thereby reducing the processing pressure of the center. Based on the logistics node data obtained in step S11 and the multidimensional logistics data collected in step S12, the system constructs the node data and multidimensional data into a logistics monitoring network: the storage node, the distribution center node and the transportation tool node are used as the vertices in the network, and these nodes are connected by the logistics data between the nodes. The edges represent the information such as the goods circulation path, transportation time, and goods status between the nodes. After the logistics monitoring network is formed, it can fully reflect the operation of the entire logistics chain and provide real-time monitoring between different nodes and paths. The data is preprocessed in the logistics monitoring network to remove the abnormal values, missing values ​​and other illogical data in the data. For example, if the GPS position of a node deviates, the system will detect and remove these data. The sensor data is denoised by the edge computing device to remove the influencing factors such as sensor error and network transmission noise to ensure the accuracy of the data. The data of different dimensions are normalized to unify the magnitude and range of the data for subsequent analysis. For example, different types of data such as temperature, humidity, and location are normalized to the same scale. The data is converted using standardized rules so that data from different sources and in different formats are presented under the same standard, thereby improving data comparability and processing efficiency. Finally, after data cleaning, denoising, normalization and standardization, standard multi-dimensional logistics monitoring data is generated, which not only ensures its accuracy, but also provides strong support for subsequent logistics optimization and decision-making.

[0093] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0094] Step S21: extracting logistics orders from standard multi-dimensional logistics monitoring data through the distribution center node to obtain logistics order data; performing destination analysis on the logistics order data to generate logistics cargo transportation destination data;

[0095] Step S22: searching for the location of adjacent storage nodes for the logistics cargo transportation destination data to generate adjacent storage node location information data; performing initial transportation path planning for the logistics cargo transportation destination data using the adjacent storage node location information data to generate initial logistics cargo transportation planning path data;

[0096] Step S23: extracting storage space from the standard multi-dimensional logistics monitoring data through the storage node to obtain storage space data; performing available space analysis on the storage space data to generate available storage space data;

[0097] Step S24: Perform a warehouse scheduling retention analysis on the warehouse storage space available data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

[0098] In an embodiment of the present invention, the standard multi-dimensional logistics monitoring data is extracted by the distribution center node. Specifically, the distribution center node can screen out the logistics orders that need to be processed currently according to the real-time monitored information such as goods, vehicles, storage status, etc., and generate logistics order data. For example, when the goods arrive at the distribution center, the system extracts the relevant information of the goods according to the logistics order data, including the quantity of goods, transportation requirements, customer information, etc. After generating the logistics order data, the system analyzes the destination in each order and identifies the logistics cargo transportation destination data of each order. This data reflects the final delivery location of each cargo and is used for subsequent transportation route planning. For example, the destination of an order is Beijing, and the system will generate relevant transportation destination data based on this information. According to the logistics cargo transportation destination data generated in step S21, the system searches for the location information of adjacent storage nodes. This operation determines the storage node closest to the destination through the location data in the geographic information system (GIS) or the storage network, and generates the location information data of the adjacent storage nodes. For example, for orders sent to Beijing, the system searches for the location information of storage nodes around Beijing in order to find the best storage or transfer point for the goods. Using the location information data of adjacent storage nodes, the system performs initial path planning for logistics goods. The planning is based on the existing traffic routes, storage node layout and real-time traffic conditions to generate initial logistics cargo transportation planning path data. For example, if a certain cargo is sent from Shanghai to Beijing, the system will plan an initial transportation route and consider the storage nodes along the way as transit or backup routes. The system analyzes the standard multi-dimensional logistics monitoring data through the storage nodes and extracts the storage space data of the current storage nodes. The data includes information such as the existing storage capacity, cargo types, and storage conditions in each storage node. For example, the system will analyze whether a storage node currently has free storage space to accept the upcoming cargo. Based on the extracted storage space data, the system performs available space analysis and generates storage space available data. This data reflects the storage capacity that the storage node can currently allocate to new cargo, ensuring that the cargo can smoothly enter the appropriate storage in the logistics chain. For example, if a storage node has 500 cubic meters of free space, but requires special refrigeration conditions, the system will analyze and generate specific available space data accordingly. The system further analyzes the available data of warehouse storage space to evaluate whether the warehouse node needs to reserve storage space for future orders or existing goods, and generates warehouse scheduling retention data. This step helps the system ensure that the warehouse node will not be overloaded or logistics will not be stagnant due to lack of space. For example, a warehouse node needs to reserve space for a large number of goods that are about to arrive, so the system will make space reservation decisions based on future logistics needs. Finally, based on the generated warehouse scheduling retention data, the system screens and optimizes the previous initial logistics cargo transportation planning path data.By excluding those routes with insufficient storage nodes or tight scheduling, the system can generate more efficient logistics cargo transportation planning and optimization route data. For example, if the initial planning route passes through a storage node with insufficient space, the system will re-plan the route and deliver the goods to another storage node to ensure smooth transportation and reasonable use of storage.

[0099] Preferably, step S24 includes the following steps:

[0100] Step S241: Performing space availability time series trend analysis on the warehouse storage space availability data to generate space availability time series trend data; performing curve conversion on the space availability time series trend data to generate a space availability time series trend curve;

[0101] Step S242: Calculate the curvature of the space available time series trend curve to obtain storage space available curvature data; distinguish the positive and negative values ​​of the storage space available curvature data, and when the storage space available curvature data is positive, mark the corresponding storage node as congested based on the storage space available curvature data to generate congested storage node data;

[0102] Step S243: when the storage space available curvature data is a negative value, the corresponding storage node is marked as idle based on the storage space available curvature data to generate idle storage node data; the initial logistics cargo transportation planning path data is screened for the first time according to the congested storage node data and the idle storage node data, and the congested storage node data is removed, thereby generating a logistics cargo transportation planning screening path;

[0103] Step S244: Perform a warehouse scheduling retention analysis on the available data of the warehouse storage space according to the idle warehouse node data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

[0104] In an embodiment of the present invention, by performing a time series analysis on the available data of the storage space, the space availability of each storage node at different times is obtained. This analysis helps to understand the usage trend, change law and future available space forecast of the storage space, and generate space availability time series trend data, which includes the usage of the storage space at each time point. The space availability time series trend data is converted into a curve form to generate a space availability time series trend curve. This curve can intuitively reflect the usage trend of the storage space, which is convenient for subsequent analysis. Interpolation, spline curves and other methods can be used to convert the curve to smooth data fluctuations. The curvature of the generated space availability time series trend curve is calculated to obtain the storage space availability curvature data. The curvature reflects the speed of the curve change and can indicate the changing trend of the space availability status. The curvature is calculated using the relationship between the first-order derivative and the second-order derivative, and the formula is as follows: Curvature = (1 + (y′)2)3 / 2y″, where y′ and y″ are the first and second derivatives of the curve respectively. The storage space availability curvature data is distinguished as positive and negative values. When the curvature data is positive, it means that the space availability is increasing, indicating that the storage node is gradually crowded. Based on this judgment, the corresponding storage node is marked as crowded and the crowded storage node data is generated. When the storage space available curvature data is negative, it means that the space availability is decreasing, which means that the storage node is relatively idle. In this regard, the system marks the corresponding storage node as idle and generates idle storage node data. According to the generated crowded storage node data and idle storage node data, the initial logistics cargo transportation planning path data is screened for the first time. The crowded storage nodes passed in the path are eliminated to generate the logistics cargo transportation planning screening path. This screening ensures that the goods avoid crowded storage nodes as much as possible during transportation and reduce the risk of delays. The idle storage node data is used to conduct an in-depth analysis of the storage storage space available data to generate storage scheduling retention data. This data shows the amount of space that can be reserved by the idle storage node in the future period of time, as well as the need to reserve for new goods. For example, if a certain idle storage node is predicted to have goods arriving in the future, a certain amount of space needs to be reserved to cope with the arrival of new goods. Based on the generated warehousing scheduling retention data, the previous initial logistics cargo transportation planning path data is further screened to ensure that each transportation path passes through a storage node with sufficient storage space. Finally, the logistics cargo transportation planning optimization path data is generated, which will maximize the transportation efficiency and ensure the smooth and timely transportation of goods.

[0105] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0106] Step S31: extracting logistics cargo features from logistics order data to obtain logistics cargo feature data; evaluating cargo importance from the logistics cargo feature data to generate cargo importance evaluation data; and prioritizing cargo transportation from the logistics order data based on the cargo importance evaluation data and optimizing the path data according to the logistics cargo transportation planning to generate cargo transportation priority ranking data;

[0107] Step S32: confirming the mode of transport according to the cargo transport priority ranking data, and generating a normal cargo transport scheduling strategy; extracting transport data from the standard multi-dimensional logistics monitoring data based on the transport tool node through the normal cargo transport scheduling strategy, and obtaining logistics transport data;

[0108] Step S33: performing transportation anomaly detection on the logistics transportation data to generate logistics transportation anomaly detection data; performing abnormal delay prediction on the logistics order data based on the logistics transportation anomaly detection data to generate abnormal delay prediction data;

[0109] Step S34: Perform freight transport multimodal transport scheduling on the logistics transportation data through the abnormal delay prediction data to generate freight transport multimodal transport scheduling data; perform scheduling simulation on the freight transport multimodal transport scheduling data to generate freight transport multimodal transport scheduling simulation data.

[0110] In the embodiment of the present invention, by extracting features from the logistics order data, the attributes of each cargo (such as weight, volume, fragility, shelf life, etc.) are analyzed to generate logistics cargo feature data, which can be achieved through data mining and machine learning technology to extract features related to transportation efficiency and risk. Based on the extracted logistics cargo feature data, the importance of the cargo is evaluated. Using a weight algorithm, factors including the value, urgency, and customer demand of the cargo are considered to generate cargo importance evaluation data. According to the cargo importance evaluation data, the logistics orders are prioritized for transportation, and cargo transportation priority ranking data is generated. High-priority cargo should be arranged for transportation first to improve overall transportation efficiency and customer satisfaction. According to the cargo transportation priority ranking data, the appropriate mode of transportation (such as land transportation, sea transportation, air transportation, etc.) is confirmed, and a normal cargo transportation scheduling strategy is generated. This strategy should take into account factors such as the characteristics of the cargo, transportation time, and cost. Based on the transportation tool node, the standard multi-dimensional logistics monitoring data is analyzed using the normal cargo transportation scheduling strategy to extract relevant logistics transportation data, including transportation time, route, transportation tool status, etc. Analyze the extracted logistics and transportation data, identify potential transportation anomalies (such as delays, damage, loss, etc.), and generate logistics and transportation anomaly detection data. Anomaly detection algorithms (such as isolation forests, support vector machines, etc.) can be used for model training and identification. Based on the generated logistics and transportation anomaly detection data, abnormal delay prediction is performed on logistics orders to generate abnormal delay prediction data. This prediction can use time series analysis, regression models and other methods to predict possible delay times and their causes. Using abnormal delay prediction data, multimodal transport scheduling of cargo transportation is performed on logistics and transportation data to generate multimodal transport scheduling data for cargo transportation. This scheduling should take into account the advantages and disadvantages of different modes of transportation and how to efficiently switch between different modes of transportation. The generated multimodal transport scheduling data for cargo transportation is simulated to generate multimodal transport scheduling simulation data for cargo transportation. The scheduling results can be evaluated through simulation models to analyze the transportation time, cost and efficiency under different circumstances for optimization and adjustment.

[0111] Preferably, step S33 includes the following steps:

[0112] Step S331: confirming the transport location coordinates of the logistics transport data to obtain the logistics transport location coordinates; updating the logistics transport location coordinates based on a preset time interval to generate updated logistics transport location coordinates;

[0113] Step S332: constructing coordinate movement trajectories for the logistics transportation location coordinates and the updated logistics transportation location coordinates to generate coordinate update trajectory data; discretizing the coordinate update trajectory data to generate coordinate update trajectory discrete points; calculating the distance between adjacent discrete points of the coordinate update trajectory discrete points to obtain coordinate update discrete point distance data;

[0114] Step S333: comparing the coordinate updated discrete point distance data with a preset discrete point distance threshold, when the coordinate updated discrete point distance data is less than the preset discrete point distance threshold, marking the corresponding coordinate updated trajectory data as abnormal coordinate trajectory data;

[0115] Step S334: perform transportation anomaly detection on the logistics transportation data based on the abnormal coordinate trajectory data to generate logistics transportation anomaly detection data; divide the logistics transportation anomaly detection data into data sets to generate a model training set and a model test set; perform model training on the model training set through a long short-term memory network algorithm to generate an abnormal delay prediction pre-model;

[0116] Step S335: perform model optimization iteration on the abnormal delay prediction pre-model through the model test set to generate an abnormal delay prediction model; import the logistics transportation abnormality detection data and logistics order data into the abnormal delay prediction model to perform abnormal delay prediction and generate abnormal delay prediction data.

[0117] In an embodiment of the present invention, the location information of each transport node is extracted from the logistics transportation data to determine the logistics transportation location coordinates. The location coordinates are usually obtained by a GPS device to ensure the accuracy of the location. Based on a preset time interval (for example, every 5 minutes, 10 minutes, etc.), the acquired logistics transportation location coordinates are regularly updated to generate updated logistics transportation location coordinates. This process can be implemented by a scheduled task or real-time data stream processing. Using the current logistics transportation location coordinates and the updated logistics transportation location coordinates, a coordinate movement trajectory of the transportation path is constructed to generate coordinate update trajectory data. This trajectory represents the position change at each time point during the transportation process. The coordinate update trajectory data is discretized to generate coordinate update trajectory discrete points. This step can sample trajectory points at a certain interval. The distance between the coordinate update trajectory discrete points is calculated to generate coordinate update discrete point distance data. This step can be implemented by the Euclidean distance formula or other suitable distance calculation methods. According to the coordinate update discrete point distance data and the preset discrete point distance threshold, if the distance data is less than the preset threshold, it is determined to be an abnormal situation. The coordinate update trajectory data that meets the abnormal standard is marked as abnormal coordinate trajectory data for subsequent analysis and processing. Based on the marked abnormal coordinate trajectory data, the logistics and transportation data is comprehensively analyzed to generate logistics and transportation anomaly detection data. This process can be combined with machine learning algorithms to improve detection accuracy. The generated logistics and transportation anomaly detection data is divided into a model training set and a model test set, generally 70% for training and 30% for testing. The model training set is trained using the long short-term memory network (LSTM) algorithm to generate an abnormal delay prediction pre-model. This model is able to capture long-term dependencies in time series data. The generated abnormal delay prediction pre-model is optimized and iterated through the model test set, the model performance is evaluated, and the model parameters are adjusted according to the test results to finally generate an abnormal delay prediction model. The logistics and transportation anomaly detection data and logistics order data are imported into the abnormal delay prediction model, and the actual abnormal delay prediction is performed to generate the final abnormal delay prediction data.

[0118] Preferably, step S34 includes the following steps:

[0119] Step S341: performing an initial transportation mode analysis on the logistics transportation data to generate initial transportation mode data; performing delay segment segmentation on the logistics transportation data using the abnormal delay prediction data to generate delay segment segmentation data;

[0120] Step S342: performing intermodal transport mode matching analysis on the delayed segment data to generate multimodal transport feasibility data, where the intermodal transport modes include land transport, sea transport, air transport and rail transport; performing transport resource allocation on the delayed segment data according to the multimodal transport feasibility data to generate multimodal transport dispatching data for freight transport;

[0121] Step S343: reconstructing the path of the delayed section segmentation data according to the freight transport multimodal transport scheduling data to generate multimodal transport reconstructed path data;

[0122] Step S344: Perform scheduling simulation on the delayed road segmentation data through multimodal transport reconstruction path data to generate multimodal transport scheduling simulation data for freight transportation.

[0123] In an embodiment of the present invention, by comprehensively analyzing the logistics and transportation data, the current mode of cargo transportation (such as land transportation, sea transportation, air transportation, rail transportation, etc.) is identified, and initial transportation mode data is generated. The analysis can be comprehensively evaluated using historical data, transportation time and cost and other factors. Using abnormal delay prediction data, the logistics and transportation data is divided into delayed sections, the transportation sections with delay risks are identified, and delayed section segmentation data is generated, which can mark those sections that exceed the normal transportation time by setting thresholds. For the delayed section segmentation data, intermodal mode matching analysis is performed to generate multimodal feasibility data. This step requires evaluating the applicability of different modes of transportation in specific sections, including the feasibility of land transportation, sea transportation, air transportation and rail transportation, taking into account time, cost and cargo characteristics. According to the generated multimodal feasibility data, the delayed section segmentation data is optimized for resource allocation to ensure the selection of appropriate transportation modes and resources to generate multimodal dispatching data for cargo transportation. This process requires comprehensive consideration of the availability and timeliness of transportation tools. Using the generated freight transport multimodal transport scheduling data, the delayed section segmentation data is replanned and reconstructed to generate multimodal transport reconstruction path data. This step can be combined with the route optimization algorithm to ensure the optimal transportation path while reducing delays. Based on the multimodal transport reconstruction path data, the delayed section segmentation data is dispatched and simulated to generate freight transport multimodal transport scheduling simulation data. This step can use simulation models to test new transportation scheduling strategies and analyze their effectiveness and efficiency in actual operation.

[0124] Preferably, step S4 comprises the following steps:

[0125] Step S41: extracting the dispatch delivery time from the freight transport multimodal transport dispatch simulation data to generate dispatch delivery time data; performing dispatch capacity analysis on the dispatch delivery time data and abnormal delay prediction data to generate dispatch capacity analysis data;

[0126] Step S42: constructing a joint scheduling scheme for the freight transport multimodal transport scheduling simulation data through the scheduling capacity analysis data to generate joint scheduling scheme data;

[0127] Step S43: Match the joint transportation plan data and the logistics transportation anomaly detection data to the joint transportation plan, and generate the abnormal transportation joint scheduling strategy; perform full-link transportation feedback on the logistics monitoring network according to the normal cargo transportation scheduling strategy and the abnormal transportation joint scheduling strategy, and generate full-link transportation feedback data to execute the logistics cargo transportation scheduling operation.

[0128] In an embodiment of the present invention, by extracting key dispatch delivery time information, including the estimated arrival time, transportation start and end time, etc., from the cargo transportation multimodal transport dispatch simulation data, dispatch delivery time data is generated. In this step, historical transportation records can be analyzed by data mining technology to improve the accuracy of delivery time. The extracted dispatch delivery time data and abnormal delay prediction data are analyzed for capacity to evaluate whether the capacity of each dispatch scheme meets the expected demand. By combining factors such as the load capacity, availability and expected transportation time of the transport tool, dispatch capacity analysis data is generated, which will help identify potential capacity shortage or excess problems. Based on the generated dispatch capacity analysis data, the cargo transportation multimodal transport dispatch simulation data is comprehensively analyzed to construct joint dispatch scheme data. In this step, the dispatch of each mode of transportation is integrated to ensure the effective dispatch and delivery of goods in the shortest time. The generated joint dispatch scheme data is matched with the logistics transportation anomaly detection data to identify the best dispatch strategy under abnormal transportation conditions and generate an abnormal transportation joint dispatch strategy. In this step, the dispatch schemes under normal and abnormal conditions are compared to ensure that transportation challenges can be effectively addressed under abnormal conditions. According to the normal cargo transportation scheduling strategy and the abnormal transportation joint scheduling strategy, the logistics monitoring network is fed back with full-link transportation, which includes real-time updating of transportation status, tracking of cargo location, and adjustment of transportation plans. Finally, full-link transportation feedback data is generated to ensure the efficient execution of logistics cargo transportation scheduling operations. Feedback data can be used to optimize future scheduling plans and further improve overall transportation efficiency.

[0129] In this specification, a logistics cargo transportation scheduling system based on artificial intelligence is provided, which is used to execute the above-mentioned logistics cargo transportation scheduling method based on artificial intelligence. The logistics cargo transportation scheduling system based on artificial intelligence includes:

[0130] The logistics network construction module is used to obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data;

[0131] The path planning module is used to search for the positions of adjacent logistics storage nodes on the standard multi-dimensional logistics monitoring data through the distribution center node, and generate the position information data of adjacent storage nodes; to perform initial transportation path planning on the logistics cargo transportation destination data through the position information data of adjacent storage nodes, and generate the initial logistics cargo transportation planning path data; to perform storage scheduling retention analysis on the standard multi-dimensional logistics monitoring data through the storage node, and generate the storage scheduling retention data; to perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data, and generate the logistics cargo transportation planning optimization path data;

[0132] The transport scheduling module is used to prioritize the logistics order data according to the optimized path data of the logistics cargo transportation plan, and generate the cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate the normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multi-dimensional logistics monitoring data based on the transportation tool node, and generate the abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation based on the abnormal delay prediction data, and generate the multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate the multimodal transportation scheduling simulation data of cargo transportation;

[0133] The link feedback module is used to perform scheduling capacity analysis on the multimodal freight transport scheduling simulation data and generate scheduling capacity analysis data; to construct a joint scheduling plan for the multimodal freight transport scheduling simulation data through the scheduling capacity analysis data and generate joint scheduling plan data; to perform full-link transportation feedback on the logistics monitoring network based on the joint scheduling plan data and logistics transportation anomaly detection data, and generate full-link transportation feedback data to execute logistics freight transport scheduling operations.

[0134] The beneficial effect of the present invention is that by fully acquiring logistics node data, it can ensure that the system has a rich information foundation, which is helpful for subsequent analysis and decision-making, generates multi-dimensional logistics data and builds a monitoring network, realizes real-time monitoring of warehousing, distribution centers and transportation tools, and improves logistics transparency. Standardize multi-dimensional logistics monitoring data to make it suitable for subsequent analysis, ensure data quality and consistency, and thus improve the accuracy of decision-making. Through adjacent node search, the system can quickly identify the best storage location, provide a basis for path planning, reduce transportation time and cost, and the generated initial transportation planning path provides a reasonable transportation plan for logistics goods to ensure that the goods are delivered on time. By analyzing warehousing scheduling data, identifying potential scheduling problems, it is helpful to optimize inventory management and resource allocation. Path optimization is performed based on scheduling retention data to ensure that the best transportation route is selected, improve transportation efficiency and customer satisfaction. Prioritize transportation orders so that important orders are processed in a timely manner and optimize resource utilization. The normal cargo transportation scheduling strategy formulated based on priority ensures that transportation is carried out in the most appropriate way and improves response speed. Predict transportation delay risks, take countermeasures in advance, and reduce losses caused by delays. Improve transportation flexibility and efficiency through multimodal transport scheduling, making the connection between different modes of transport smoother. Simulating scheduling can evaluate the feasibility of different plans, optimize actual scheduling decisions, and reduce risks. Analyzing the scheduling capacity can identify potential bottlenecks and waste of resources, thereby achieving more efficient resource allocation. Generate a joint scheduling plan based on capacity analysis, so that multiple transportation links can work in coordination and improve overall efficiency. Through full-link transportation feedback, the logistics process can be monitored in real time, and the strategy can be adjusted in time to ensure smooth and efficient logistics operations, and improve the adaptability and response speed of the overall supply chain. Therefore, the present invention improves the efficiency and reliability of logistics cargo scheduling through multi-dimensional data collection, dynamic path optimization, delay prediction and full-link feedback mechanism.

[0135] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0136] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A logistics cargo transportation scheduling method based on artificial intelligence, characterized in that: The following steps are involved: Step S1: Obtain logistics node data; Collect multi-dimensional logistics data from logistics node data to generate multi-dimensional logistics data; The logistics monitoring network is constructed based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; the multi-dimensional logistics data is pre-processed through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data; Step S2: Search the positions of adjacent logistics storage nodes for standard multi-dimensional logistics monitoring data through the distribution center node to generate adjacent storage node position information data; perform initial transportation path planning on the logistics cargo transportation destination data through the adjacent storage node position information data to generate initial logistics cargo transportation planning path data; perform storage scheduling retention analysis on the standard multi-dimensional logistics monitoring data through the storage node to generate storage scheduling retention data; Based on the warehouse scheduling retention data, the initial logistics cargo transportation planning path data is screened to generate logistics cargo transportation planning optimization path data; Step S3: prioritize the logistics order data for cargo transportation according to the optimized path data of the logistics cargo transportation planning, and generate cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate a normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multidimensional logistics monitoring data based on the transportation tool node, and generate abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation according to the abnormal delay prediction data, and generate multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate multimodal transportation scheduling simulation data of cargo transportation; Step S4: Perform a dispatching capacity analysis on the multimodal dispatching simulation data for freight transport to generate dispatching capacity analysis data; construct a joint dispatching plan for the multimodal dispatching simulation data for freight transport through the dispatching capacity analysis data to generate joint dispatching plan data; perform full-link transport feedback on the logistics monitoring network based on the joint dispatching plan data and the logistics transport anomaly detection data to generate full-link transport feedback data to execute logistics freight transport dispatching operations.

2. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Obtain logistics node data; Step S12: Using edge computing network technology to deploy sensor equipment on logistics node data to obtain logistics node sensor equipment deployment data; performing multi-dimensional logistics data collection on the logistics node sensor equipment deployment data to generate multi-dimensional logistics data; Step S13: constructing a logistics monitoring network by using logistics node data as nodes and multi-dimensional logistics data as edges, and generating a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; Step S14: preprocessing the multidimensional logistics data through the logistics monitoring network to generate standard multidimensional logistics monitoring data, wherein the data preprocessing includes data cleaning, data denoising, data normalization and data standardization.

3. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: extracting logistics orders from standard multi-dimensional logistics monitoring data through the distribution center node to obtain logistics order data; performing destination analysis on the logistics order data to generate logistics cargo transportation destination data; Step S22: searching for the location of adjacent storage nodes for the logistics cargo transportation destination data to generate adjacent storage node location information data; performing initial transportation path planning for the logistics cargo transportation destination data using the adjacent storage node location information data to generate initial logistics cargo transportation planning path data; Step S23: extracting storage space from the standard multi-dimensional logistics monitoring data through the storage node to obtain storage space data; performing available space analysis on the storage space data to generate available storage space data; Step S24: Perform a warehouse scheduling retention analysis on the warehouse storage space available data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

4. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 3 is characterized in that: Step S24 includes the following steps: Step S241: Performing space availability time series trend analysis on the warehouse storage space availability data to generate space availability time series trend data; performing curve conversion on the space availability time series trend data to generate a space availability time series trend curve; Step S242: Calculate the curvature of the space available time series trend curve to obtain storage space available curvature data; distinguish the positive and negative values ​​of the storage space available curvature data, and when the storage space available curvature data is positive, mark the corresponding storage node as congested based on the storage space available curvature data to generate congested storage node data; Step S243: when the storage space available curvature data is a negative value, the corresponding storage node is marked as idle based on the storage space available curvature data to generate idle storage node data; the initial logistics cargo transportation planning path data is screened for the first time according to the congested storage node data and the idle storage node data, and the congested storage node data is removed, thereby generating a logistics cargo transportation planning screening path; Step S244: Perform a warehouse scheduling retention analysis on the available data of the warehouse storage space according to the idle warehouse node data to generate warehouse scheduling retention data; perform path screening on the initial logistics cargo transportation planning path data based on the warehouse scheduling retention data to generate logistics cargo transportation planning optimized path data.

5. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: extracting logistics cargo features from logistics order data to obtain logistics cargo feature data; evaluating cargo importance from the logistics cargo feature data to generate cargo importance evaluation data; and prioritizing cargo transportation from the logistics order data based on the cargo importance evaluation data and optimizing the path data according to the logistics cargo transportation planning to generate cargo transportation priority ranking data; Step S32: confirming the mode of transport according to the cargo transport priority ranking data, and generating a normal cargo transport scheduling strategy; extracting transport data from the standard multi-dimensional logistics monitoring data based on the transport tool node through the normal cargo transport scheduling strategy, and obtaining logistics transport data; Step S33: performing transportation anomaly detection on the logistics transportation data to generate logistics transportation anomaly detection data; performing abnormal delay prediction on the logistics order data based on the logistics transportation anomaly detection data to generate abnormal delay prediction data; Step S34: Perform freight transport multimodal transport scheduling on the logistics transportation data through the abnormal delay prediction data to generate freight transport multimodal transport scheduling data; perform scheduling simulation on the freight transport multimodal transport scheduling data to generate freight transport multimodal transport scheduling simulation data.

6. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 5 is characterized in that: Step S33 includes the following steps: Step S331: confirming the transport location coordinates of the logistics transport data to obtain the logistics transport location coordinates; updating the logistics transport location coordinates based on a preset time interval to generate updated logistics transport location coordinates; Step S332: constructing coordinate movement trajectories for the logistics transportation location coordinates and the updated logistics transportation location coordinates to generate coordinate update trajectory data; discretizing the coordinate update trajectory data to generate coordinate update trajectory discrete points; calculating the distance between adjacent discrete points of the coordinate update trajectory discrete points to obtain coordinate update discrete point distance data; Step S333: comparing the coordinate updated discrete point distance data with a preset discrete point distance threshold, when the coordinate updated discrete point distance data is less than the preset discrete point distance threshold, marking the corresponding coordinate updated trajectory data as abnormal coordinate trajectory data; Step S334: perform transportation anomaly detection on the logistics transportation data based on the abnormal coordinate trajectory data to generate logistics transportation anomaly detection data; divide the logistics transportation anomaly detection data into data sets to generate a model training set and a model test set; perform model training on the model training set through a long short-term memory network algorithm to generate an abnormal delay prediction pre-model; Step S335: perform model optimization iteration on the abnormal delay prediction pre-model through the model test set to generate an abnormal delay prediction model; import the logistics transportation abnormality detection data and logistics order data into the abnormal delay prediction model to perform abnormal delay prediction and generate abnormal delay prediction data.

7. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 6 is characterized in that: Step S34 includes the following steps: Step S341: performing an initial transportation mode analysis on the logistics transportation data to generate initial transportation mode data; performing delay segment segmentation on the logistics transportation data using the abnormal delay prediction data to generate delay segment segmentation data; Step S342: performing intermodal transport mode matching analysis on the delayed segment data to generate multimodal transport feasibility data, where the intermodal transport modes include land transport, sea transport, air transport and rail transport; performing transport resource allocation on the delayed segment data according to the multimodal transport feasibility data to generate multimodal transport dispatching data for freight transport; Step S343: reconstructing the path of the delayed section segmentation data according to the freight transport multimodal transport scheduling data to generate multimodal transport reconstructed path data; Step S344: Perform scheduling simulation on the delayed road segmentation data through multimodal transport reconstruction path data to generate multimodal transport scheduling simulation data for freight transportation.

8. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 7 is characterized in that: Step S4 includes the following steps: Step S41: extracting the dispatch delivery time from the freight transport multimodal transport dispatch simulation data to generate dispatch delivery time data; performing dispatch capacity analysis on the dispatch delivery time data and abnormal delay prediction data to generate dispatch capacity analysis data; Step S42: constructing a joint scheduling scheme for the freight transport multimodal transport scheduling simulation data through the scheduling capacity analysis data to generate joint scheduling scheme data; Step S43: Match the joint transportation plan data and the logistics transportation anomaly detection data to the joint transportation plan, and generate the abnormal transportation joint scheduling strategy; perform full-link transportation feedback on the logistics monitoring network according to the normal cargo transportation scheduling strategy and the abnormal transportation joint scheduling strategy, and generate full-link transportation feedback data to execute the logistics cargo transportation scheduling operation.

9. The method for logistics cargo transportation scheduling based on artificial intelligence according to claim 8 is characterized in that: Step S41 includes the following steps: Step S411: extracting time nodes from the freight transport multimodal transport scheduling simulation data to generate scheduling time node data; calculating the delivery time from the scheduling time node data to generate scheduling delivery time data; Step S412: Compare the dispatch delivery time data with the abnormal delay prediction data to generate time difference analysis data; perform capacity dispatch analysis on the time difference analysis data to generate preliminary dispatch capacity analysis data, wherein the capacity dispatch analysis formula is as follows: In the formula, U represents the capacity utilization rate, D represents the transportation demand, C represents the available capacity, E represents the scheduling efficiency factor, and T d Indicated as the actual delay time, T s It is indicated as the scheduled delivery time; Step S413: Optimize the capacity of the preliminary dispatch capacity analysis data to generate dispatch capacity optimization data; perform cluster analysis of actual transportation conditions on the dispatch capacity optimization data to generate dispatch capacity analysis data.

10. A logistics cargo transportation scheduling system based on artificial intelligence, characterized in that: Used to execute the artificial intelligence-based logistics cargo transportation scheduling method as claimed in claim 1, the artificial intelligence-based logistics cargo transportation scheduling system includes: The logistics network construction module is used to obtain logistics node data; perform multi-dimensional logistics data collection on the logistics node data to generate multi-dimensional logistics data; construct a logistics monitoring network based on the logistics node data as nodes and multi-dimensional logistics data to generate a logistics monitoring network, wherein the logistics monitoring network includes storage nodes, distribution center nodes and transportation tool nodes; perform data preprocessing on the multi-dimensional logistics data through the logistics monitoring network to generate standard multi-dimensional logistics monitoring data; The path planning module is used to search for the positions of adjacent logistics storage nodes on the standard multi-dimensional logistics monitoring data through the distribution center node, and generate the position information data of adjacent storage nodes; to perform initial transportation path planning on the logistics cargo transportation destination data through the position information data of adjacent storage nodes, and generate the initial logistics cargo transportation planning path data; to perform storage scheduling retention analysis on the standard multi-dimensional logistics monitoring data through the storage node, and generate the storage scheduling retention data; to perform path screening on the initial logistics cargo transportation planning path data based on the storage scheduling retention data, and generate the logistics cargo transportation planning optimization path data; The transport scheduling module is used to prioritize the logistics order data according to the optimized path data of the logistics cargo transportation plan, and generate the cargo transportation priority ranking data; confirm the transportation mode according to the cargo transportation priority ranking data, and generate the normal cargo transportation scheduling strategy; predict the abnormal transportation delay of the standard multi-dimensional logistics monitoring data based on the transportation tool node, and generate the abnormal delay prediction data; perform multimodal transportation scheduling of cargo transportation based on the abnormal delay prediction data, and generate the multimodal transportation scheduling data of cargo transportation; perform scheduling simulation on the multimodal transportation scheduling data of cargo transportation, and generate the multimodal transportation scheduling simulation data of cargo transportation; The link feedback module is used to perform scheduling capacity analysis on the multimodal freight transport scheduling simulation data and generate scheduling capacity analysis data; to construct a joint scheduling plan for the multimodal freight transport scheduling simulation data through the scheduling capacity analysis data and generate joint scheduling plan data; to perform full-link transportation feedback on the logistics monitoring network based on the joint scheduling plan data and logistics transportation anomaly detection data, and generate full-link transportation feedback data to execute logistics freight transport scheduling operations.

Images