A method and apparatus for managing transportation information

Abstract

The application discloses a kind of transport information management method and device, it is related to wisdom logistics technical field.The specific embodiment of the application includes: the historical transport data of certain transport line can be acquired, the fluctuation value of historical transport data is evaluated, and according to fluctuation value, from multiple transport models, the target transport model matched with fluctuation value is selected;Initial transport plan corresponding to the transport line is generated based on the target transport model;Further, the actual transport data of transport line is acquired, and the transport vehicle data and / or vehicle loading data contained in initial transport plan are dynamically adjusted according to actual transport data.The transport plan decision of the transport line of the embodiment of the application is divided into two stages, after generating initial transport plan using adaptive model, initial transport plan can be adjusted in real time by monitoring actual transport data, improve the accuracy, real-time performance and rationality of determining transport plan, so as to improve transport efficiency.

Description

Technical Field

[0001] This invention relates to the field of smart logistics technology, and in particular to a method and apparatus for managing transportation information. Background Technology

[0002] Existing methods for developing logistics transportation plans typically rely on historical transportation data. However, existing static transportation plans are difficult to adapt to dynamically changing logistics demands. In cases where complex transportation demands cause static transportation plans to fail, it can lead to adverse effects such as cargo transportation delays and increased transportation costs.

[0003] In addition, in actual transportation scenarios, there may be issues where the amount of goods exceeds the vehicle's loading capacity or the amount of goods transported by the vehicle is far less than the vehicle's loading capacity. Existing methods lack the detection and scheduling of transport data during transportation, which affects the efficiency of cargo transportation. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a transportation information management method and apparatus, capable of generating a transportation plan for any transportation route, acquiring historical transportation data corresponding to the transportation route, evaluating the fluctuation value of the historical transportation data, selecting a target transportation model matching the fluctuation value from multiple transportation models based on the fluctuation value, generating an initial transportation plan corresponding to the transportation route based on the target transportation model, further acquiring actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data. By dividing the transportation plan decision-making for transportation routes into two stages, after generating the initial transportation plan using an adapted model, the initial transportation plan can be adjusted in real time by monitoring actual transportation data, improving the accuracy, real-time nature, and rationality of determining the transportation plan, thereby improving transportation efficiency.

[0005] To achieve the above objectives, according to one aspect of the present invention, a transportation information management method is provided, comprising: in response to generating a transportation plan for any transportation route, acquiring historical transportation data corresponding to the transportation route, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data; evaluating the fluctuation value of the historical transportation data, selecting a target transportation model matching the fluctuation value from a variety of transportation models based on the fluctuation value; generating an initial transportation plan corresponding to the transportation route based on the target transportation model; acquiring one or more actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

[0006] Optionally, evaluating the fluctuation values ​​of the historical transportation data and selecting a target transportation model matching the fluctuation values ​​from multiple transportation models includes: determining a first fluctuation value of the historical transportation vehicle data in the time dimension and a second fluctuation value of the historical vehicle loading data in the time dimension; obtaining transportation change factors and transportation interference factors of the transportation route; and selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors.

[0007] Optionally, selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor includes: selecting a benchmark compensation model as the target transportation model from multiple transportation models when it is determined that the first fluctuation value is lower than a first fluctuation threshold, the second fluctuation value is lower than a second fluctuation threshold, there is no transportation change factor, and there is no transportation interference factor.

[0008] Optionally, selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor includes: selecting a time-series prediction model as the target transportation model when it is determined that the first fluctuation value is higher than a third fluctuation threshold, the second fluctuation value is lower than a fourth fluctuation threshold, there is a transportation change factor with time-series characteristics, and there is no transportation interference factor.

[0009] Optionally, selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor includes: selecting a large language model as the target transportation model from multiple transportation models when it is determined that the first fluctuation value is higher than a third fluctuation threshold, the second fluctuation value is higher than a fourth fluctuation threshold, there is a transportation change factor, and there is a transportation interference factor.

[0010] Optionally, the historical vehicle loading data further includes historical predicted cargo volume; generating the initial transportation plan corresponding to the transportation route based on the target transportation model includes: for the case where the target transportation model is a time-series prediction model, obtaining the deviation between historical predicted cargo volume and vehicle loading capacity, using the deviation as a target variable, using the historical predicted cargo volume as a dynamic covariate, using the time-series prediction model to analyze the time-series correlation between the target variable and the dynamic covariate, and outputting the prediction deviation based on the time-series correlation; and determining the transportation vehicle data included in the initial transportation plan based on the prediction deviation and the historical vehicle loading data.

[0011] Optionally, generating the initial transportation plan corresponding to the transportation route based on the target transportation model includes: for the case where the target transportation model is a large language model, obtaining the current cargo volume to be transported for the transportation route, the transportation change factors of the transportation route, and the transportation interference factors; constructing prompt words for the large language model based on the current cargo volume to be transported, the historical transportation data, the transportation change factors, the transportation interference factors, and one or more of various decision constraints; and using the large language model to perform reasoning based on the prompt words to output the initial transportation plan corresponding to the transportation route.

[0012] Optionally, the step of acquiring one or more actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data, includes: acquiring one or more data among vehicle allocation information, cargo loading information, and transportation feedback information included in the actual transportation data; updating the transportation vehicle data included in the initial transportation plan based on the vehicle allocation information; and / or determining the cargo loading progress based on the cargo loading information; determining the remaining loading capacity of the vehicle based on the cargo loading progress and vehicle loading capacity, and updating the vehicle loading data included in the initial transportation plan if it is determined that the remaining loading capacity exceeds a preset capacity threshold; and / or identifying transportation anomaly information indicated in the transportation feedback information, outputting a real-time transportation plan based on the transportation anomaly information using a large language model, and adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the real-time transportation plan.

[0013] Optionally, determining the cargo loading progress based on the cargo loading information includes: acquiring real-time cargo loading progress images at set time intervals; identifying real-time depth information of the loaded cargo in the vehicle's cargo compartment based on the real-time cargo loading progress images; determining the cargo loading progress based on the ratio of the real-time depth information of the loaded cargo to the total depth of the vehicle's cargo compartment; and / or constructing a virtual cargo compartment image; generating a virtual cargo image superimposed on the virtual cargo compartment image based on the real-time depth information of the loaded cargo; and displaying the cargo loading progress using the virtual cargo compartment image and the virtual cargo image.

[0014] To achieve the above objectives, according to a second aspect of the present invention, a transportation information management device is provided, comprising: a data acquisition module, a plan generation module, and a plan adjustment module; wherein, The data acquisition module is used to acquire historical transportation data corresponding to the transportation route in response to generating a transportation plan for any transportation route, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data. The plan generation module is used to evaluate the fluctuation value of the historical transportation data, select a target transportation model that matches the fluctuation value from multiple transportation models, and generate an initial transportation plan corresponding to the transportation route based on the target transportation model. The adjustment plan module is used to acquire one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

[0015] Optionally, a transportation information management device is used to evaluate the fluctuation values ​​of the historical transportation data and select a target transportation model that matches the fluctuation values ​​from multiple transportation models, including: determining a first fluctuation value of the historical transportation vehicle data in the time dimension and a second fluctuation value of the historical vehicle loading data in the time dimension; acquiring transportation change factors and transportation interference factors of the transportation route; and selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors.

[0016] Optionally, the transportation information management device is used to select the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor, including: selecting a benchmark compensation model as the target transportation model from multiple transportation models when it is determined that the first fluctuation value is lower than a first fluctuation threshold, the second fluctuation value is lower than a second fluctuation threshold, there is no transportation change factor, and there is no transportation interference factor.

[0017] Optionally, the transportation information management device is used to select the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor, including: selecting a time-series prediction model as the target transportation model from multiple transportation models when it is determined that the first fluctuation value is higher than a third fluctuation threshold, the second fluctuation value is lower than a fourth fluctuation threshold, there is a transportation change factor with time-series characteristics, and there is no transportation interference factor.

[0018] Optionally, the transportation information management device is used to select the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factor, and the transportation interference factor, including: selecting a large language model as the target transportation model from multiple transportation models when it is determined that the first fluctuation value is higher than a third fluctuation threshold, the second fluctuation value is higher than a fourth fluctuation threshold, there is a transportation change factor, and there is a transportation interference factor.

[0019] Optionally, the historical vehicle loading data further includes historical predicted cargo volume; the transportation information management device is used to generate an initial transportation plan corresponding to the transportation route based on the target transportation model, including: for the case where the target transportation model is a time-series prediction model, obtaining the deviation between historical predicted cargo volume and vehicle loading capacity, using the deviation as a target variable, using the historical predicted cargo volume as a dynamic covariate, using the time-series prediction model to analyze the time-series correlation between the target variable and the dynamic covariate, and outputting the prediction deviation based on the time-series correlation; and determining the transportation vehicle data included in the initial transportation plan based on the prediction deviation and the historical vehicle loading data.

[0020] Optionally, the transportation information management device is used to generate an initial transportation plan corresponding to the transportation route based on the target transportation model, including: for the case where the target transportation model is a large language model, obtaining the current cargo volume to be transported for the transportation route, the transportation change factors of the transportation route, and the transportation interference factors; constructing prompt words for the large language model based on the current cargo volume to be transported, the historical transportation data, the transportation change factors, the transportation interference factors, and one or more data from multiple decision constraints; and using the large language model to perform reasoning based on the prompt words to output the initial transportation plan corresponding to the transportation route.

[0021] Optionally, the transportation information management device is used to acquire one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data, including: acquiring one or more data among vehicle allocation information, cargo loading information, and transportation feedback information included in the actual transportation data; updating the transportation vehicle data included in the initial transportation plan based on the vehicle allocation information; and / or determining the cargo loading progress based on the cargo loading information; determining the remaining loading capacity of the vehicle based on the cargo loading progress and the vehicle loading capacity, and updating the vehicle loading data included in the initial transportation plan if it is determined that the remaining loading capacity exceeds a preset capacity threshold; and / or identifying transportation anomaly information indicated in the transportation feedback information, outputting a real-time transportation plan based on the transportation anomaly information using a large language model, and adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the real-time transportation plan.

[0022] Optionally, a transportation information management device is used to determine the cargo loading progress based on the cargo loading information, including: acquiring real-time cargo loading progress images at set time intervals; identifying real-time depth information of the loaded cargo in the vehicle's compartment based on the real-time cargo loading progress images; determining the cargo loading progress based on the ratio of the real-time depth information of the loaded cargo to the total depth of the vehicle's compartment; and / or constructing a virtual vehicle compartment image; generating a virtual cargo image superimposed on the virtual vehicle compartment image based on the real-time depth information of the loaded cargo; and displaying the cargo loading progress using the virtual vehicle compartment image and the virtual cargo image.

[0023] To achieve the above objectives, according to a third aspect of the present invention, an electronic device for transport information management is provided, characterized in that it includes: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described above for transport information management.

[0024] To achieve the above objectives, according to a fourth aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, characterized in that, when the program is executed by a processor, it implements any of the methods described in the above-described transportation information management methods.

[0025] To achieve the above objectives, according to a fifth aspect of the present invention, a computer program product is provided, comprising a computer program, characterized in that, when executed by a processor, the computer program implements any of the methods described above for transport information management.

[0026] One embodiment of the above invention has the following advantages or beneficial effects: It can respond to generating a transportation plan for any transportation route by acquiring historical transportation data corresponding to the transportation route, evaluating the fluctuation value of the historical transportation data, selecting a target transportation model matching the fluctuation value from multiple transportation models based on the fluctuation value, generating an initial transportation plan corresponding to the transportation route based on the target transportation model, and further acquiring the actual transportation data corresponding to the transportation route, dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data. This embodiment of the invention divides the transportation plan decision-making for transportation routes into two stages. After generating the initial transportation plan using an adapted model, the initial transportation plan can be adjusted in real time by monitoring actual transportation data, improving the accuracy, real-time nature, and rationality of determining the transportation plan, thereby improving transportation efficiency.

[0027] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0028] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein: Figure 1 This is a flowchart illustrating a transportation information management method provided in an embodiment of the present invention; Figure 2 is a flowchart illustrating a transportation information management method provided in an embodiment of the present invention; Figure 3A This is a schematic diagram illustrating how to determine the loading progress of goods, provided by an embodiment of the present invention. Figure 3B This is a schematic diagram illustrating the progress of cargo loading, provided by an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a transportation information management device provided in an embodiment of the present invention; Figure 5 is an exemplary system architecture diagram in which embodiments of the present invention can be applied; Figure 6This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0029] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0030] Since the transportation plan (parcel diversion) directly determined by the dynamic loading algorithm has a significant impact on operations, an unreasonable diversion plan may not only increase the number of parcel transfers but may also drive up the overall transportation cost.

[0031] The existing methods have the following problems: 1) The accuracy of cargo volume forecasting is low; 2) The real-time generation of vehicle plans is low, and due to uncertainties such as warehouse filling and point misalignment during transportation, the real-time and accuracy of vehicle dispatching are low; 3) The real-time and accuracy of determining vehicle loading data are low; 4) The transportation plan is not reasonable and accurate because it does not take into account transportation interference factors (including transportation diversion caused by other transportation methods such as rail and air transportation in addition to road transportation).

[0032] In view of this, such as Figure 1 As shown, this embodiment of the invention provides a transportation information management method, which may include the following steps: Step S101: In response to generating a transportation plan for any transportation route, obtain the historical transportation data corresponding to the transportation route, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data.

[0033] Step S102: Evaluate the fluctuation value of the historical transportation data, select a target transportation model that matches the fluctuation value from multiple transportation models, and generate an initial transportation plan corresponding to the transportation route based on the target transportation model.

[0034] Specifically, due to the influence of various transportation factors, such as sudden changes in the total amount of goods, abnormal weather, sudden changes in road conditions, and adjustments in transport capacity, there is a need to dynamically generate transport plans for each transport route in the transport network. This necessitates the advance prediction and handling of leftover cargo and idle space. Leftover cargo refers to the excess cargo volume caused when the total amount of goods exceeds the total available space of the vehicle. Idle space refers to the space remaining when the amount of goods is less than the total available space of the vehicle, resulting in the vehicle not being fully loaded.

[0035] Furthermore, in this embodiment of the invention, historical transportation data corresponding to the transportation route is obtained, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data; specifically, the historical transportation data of a certain transportation route is, for example, historical transportation vehicle data (e.g., including vehicle type, number of vehicles, etc. for the transportation route) and historical vehicle loading data (e.g., including total vehicle loading, amount of transported goods, etc.) within the past 30 days.

[0036] Furthermore, the fluctuation values ​​of the historical transportation data are evaluated, and a target transportation model matching the fluctuation values ​​is selected from multiple transportation models based on the fluctuation values.

[0037] Specifically, the method for evaluating the fluctuation value of the historical transportation data can perform cluster analysis on the transportation characteristics of historical transportation data for a certain transportation route, determine the coefficient of variation based on the clustering results, and use the coefficient of variation as the fluctuation value. That is, evaluating the fluctuation value of the historical transportation data and selecting a target transportation model that matches the fluctuation value from multiple transportation models includes: determining a first fluctuation value of the historical transportation vehicle data in the time dimension and a second fluctuation value of the historical vehicle loading data in the time dimension; obtaining transportation change factors and transportation interference factors of the transportation route; and selecting the target transportation model from multiple transportation models based on one or more of the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors. Among them, the first fluctuation value of historical transport vehicle data in the time dimension represents the deviation between the predicted data and the actual data of transport vehicles in the past period; the second fluctuation value of historical vehicle loading data in the time dimension represents the deviation between the predicted data and the actual total amount of cargo loaded by vehicles in the past period; transport change factors include filling (temporarily adding cargo volume, similar to order consolidation) and point-of-departure (adjusting the number of nodes in the route, similar to carpooling) and other interference factors, including transport diversion caused by other means of transport besides road transport such as rail transport and air transport.

[0038] Specifically, based on the transportation fluctuation characteristics indicated by one or more of the data, including the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors, a target transportation model for generating the initial transportation plan is selected. The target transportation model includes a baseline compensation model, a time series prediction model, and a large language model.

[0039] For the benchmark compensation model: Given that the first fluctuation value is below the first fluctuation threshold, the second fluctuation value is below the second fluctuation threshold, there are no transportation change factors, and there are no transportation interference factors, a benchmark compensation model is selected from multiple transportation models as the target transportation model. The fact that the first fluctuation value is below the first fluctuation threshold, the second fluctuation value is below the second fluctuation threshold, there are no transportation change factors, and there are no transportation interference factors indicates that the transportation flow has a stable pattern. For example, the vehicle type and quantity are basically fixed, the volume fluctuation of vehicle loading data is small, there is no complex planning interference such as filling or cross-points, and there is no diversion interference such as railway diversion or air diversion. Based on this stable pattern, fine-tuning can be performed based on historical transportation data to generate the current initial transportation plan. For example, based on the historical transportation plan of the previous day, combined with the predicted cargo quantity of the current day, the historical transportation plan is compensated to obtain the initial transportation plan. The first fluctuation threshold (e.g., coefficient of variation) is 20%, and the second fluctuation threshold is, for example, 5%.

[0040] For the time-series prediction model: Based on one or more of the following data: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors, the target transportation model is selected from multiple transportation models. This includes selecting the time-series prediction model as the target transportation model when it is determined that the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is lower than the fourth fluctuation threshold, there are transportation change factors with time-series characteristics, and there are no transportation interference factors. Specifically, determining that the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is lower than the fourth fluctuation threshold, there are transportation change factors with time-series characteristics, and there are no transportation interference factors indicates that the transportation flow has the characteristic of stable total cargo volume, that is, the vehicle type and quantity change significantly, the vehicle loading data fluctuates little, there is no interference from complex plans such as filling or switching points (or the plan is fixed or has time-series characteristics, for example, the interval between two adjacent filling / switching plans is fixed in days), and there is no interference from railway diversion, air diversion, etc.; wherein, the third fluctuation threshold is, for example, 20%~30%, and the fourth fluctuation threshold is, for example, 15%.

[0041] For the large language model: Based on one or more of the following data: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors, the target transportation model is selected from multiple transportation models. This includes selecting the large language model as the target transportation model when it is determined that the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is higher than the fourth fluctuation threshold, there are transportation change factors, and there are transportation interference factors. Specifically, determining that the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is higher than the fourth fluctuation threshold, there are transportation change factors, and there are transportation interference factors indicates that the transportation flow has complex fluctuation characteristics, that is, the vehicle type and quantity change significantly, the vehicle loading data fluctuates significantly, there are complex plan interferences such as filling or cross-points, and there are diversion interferences such as railway diversion and air diversion. In this case, the reasoning ability of the large language model is used to perform fine-grained scheduling.

[0042] Furthermore, for different situations, the corresponding target transportation model is used to generate the initial transportation plan for the transportation route.

[0043] For cases where the target transportation model serves as the baseline compensation model: Based on historical transportation data (historical vehicle data and historical vehicle loading data), and combined with the current predicted cargo volume (i.e., the total amount of cargo to be transported) for that transportation route, the historical transportation data is compensated and corrected (fine-tuned) to generate the current initial transportation plan. For example: Obtain yesterday's actual transportation data for that transportation route from the transportation production system, and perform vehicle-cargo matching detection by comparing the vehicle departure and arrival times with the sorting and inspection times of the loaded cargo. If there is a discrepancy, the actual package loading information of the dispatching vehicle is corrected according to the sorting and inspection times to obtain accurate vehicle loading data; furthermore, yesterday's dispatch plan is used as the baseline for today's dispatch plan. Based on this, and combined with today's predicted cargo volume to be transported and the loading cargo volume range of historical dispatching vehicle types for this flow, today's dispatch plan is fine-tuned. Fine-tuning includes, for example, adjusting the dispatch plan if the predicted cargo volume to be transported is greater than the existing dispatch plan. If the total loading capacity is greater than the existing vehicle dispatch plan's total loading capacity but does not meet the criteria for additional vehicle dispatch, then additional vehicles will be dispatched. If the predicted cargo volume is greater than the existing vehicle dispatch plan's total loading capacity but does not meet the criteria for additional vehicle dispatch, then the excess cargo volume will be considered as leftover cargo. If the predicted cargo volume is significantly lower than the existing vehicle dispatch plan's total loading capacity and there are empty vehicles, then the number of vehicles will be reduced. If the predicted cargo volume is significantly lower than the existing vehicle dispatch plan's total loading capacity but there are no empty vehicles, then the existing vehicle dispatch plan will not be adjusted, and the remaining space of the current last vehicle will be considered as available empty space for diversion, etc. Finally, today's vehicle dispatch plan (i.e., the initial transportation plan) and the estimated cargo volume and available space information for transportation flow will be determined.

[0044] For the case where the target transportation model is a time-series prediction model: the historical vehicle loading data also includes historical predicted freight volume; generating the initial transportation plan corresponding to the transportation route based on the target transportation model includes: for the case where the target transportation model is a time-series prediction model, obtaining the deviation between historical predicted freight volume and vehicle loading capacity, using the deviation as a target variable, using the historical predicted freight volume as a dynamic covariate, using the time-series prediction model to analyze the time-series correlation between the target variable and the dynamic covariate, and outputting the prediction deviation based on the time-series correlation; and determining the transportation vehicle data included in the initial transportation plan based on the prediction deviation and the historical vehicle loading data. Specifically, the time-series prediction model can be a Moirai model, which learns and analyzes the time-series correlation between the target variable and the dynamic covariate, outputs transportation vehicle data (including vehicle type, quantity, total loading, etc.), and further uses a rule engine to generate the initial transportation plan.

[0045] Furthermore, regarding the analysis of the temporal correlation between the target variable and the dynamic covariate using the aforementioned time-series prediction model, specifically, two key time-series sequences are constructed, with each independent transportation route's corresponding flow direction (e.g., "from A collection and sorting center to B collection and sorting center") as an analysis unit: 1) Target variable (Y): The deviation between historically predicted freight volume and vehicle loading capacity. For example, the daily deviation between predicted freight volume and actual total available vehicle capacity. If the target variable is positive, it indicates that freight demand exceeds expectations, leading to congestion. In this case, the congestion data = deviation = predicted freight volume – actual total available capacity, and the remaining idle space = 0. If the target variable is negative, it indicates that vehicle capacity is excessive. In this case, the congestion data = 0, and the remaining idle space = (-1). Deviation = Actual Total Available Capacity – Forecast Cargo Quantity; 2) Dynamic covariate (X): Historical predicted cargo volume, which serves as the main input feature of the model to capture the impact of cargo volume prediction on the deviation; further, the above-aligned historical time series data (target variable and dynamic covariate) are input into the Moirai model for training. The Moirai model can learn the complex nonlinear relationship between the dynamic covariate (predicted cargo volume) and the target variable (deviation) over time, and model the periodicity, trend and other patterns of the sequence.

[0046] Furthermore, using the trained Moirai model, based on historical sequences within a historical time period (e.g., 90 days) and today's predicted cargo volume, the predicted value of today's deviation is output. Further, today's predicted total available capacity equals today's predicted cargo volume minus today's prediction deviation. Today's predicted total available capacity is input into the rules engine, and combined with the historical loading capacity and cargo volume ranges of each transport vehicle type for each destination, the current initial transport plan is generated. The rules engine's calculation logic includes adjusting the vehicle type and number based on yesterday's actual shipments (i.e., determining the transport vehicle data included in the initial transport plan), with the adjustment goal of being as consistent as possible with today's predicted available capacity.

[0047] For the case where the target transportation model is a large language model: The step of generating the initial transportation plan corresponding to the transportation route based on the target transportation model includes: obtaining the current cargo volume to be transported for the transportation route, transportation change factors, and transportation interference factors; constructing prompt words for the large language model based on the current cargo volume to be transported (e.g., predicted cargo demand), historical transportation data, transportation change factors (e.g., point-of-delivery or warehouse-filling tasks), transportation interference factors (e.g., railway diversion data, air diversion data), and various decision constraints (e.g., weather data); and using the large language model to perform reasoning based on the prompt words to output the initial transportation plan corresponding to the transportation route. Specifically, in this embodiment of the invention, prompt words are constructed to output the initial transportation plan corresponding to the transportation route for the large language model, shaping the large language model into a seasoned dispatcher through the prompt words. The prompt words include: role description, task objectives, business background knowledge, decision rules and basis, and constraints. Simultaneously, based on today's predicted freight volume, a small number of the most similar cases of transportation planning decisions are retrieved and input from historical transportation data (Few-shot learning). Weather information of relevant transportation nodes is obtained from an external weather interface to guide the large model, like a human expert, to comprehensively consider various complex factors and deduce the most reasonable initial transportation plan for today (including vehicle loading data). After obtaining the output information of the large language model, the prompt words constructed in this embodiment of the invention are strictly validated in JSON format and the decision results are initially evaluated. If there are errors that do not meet the format requirements, or that are obviously unreasonable or violate basic constraints or common sense, supplementary prompt words are added to allow the large model to re-decide, which greatly reduces the influence of the large language model's illusion and improves the decision accuracy of the large language model.

[0048] In this embodiment of the invention, the prompt words include system prompt and human prompt; for example, one prompt word is as follows (the prompt word content is contained in quotation marks): System Prompt Example: "###Role You are an experienced vehicle dispatcher. Your daily task is to rationally arrange vehicle transportation plans based on the daily forecast of freight demand and relevant historical cases, and to estimate the remaining amount of cargo and the remaining available vehicle spaces.

[0049] ### Input data content and meaning: (1) Forecast_vol: The sum of the volumes of all packages in cubic meters.

[0050] (2) Start City Weather: Weather in the city where the start city is located.

[0051] (3) destination weather end_city_weather: the weather in the city where the destination is located.

[0052] ### Output data content and meaning: (1) Decision process: Simply output the actual volume estimation process, the full load calculation process of each vehicle, the decision reason for each vehicle, and whether to remove the last vehicle.

[0053] (2) Actual volume of goods today: The actual volume of goods on the day. Forecast_vol and actual_vol may differ. Depending on the mode of transport, actual_vol is equal to the sum of rail_vol, air_vol and road_vol.

[0054] (3) Rail diversion volume rail_vol: The sum of the volumes of parcels selected for rail transport in today's actual freight volume, in cubic meters. Whether or not rail transport is chosen is affected by the weather.

[0055] (4) Air Diversion Volume air_vol: The sum of the volumes of parcels selected for air transport in today's actual cargo volume, in cubic meters. Whether air transport will be selected is affected by the weather.

[0056] (5) Road transport volume road_vol: The sum of the volumes of parcels selected for road transport in today's actual volume, in cubic meters. The road transport volume determines the vehicle transport plan transport_plan.

[0057] (6) Vehicle transportation plan: This needs to include the dispatch order, car model, maximum actual dispatch volume (max_actual_send_vol) for each vehicle, and whether there are any cross-string or fill tasks (has_string_or_fill_task). When cross-string or fill tasks exist, these tasks will occupy vehicle capacity. With this flag present, the maximum actual dispatch volume (max_actual_send_vol) for the same vehicle model will be significantly smaller.

[0058] (7) tail_volume: When the total full load capacity of all vehicles is less than the road transport cargo demand road_vol, there will be a portion of cargo that cannot be loaded. This portion of cargo is called the remaining tail cargo, and the sum of their volumes is called the tail cargo volume, in cubic meters.

[0059] (8) Free space: When the total full load capacity of all vehicles is greater than the road transport cargo volume demand road_vol, there will be some vehicles that are not fully loaded. The total remaining space of these vehicles is called the free space of the vehicles. Its size is equal to the sum of the volumes of the packages that the vehicles can still load, in cubic meters.

[0060] ###Historical Case Explanation: Each line represents a historical decision case for a given date, where stat_dt is the date.

[0061] ### Decision-making principles and precautions: (1) Based on the input forecast demand forecast_vol, and combined with the relationship between the historical forecast demand forecast_vol and the historical actual demand actual_vol in historical cases, decide the actual demand actual_vol.

[0062] (2) The actual volume of goods transported is strictly equal to the sum of the rail diversion volume rail_vol, the air diversion volume air_vol, and the road transport volume road_vol.

[0063] (3) The decision on rail diversion freight volume rail_vol and air diversion freight volume air_vol needs to refer to the weather at the originating point start_city_weather and the weather at the destination point end_city_weather.

[0064] (4) The rail diversion volume rail_vol is limited by available space, and the diversion volume shall not exceed the maximum rail_vol in historical cases. The air diversion volume air_vol is also limited by available space, and the diversion volume shall not exceed the maximum air_vol in historical cases.

[0065] (5) Generate a transport plan based on the road transport volume (road_vol). When making decisions, output the plans sequentially. After dispatching each vehicle, calculate the remaining cargo volume. If the remaining cargo volume > (max_actual_send_vol of the next vehicle)... Only when the volume of cargo reaches 30% or more can another vehicle be dispatched; otherwise, no additional vehicle can be added. For example, if after dispatching the first vehicle, the remaining cargo volume is calculated to be 10 cubic meters, and when dispatching the second vehicle, it is found that the remaining 10 cubic meters is less than the maximum dispatchable volume of the second vehicle, max_actual_send_vol. If 20% are not met, then no second vehicle will be shipped.

[0066] (6) During the dispatching process, it is also necessary to consider the daily dispatching times in history. If the history has always been single dispatching, the probability of dispatching two dispatching times today is not high. Therefore, it is necessary to make a judgment based on historical data.

[0067] (7) The fill task has_string_or_fill_task marks the corresponding train for today according to the historical shipping pattern. In practice, it is necessary to prioritize the completion of the fill task. Therefore, if the frequency of fill has been high in the past, there should also be a fill task today.

[0068] (8) When arranging vehicle transportation plans, you need to adhere to the principle of cost-saving, ensuring each vehicle is loaded to the maximum extent possible and avoiding underloaded shipments. Please note that at most one value between the remaining cargo quantity and available space can be non-zero.

[0069] Human Prompt Example: "###Historical Case Reference" {example_prompt} ### Supplementary prompts (derived from JSON format verification agent and result evaluation agent) {In the output, set the maximum shipment volume (max_actual_send_vol) for each vehicle. Do not directly use historical values; calculate a reasonable value based on historical patterns for the same vehicle model. Do not directly use the historical maximum value for that vehicle model.} ###Task Today's forecasted freight demand is `forcast_vol` ({forcast_vol}), the originating point weather is `start_city_weather` ({start_city_weather}), and the destination point weather is `end_city_weather` ({end_city_weather}). Please make decisions regarding today's vehicle transportation plan, remaining cargo volume, and available parking spaces. Note that the output format must be consistent with the example, and the output fields must match the specifications in `output`, outputting in JSON format. Your answer is: By constructing prompt words and a large language model, the rationality and effectiveness of transportation planning under complex and fluctuating conditions have been improved.

[0070] Step S103: Obtain one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

[0071] In this embodiment of the invention, the initial transportation plan generated in steps S101 to S102 is further adjusted and optimized by combining the actual transportation data acquired in real time.

[0072] Specifically, the step of acquiring one or more actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data, includes: acquiring one or more data among vehicle allocation information, cargo loading information, and transportation feedback information included in the actual transportation data; updating the transportation vehicle data included in the initial transportation plan based on the vehicle allocation information; and / or determining the cargo loading progress based on the cargo loading information; determining the remaining loading capacity of the vehicle based on the cargo loading progress and the vehicle loading capacity; and updating the vehicle loading data included in the initial transportation plan if it is determined that the remaining loading capacity exceeds a preset capacity threshold.

[0073] This can involve obtaining structured vehicle allocation information from a TMS (Transportation Management System), including vehicle type, vehicle identification (license plate number), vehicle loading capacity, point-to-point tasks, and warehouse filling tasks. Furthermore, the initial transportation plan's vehicle data is updated based on this vehicle allocation information. For example, if there's a need for fine-tuning due to leftover cargo / vacant space, the transportation vehicle data (vehicle type, vehicle loading capacity, etc.) is updated to address the issue of leftover cargo / vacant space.

[0074] Furthermore, images of cargo loading are captured using imaging equipment to identify the cargo loading progress.

[0075] Specifically, according to a set loading time interval (e.g., 5 minutes), a binocular starlight camera installed at each sorting and loading bay can be used to collect video streams, acquire the video streams, and perform keyframe sampling. A depth map is generated for each sampled frame. The system identifies the internal space boundaries of the vehicle's cargo compartment, segments and identifies point cloud clusters belonging to the loaded packages, and filters out dynamic interference such as personnel, handling equipment, and conveyor belts. The segmented "cargo point cloud clusters" are then fitted with minimum volume bounding boxes. Based on the fitting results, the actual total volume occupied by the loaded cargo (real-time depth information) is estimated in real time. Furthermore, the real-time depth information of the loaded cargo is compared with the total volume of the vehicle's cargo compartment. The cargo loading progress is determined by the proportion of depth; that is, determining the cargo loading progress based on the cargo loading information includes: acquiring real-time cargo loading progress images at set time intervals; identifying real-time depth information of the loaded cargo in the vehicle's compartment based on the real-time cargo loading progress images; determining the cargo loading progress based on the proportion of the real-time depth information of the loaded cargo to the total depth of the vehicle's compartment; and / or constructing a virtual vehicle compartment image; generating a virtual cargo image superimposed on the virtual vehicle compartment image based on the real-time depth information of the loaded cargo; and displaying the cargo loading progress using the virtual vehicle compartment image and the virtual cargo image.

[0076] Specifically, Figure 3A The diagram illustrates a method for determining cargo loading progress, which includes a camera and a vehicle diagram. The system can identify the vehicle's license plate number by capturing images of the vehicle and retrieve basic vehicle information (such as vehicle type, loading capacity, and total depth of the cargo compartment) from the TMS based on the license plate number. Furthermore, AI vision can be used to identify real-time depth information (i.e., loaded depth) and remaining depth (i.e., available space in the cargo compartment) based on image frames captured by the camera video. The cargo loading progress can also be determined by the ratio of the loaded depth to the total vehicle length (i.e., total depth of the cargo compartment).

[0077] Furthermore, based on vehicle loading capacity or vehicle compartment volume data, a virtual vehicle compartment image is constructed; based on the real-time depth information of the loaded goods, a virtual cargo image is generated and superimposed on the virtual vehicle compartment image; using the virtual compartment image and the virtual cargo image, the loading progress of the goods is displayed. Specifically, embodiments of the present invention can construct a virtual simulation image of the vehicle (virtual compartment image), and combine it with camera calibration parameters to construct an instantaneous three-dimensional point cloud scene inside the compartment; based on the real-time depth information of the loaded goods, a virtual cargo image is generated and superimposed on the virtual vehicle compartment image; the loading progress of the goods is displayed. Figure 3BA schematic diagram is shown of a virtual truck bed image of a vehicle displaying the progress of cargo loading. By using the virtual truck bed image and the virtual cargo image superimposed on the virtual truck bed image, the loading progress can be visualized and the changes in the loading progress can be monitored in real time.

[0078] Furthermore, in this embodiment of the invention, the abnormal transportation situation information indicated in the transportation feedback information is identified, a real-time transportation plan is output based on the abnormal transportation situation information using a large language model, and the transportation vehicle data and / or vehicle loading data included in the initial transportation plan are adjusted based on the real-time transportation plan.

[0079] Specifically, transportation feedback information can include voice data from staff performing cargo loading. Staff can report voice data through dialogue with robots (text) or smart terminals, providing feedback on temporary unforeseen circumstances during loading or transportation (i.e., transportation anomaly information). For example, transportation feedback information may include: "An extra loading truck has been added to route XX," "The loading truck for route XX has arrived at the platform, with an estimated cargo volume of 60 cubic meters," "The connecting point plan for route XX has been cancelled," "The cargo volume on the XX loading line is severely backlogged," "A large amount of cargo has arrived at the site, and the cargo volume for route XX has suddenly increased, estimated at 20 cubic meters," "No trucks are available at the platform for route XX, and a temporary diversion plan is needed," etc.

[0080] Furthermore, for voice feedback information, the ASR (Automatic Speech Recognition) service can be invoked to convert it into text information; then, the text information is fed into a large language model for semantic parsing. Considering that frontline feedback information is often extremely brief, in order to achieve semantic parsing of it by the large language model, when frontline feedback information is obtained, the corresponding initial transportation plan is used to query plan details, algorithm decision logs, and algorithm input data snapshots, etc., to enhance and expand the context of the feedback information, providing it to the large language model to accurately identify the information of the plan to be adjusted, and converting it into structured flow dimension data (such as: transportation route flow direction identifier, adjustment object, adjustment type, adjustment amount) and plan dimension data (such as: plan identifier, operation instructions: add / pause / modify / terminate, adjustment content), driving the dynamic loading algorithm engine to adjust the diversion plan and loading instructions (i.e., the transportation vehicle data and / or vehicle loading data included in the initial transportation plan) in real time.

[0081] This invention employs a transportation planning management framework of "scenario-adaptive algorithm matching + multimodal data dynamic correction," dividing the entire transportation plan into two stages: initial transportation plan generation and real-time dynamic correction. In the initial transportation plan generation stage, the most suitable model (baseline compensation model / time-series prediction model / large language model) is matched based on the "vehicle dispatch stability" characteristics of the route flow. In the real-time dynamic correction stage, the initial transportation plan is dynamically corrected in real time using multimodal data such as transportation system data, loading video recognition, and on-site human-machine voice / text interaction (feedback information), forming a complete closed loop from "static prediction" to "dynamic perception." This improves the accuracy of predictions regarding "leftover cargo or idle loading spaces" and the rationality of transportation scheduling.

[0082] like Figure 2 As shown in the figure, an embodiment of the present invention provides a flowchart of a transportation information management method, which may include the following steps: Step S201: Generate an initial transportation plan.

[0083] Specifically, in response to generating a transportation plan for any transportation route, historical transportation data corresponding to the transportation route is obtained, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data; the fluctuation value of the historical transportation data is evaluated, and a target transportation model matching the fluctuation value is selected from multiple transportation models (baseline compensation model, time series prediction model, and large language model) based on the fluctuation value; and an initial transportation plan corresponding to the transportation route is generated based on the target transportation model.

[0084] Step S202: Dynamically adjust the initial transportation plan.

[0085] Specifically, one or more actual transportation data corresponding to the transportation route are obtained, and the transportation vehicle data and / or vehicle loading data included in the initial transportation plan are dynamically adjusted based on the actual transportation data.

[0086] This includes obtaining vehicle allocation information (i.e., actual transportation data), comparing the vehicle allocation information with the transportation vehicle data and / or vehicle loading data included in the initial transportation plan, and adjusting the initial transportation plan accordingly.

[0087] Obtain cargo loading information (i.e., actual transportation data), determine the cargo loading progress based on the cargo loading information, and adjust the initial transportation plan based on the cargo loading progress.

[0088] Obtain transportation feedback information (i.e., actual transportation data), identify transportation anomalies indicated in the transportation feedback information, and adjust the initial transportation plan based on the transportation anomaly information.

[0089] The descriptions of steps S201 to S202 are the same as those of steps S101 to S103, and will not be repeated here.

[0090] The embodiments of the present invention construct an intelligent prediction system capable of sensing, reasoning, and rapidly responding to changes on-site through the technical solution of "scenario adaptive algorithm matching + multimodal data dynamic correction" described in steps S201 to S202. This fundamentally solves the problem of low accuracy of transportation plans in existing methods (e.g., inaccurate prediction of leftover cargo / idle space), and improves the accuracy and efficiency of dynamic loading schemes.

[0091] like Figure 4 As shown, this embodiment of the invention provides a transportation information management device 400, including: a data acquisition module 401, a plan generation module 402, and a plan adjustment module 403; wherein, The data acquisition module 401 is used to acquire historical transportation data corresponding to the transportation route in response to generating a transportation plan for any transportation route, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data. The plan generation module 402 is used to evaluate the fluctuation value of the historical transportation data, select a target transportation model that matches the fluctuation value from multiple transportation models, and generate an initial transportation plan corresponding to the transportation route based on the target transportation model. The adjustment plan module 403 is used to acquire one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

[0092] This invention also provides an electronic device for transportation information management, comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in any of the above embodiments.

[0093] This invention also provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0094] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the methods described above for transportation information management.

[0095] Figure 5 An exemplary system architecture 500 is shown for which the transportation information management method or transportation information management device of the present invention can be applied.

[0096] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, and 503, a network 504, and a server 505. Network 504 serves as the medium for providing communication links between terminal devices 501, 502, and 503 and server 505. Network 504 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0097] Users can use terminal devices 501, 502, and 503 to interact with server 505 via network 504 to receive or send messages, etc. Various client applications can be installed on terminal devices 501, 502, and 503, such as e-commerce client applications, web browser applications, search applications, instant messaging tools, and email clients.

[0098] Terminal devices 501, 502, and 503 can be various electronic devices with displays and supporting various client applications, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0099] Server 505 can be a server that provides various services, such as a backend management server that supports client applications used by users through terminal devices 501, 502, and 503. The backend management server can process received requests to generate transportation plans for any transportation route and feed back the initial or adjusted transportation plan for that route to the terminal device.

[0100] It should be noted that the transportation information management method provided in this embodiment of the invention is generally executed by server 505, and correspondingly, the transportation information management device is generally set in server 505.

[0101] It should be understood that Figure 5 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0102] The following is for reference. Figure 6 It shows a schematic diagram of the structure of a computer system 600 suitable for implementing a terminal device of the present invention. Figure 6 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0103] like Figure 6As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 602 or programs loaded from storage section 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the system 600. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0104] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0105] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs the functions defined above in the system of this invention.

[0106] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0107] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0108] The modules and / or units described in the embodiments of the present invention can be implemented in software or hardware. The described modules and / or units can also be housed in a processor; for example, a processor can be described as including a data acquisition module, a plan generation module, and a plan adjustment module. The names of these modules do not necessarily limit the module itself; for example, the data acquisition module can also be described as a "module for acquiring historical transportation data corresponding to a transportation route."

[0109] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs that, when executed by the device, cause the device to include: responding to generate a transportation plan for any transportation route by acquiring historical transportation data corresponding to the transportation route, evaluating the fluctuation value of the historical transportation data, selecting a target transportation model matching the fluctuation value from multiple transportation models based on the fluctuation value; generating an initial transportation plan corresponding to the transportation route based on the target transportation model; further acquiring actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data. By dividing the transportation plan decision for a transportation route into two stages, after generating the initial transportation plan using an adapted model, the initial transportation plan can be adjusted in real time by monitoring actual transportation data, improving the accuracy, real-time nature, and rationality of determining the transportation plan, thereby improving transportation efficiency.

[0110] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A transportation information management method, characterized in that, include: In response to generating a transportation plan for any transportation route, historical transportation data corresponding to the transportation route is obtained, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data; The fluctuation values ​​of the historical transportation data are evaluated, and a target transportation model matching the fluctuation values ​​is selected from multiple transportation models based on the fluctuation values; based on the target transportation model, an initial transportation plan corresponding to the transportation route is generated. Obtain one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

2. The method according to claim 1, characterized in that, The evaluation of the fluctuation values ​​of the historical transportation data, and the selection of a target transportation model that matches the fluctuation values ​​from a variety of transportation models, includes: Determine the first fluctuation value of the historical transport vehicle data in the time dimension and the second fluctuation value of the historical vehicle loading data in the time dimension; Obtain the factors affecting transportation changes and factors causing transportation disruptions to the transportation route; Based on one or more of the following data: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors, the target transportation model is selected from multiple transportation models.

3. The method according to claim 2, characterized in that, The step of selecting the target transportation model from multiple transportation models based on one or more of the following: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors; If it is determined that the first fluctuation value is lower than the first fluctuation threshold, the second fluctuation value is lower than the second fluctuation threshold, there are no transportation change factors, and there are no transportation interference factors, a benchmark compensation model is selected as the target transportation model from multiple transportation models.

4. The method according to claim 2, characterized in that, The step of selecting the target transportation model from multiple transportation models based on one or more of the following: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors; If the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is lower than the fourth fluctuation threshold, there are transportation change factors with time-series characteristics, and there are no transportation interference factors, a time-series prediction model is selected from multiple transportation models as the target transportation model.

5. The method according to claim 2, characterized in that, The step of selecting the target transportation model from multiple transportation models based on one or more of the following: the first fluctuation value, the second fluctuation value, the transportation change factors, and the transportation interference factors; If it is determined that the first fluctuation value is higher than the third fluctuation threshold, the second fluctuation value is higher than the fourth fluctuation threshold, there are transportation change factors, and there are transportation interference factors, a large language model is selected as the target transportation model from multiple transportation models.

6. The method according to claim 1, characterized in that, The historical vehicle loading data also includes historical predicted cargo volume; The step of generating an initial transportation plan corresponding to the transportation route based on the target transportation model includes: For the case where the target transportation model is a time-series prediction model, The deviation between the historical predicted cargo volume and the vehicle loading capacity is obtained. The deviation is used as the target variable and the historical predicted cargo volume is used as the dynamic covariate. The time series prediction model is used to analyze the time series correlation between the target variable and the dynamic covariate, and the prediction deviation is output based on the time series correlation. Based on the prediction deviation and the historical vehicle loading data, the transport vehicle data included in the initial transport plan is determined.

7. The method according to claim 1, characterized in that, The step of generating an initial transportation plan corresponding to the transportation route based on the target transportation model includes: For the case where the target transportation model is a large language model, Obtain the current cargo volume to be transported for the transport route, the transport change factors for the transport route, and the transport interference factors; Based on the current cargo volume to be transported, the historical transportation data, the transportation change factors, the transportation interference factors, and one or more of the various decision constraints, prompt words are constructed for the large language model; The large language model is used to perform inference based on the prompt words, and the initial transportation plan corresponding to the transportation route is output.

8. The method according to claim 1, characterized in that, The step of acquiring one or more actual transportation data corresponding to the transportation route, and dynamically adjusting the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data, includes: The actual transportation data includes one or more of the following: vehicle allocation information, cargo loading information, and transportation feedback information. Update the transport vehicle data included in the initial transport plan based on the vehicle allocation information; And / or, The loading progress is determined based on the cargo loading information; the remaining loading capacity of the vehicle is determined based on the cargo loading progress and the vehicle loading capacity; if the remaining loading capacity exceeds a preset capacity threshold, the vehicle loading data included in the initial transportation plan is updated. And / or, Identify the transportation anomaly information indicated in the transportation feedback information, output a real-time transportation plan based on the transportation anomaly information using a large language model, and adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the real-time transportation plan.

9. The method according to claim 8, characterized in that, Determining the cargo loading progress based on the cargo loading information includes: Real-time cargo loading progress images are collected at set time intervals, and real-time depth information of the loaded cargo in the vehicle's compartment is identified based on the real-time cargo loading progress images. The loading progress of the cargo is determined based on the ratio of the real-time depth information of the loaded cargo to the total depth of the vehicle's cargo compartment. And / or, Construct a virtual vehicle compartment image; generate a virtual cargo image superimposed on the virtual vehicle compartment image based on the real-time depth information of the loaded cargo; and display the cargo loading progress using the virtual vehicle compartment image and the virtual cargo image.

10. A transportation information management device, characterized in that, include: The module comprises a data acquisition module, a plan generation module, and a plan adjustment module; among which, The data acquisition module is used to acquire historical transportation data corresponding to the transportation route in response to generating a transportation plan for any transportation route, wherein the historical transportation data includes historical transportation vehicle data and historical vehicle loading data. The plan generation module is used to evaluate the fluctuation value of the historical transportation data, select a target transportation model that matches the fluctuation value from multiple transportation models, and generate an initial transportation plan corresponding to the transportation route based on the target transportation model. The adjustment plan module is used to acquire one or more actual transportation data corresponding to the transportation route, and dynamically adjust the transportation vehicle data and / or vehicle loading data included in the initial transportation plan based on the actual transportation data.

11. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-9.

12. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-9.

13. A computer program product, comprising a computer program, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-9.

Images