A method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis
Through the logistics transportation forecasting and scheduling method based on data analysis and deep learning, the problem that traditional scheduling methods are difficult to cope with dynamic factors is solved, and the efficiency, safety and cost reduction of logistics transportation is achieved.
Patent Information
- Application Number
- CN202411248136.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-09-06
AI Technical Summary
Traditional logistics scheduling methods are difficult to adjust and optimize in real time, and cannot effectively deal with dynamic factors such as road congestion, weather changes and route deviations, resulting in waste, delays and increased costs of logistics and transportation resources, affecting operational efficiency and service quality.
Using the block logistics transportation prediction and scheduling optimization method based on data analysis, the road path planning is constructed by acquiring and preprocessing logistics data, and multi-dimensional scheduling is carried out by combining driving recorder video and weather and meteorological data to generate an optimized scheduling plan, and through Simulation path test verifies the feasibility of the solution.
It improves the accuracy and real-timeness of logistics transportation path planning, ensures the reliability and feasibility of the scheduling plan, significantly improves transportation efficiency and safety, reduces transportation costs, and enhances the ability to respond to emergencies.
Smart Images

Figure CN118780571B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data analysis, and in particular to a bulk logistics transportation forecasting and scheduling optimization method based on data analysis. Background Art
[0002] With the continuous growth of demand for bulk logistics transportation, traditional logistics scheduling methods have problems such as low efficiency, high cost, and slow response when dealing with complex and changing transportation environments. Traditional methods usually rely on manual experience and static planning, which are difficult to adjust and optimize in real time, and cannot effectively deal with dynamic factors such as road congestion, weather changes, and route deviations. At the same time, the existing logistics scheduling system lacks in-depth analysis and utilization of historical transportation data, and fails to give full play to the advantages of big data and artificial intelligence technology. These problems lead to waste of resources, transportation delays, and increased costs in the process of logistics transportation, seriously affecting the operational efficiency and service quality of logistics companies. To solve these problems, a bulk logistics transportation prediction and scheduling method based on data analysis and intelligent optimization is urgently needed to improve the overall efficiency and reliability of logistics transportation, reduce transportation costs, and improve the level of logistics services. Summary of the invention
[0003] Based on this, it is necessary to provide a bulk logistics transportation forecasting and scheduling optimization method based on data analysis to solve at least one of the above technical problems.
[0004] To achieve the above purpose, a method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis is provided, the method comprising the following steps:
[0005] Step S1: Obtaining logistics and transportation information data; performing data preprocessing on the logistics and transportation information data to generate standard logistics data;
[0006] Step S2: construct a deep learning model for bulk logistics; input standard logistics data into the deep learning model for bulk logistics to perform road path planning and generate road route planning data;
[0007] Step S3: Acquire dashcam video and weather data; confirm the starting point location based on the road route planning data to obtain the starting point location data; extract key frames from the dashcam video to generate dashcam key frame images; perform route fitting analysis on the road route planning data through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route; perform vehicle identification on the dashcam key frame images to generate vehicle record data; perform congestion transportation scheduling on the vehicle record data according to the preset road congestion threshold to generate a road congestion scheduling plan; perform severe transportation scheduling on the actual transportation route and the road route planning data based on the weather and meteorological data to generate a road weather scheduling plan; perform transportation destination difference scheduling based on the actual transportation route and the road route planning data to generate a road distance transportation scheduling plan; merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation forecast optimization scheduling plan;
[0008] Step S4: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation prediction optimization scheduling plan to generate logistics simulation test results; use historical transportation data to perform simulation testing on the logistics simulation test results to accurately verify and generate bulk logistics transportation test results; compare the logistics transportation indicators of the bulk logistics transportation prediction optimization scheduling plan and the logistics simulation test results to generate bulk logistics transportation indicator results; conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation indicator results to generate bulk logistics feasibility analysis results.
[0009] The present invention first obtains logistics transportation information data and pre-processes these data to generate standard logistics data. The pre-processing process includes data cleaning, data consistency check and data standardization processing to ensure the accuracy and consistency of the data. Then, a bulk logistics deep learning model is constructed, which consists of a convolutional neural network layer and a long short-term memory layer. The standard logistics data is input into the deep learning model, and road path planning is performed to generate road route planning data. Then, by obtaining a dashcam video and weather and meteorological data, the starting point location is confirmed based on the road route planning data, key frame images are extracted from the dashcam video, and the road route planning data is subjected to route fitting analysis through the vehicle-mounted GPS and the starting point location data to generate an actual transportation route. Vehicle identification is performed on the dashcam key frame image to generate vehicle record data; then, according to the actual transportation route, weather and meteorological data and vehicle record data, multi-dimensional logistics transportation scheduling is performed to generate a bulk logistics transportation forecast optimization scheduling plan. Finally, historical transportation data is obtained, and a simulation path test is performed on the bulk logistics transportation forecast optimization scheduling plan to generate a logistics simulation test result. The simulation test results are accurately verified using historical transportation data to generate bulk logistics transportation test results. The logistics transportation indicators of the predicted optimization scheduling plan and the simulation test results are then compared to generate bulk logistics transportation indicator results. The bulk logistics transportation test results and bulk logistics transportation indicator results are comprehensively evaluated for feasibility to generate bulk logistics feasibility analysis results.
[0010] The beneficial effect of the present invention is that a comprehensive process from data acquisition, processing to final scheduling optimization is realized through multiple steps. First, by acquiring logistics and transportation information data and performing preprocessing, the accuracy and consistency of the data are ensured, and standard logistics data is generated; secondly, a deep learning model for bulk logistics is constructed, and the standard logistics data is input into the model for road path planning to generate road route planning data. Next, the dashcam video and weather and meteorological data are obtained, the starting point location is confirmed through the road route planning data, and key frame images are extracted from the dashcam video; the road route planning data is subjected to route fitting analysis through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route. Vehicle identification is performed on the key frame images of the dashcam to generate vehicle recording data, and multi-dimensional scheduling of logistics transportation is performed based on the actual transportation route, weather and meteorological data, and vehicle recording data to generate a bulk logistics transportation forecast optimization scheduling plan. Finally, historical transportation data is obtained to simulate the path test of the bulk logistics transportation forecast optimization scheduling plan and generate logistics simulation test results; the logistics simulation test results are accurately verified by simulation test using historical transportation data to generate bulk logistics transportation test results, and the logistics transportation indicators of the bulk logistics transportation forecast optimization scheduling plan and the logistics simulation test results are compared to generate bulk logistics transportation indicator results, and finally, through comprehensive feasibility evaluation, the bulk logistics feasibility analysis results are generated. This method not only improves the accuracy and real-time performance of logistics transportation path planning through comprehensive analysis of multi-source data and the application of deep learning models, but also ensures the reliability and feasibility of the scheduling plan through simulation testing and historical data verification, significantly improves the efficiency and safety of bulk logistics transportation, effectively reduces transportation costs, and enhances the ability to respond to emergencies, thereby achieving all-round optimization of logistics transportation. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 A schematic diagram of the steps of a method for optimizing the forecasting and scheduling of bulk logistics transportation based on data analysis;
[0012] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;
[0013] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.
[0014] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG.
[0015] 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
[0016] The following is a clear and complete description of the technical method of the present invention 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.
[0017] 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.
[0018] 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 unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0019] To achieve this, please refer to Figures 1 to 4 , a method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis, the method comprising the following steps:
[0020] Step S1: Obtaining logistics and transportation information data; performing data preprocessing on the logistics and transportation information data to generate standard logistics data;
[0021] Step S2: construct a deep learning model for bulk logistics; input standard logistics data into the deep learning model for bulk logistics to perform road path planning and generate road route planning data;
[0022] Step S3: Acquire dashcam video and weather data; confirm the starting point location based on the road route planning data to obtain the starting point location data; extract key frames from the dashcam video to generate dashcam key frame images; perform route fitting analysis on the road route planning data through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route; perform vehicle identification on the dashcam key frame images to generate vehicle record data; perform congestion transportation scheduling on the vehicle record data according to the preset road congestion threshold to generate a road congestion scheduling plan; perform severe transportation scheduling on the actual transportation route and the road route planning data based on the weather and meteorological data to generate a road weather scheduling plan; perform transportation destination difference scheduling based on the actual transportation route and the road route planning data to generate a road distance transportation scheduling plan; merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation forecast optimization scheduling plan;
[0023] Step S4: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation prediction optimization scheduling plan to generate logistics simulation test results; use historical transportation data to perform simulation testing on the logistics simulation test results to accurately verify and generate bulk logistics transportation test results; compare the logistics transportation indicators of the bulk logistics transportation prediction optimization scheduling plan and the logistics simulation test results to generate bulk logistics transportation indicator results; conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation indicator results to generate bulk logistics feasibility analysis results.
[0024] The present invention first obtains logistics transportation information data and pre-processes these data to generate standard logistics data. The pre-processing process includes data cleaning, data consistency check and data standardization processing to ensure the accuracy and consistency of the data. Then, a bulk logistics deep learning model is constructed, which consists of a convolutional neural network layer and a long short-term memory layer. The standard logistics data is input into the deep learning model, and the road path planning is performed to generate the road route planning data. Then, by obtaining the dashcam video and weather and meteorological data, the starting point location is confirmed based on the road route planning data, the key frame image is extracted from the dashcam video, and the road route planning data is subjected to route fitting analysis through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route. The key frame image of the dashcam is subjected to vehicle identification to generate vehicle recording data; then, according to the actual transportation route, weather and meteorological data and vehicle recording data, the multi-dimensional scheduling of logistics transportation is merged, and the time series is adjusted to generate the bulk logistics transportation prediction optimization scheduling plan. Finally, the historical transportation data is obtained, the bulk logistics transportation prediction optimization scheduling plan is subjected to simulation path testing, and the logistics simulation test results are generated. The simulation test results are accurately verified using historical transportation data to generate bulk logistics transportation test results. The logistics transportation indicators of the predicted optimization scheduling plan and the simulation test results are then compared to generate bulk logistics transportation indicator results. The bulk logistics transportation test results and bulk logistics transportation indicator results are comprehensively evaluated for feasibility to generate bulk logistics feasibility analysis results.
[0025] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a bulk logistics transportation forecasting and scheduling optimization method based on data analysis of the present invention. In this example, the bulk logistics transportation forecasting and scheduling optimization method based on data analysis includes the following steps:
[0026] Step S1: Obtaining logistics and transportation information data; performing data preprocessing on the logistics and transportation information data to generate standard logistics data;
[0027] In an embodiment of the present invention, various data involved in the logistics transportation process are collected in real time through a logistics information system or an Internet of Things device (such as a sensor, RFID, GPS, etc.), including but not limited to the location information, speed, fuel consumption, cargo status, and transportation route of the transportation vehicle. Then, the collected logistics transportation information data is cleaned to remove noise data and redundant data to ensure the accuracy and completeness of the data. Next, a data consistency check is performed to ensure the consistency and reliability of the data by comparing data from different data sources. Subsequently, data conversion and standardization technology is used to convert data of different formats and sources into a unified standard format to generate standard logistics data. During the standardization process, the technical means used include data format conversion, missing value filling, data normalization, etc., to ensure that data from different sources can be effectively integrated and analyzed on a unified platform.
[0028] Step S2: construct a deep learning model for bulk logistics; input standard logistics data into the deep learning model for bulk logistics to perform road path planning and generate road route planning data;
[0029] In the embodiment of the present invention, by receiving standard logistics data as input, after training and learning, the optimal road route planning data can be predicted and generated. This method combines the advantages of deep learning in data modeling and prediction, can effectively handle complex logistics networks and large amounts of data, and provide efficient and accurate path planning solutions.
[0030] Step S3: Acquire dashcam video and weather data; confirm the starting point location based on the road route planning data to obtain the starting point location data; extract key frames from the dashcam video to generate dashcam key frame images; perform route fitting analysis on the road route planning data through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route; perform vehicle identification on the dashcam key frame images to generate vehicle record data; perform congestion transportation scheduling on the vehicle record data according to the preset road congestion threshold to generate a road congestion scheduling plan; perform severe transportation scheduling on the actual transportation route and the road route planning data based on the weather and meteorological data to generate a road weather scheduling plan; perform transportation destination difference scheduling based on the actual transportation route and the road route planning data to generate a road distance transportation scheduling plan; merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation forecast optimization scheduling plan;
[0031] In the embodiment of the present invention, by acquiring the dashcam video and weather and meteorological data, the system obtains real-time driving and weather information. The dashcam video provides visual data during the vehicle driving process, while the weather and meteorological data include key factors affecting transportation safety and efficiency such as the slippery degree of the road surface and visibility. The departure point location is confirmed based on the road route planning data, and the precise location of the departure point is determined using the vehicle-mounted GPS data. This step ensures the accuracy of subsequent route analysis and data processing. Key frames are extracted from the dashcam video to generate dashcam key frame images. These key frame images can extract key information such as vehicle identification, vehicle location, etc. through vehicle recognition technology, thereby generating vehicle record data. The road route planning data is route-fitted and analyzed through the vehicle-mounted GPS data and the departure point location data to generate the actual transportation route. This step combines the real-time vehicle location and driving path information, compares and adjusts with the preset route planning, and ensures the practicality and effectiveness of the route. Finally, based on the actual transportation route, weather and meteorological data, and vehicle record data, the system performs multi-dimensional scheduling of logistics transportation. This includes comprehensively considering factors such as the real-time traffic conditions of the route, the impact of weather conditions on transportation, and the operating status of the vehicle to generate an optimized logistics transportation scheduling plan.
[0032] Step S4: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation prediction optimization scheduling plan to generate logistics simulation test results; use historical transportation data to perform simulation testing on the logistics simulation test results to accurately verify and generate bulk logistics transportation test results; compare the logistics transportation indicators of the bulk logistics transportation prediction optimization scheduling plan and the logistics simulation test results to generate bulk logistics transportation indicator results; conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation indicator results to generate bulk logistics feasibility analysis results.
[0033] In the embodiment of the present invention, by acquiring historical transportation data, the system acquires past transportation records and related indicator data, including transportation time, cost, efficiency and other information. A simulation path test is performed on the prediction and optimization scheduling scheme for bulk logistics transportation. This step uses simulation technology to simulate decisions such as route selection and vehicle allocation in actual transportation scenarios to generate logistics simulation test results. The logistics simulation test results are accurately verified using historical transportation data. By comparing and analyzing with actual historical data, the accuracy and reliability of the simulation test are verified to generate bulk logistics transportation test results. The logistics transportation indicators of the prediction and optimization scheduling scheme for bulk logistics transportation and the logistics simulation test results are compared. This includes comparing the performance of the prediction scheme and the simulation test in terms of transportation efficiency, cost control, etc., and generating bulk logistics transportation indicator results. Finally, the bulk logistics transportation test results and the bulk logistics transportation indicator results are comprehensively evaluated for feasibility. This step considers the comprehensive impact of various indicators, evaluates the actual feasibility and economic benefits of the prediction scheme, and generates bulk logistics feasibility analysis results.
[0034] Preferably, step S1 comprises the following steps:
[0035] Step S11: Obtaining logistics transportation information data;
[0036] Step S12: performing data cleaning on the logistics transportation information data to generate logistics cleaning data;
[0037] Step S13: Performing data consistency check on the logistics cleaning data to generate logistics consistent data;
[0038] Step S14: Standardize the logistics consistency data to generate standard logistics data.
[0039] The present invention eliminates errors and inconsistencies in the data through cleaning, consistency checking and standardization processing, thereby ensuring that the basic data for subsequent analysis and decision-making is reliable and accurate. Standardization processing allows information from different data sources to be accurately understood and processed by the system, providing a consistent data basis for subsequent logistics management and decision-making. Finally, by processing and optimizing data, the system can more accurately analyze and predict transportation needs, optimize route planning and resource scheduling, thereby improving transportation efficiency and reducing costs. In summary, these data processing steps not only improve the quality and consistency of the data, but also provide key technical support for the intelligence and optimization of the logistics and transportation management system, and promote the modernization and sustainable development of the logistics industry.
[0040] In an embodiment of the present invention, data related to logistics and transportation are first obtained from various sources. These data cover various key information in the logistics process, such as the location of the goods, transportation time, transportation method, temperature records, etc. Acquiring these data involves technologies such as automated sensors, handheld devices, and GPS systems. The collected raw data is preprocessed. Data cleaning includes removing duplicate data, correcting erroneous data, filling missing data, and unifying data format operations. These steps ensure the quality and consistency of the data and prepare for subsequent analysis and application. To ensure the consistency of the cleaned data when collected from different sources or at different times. This includes comparing the difference between the data and the expected value, detecting logical errors or inconsistencies between the data, and standardizing the processing method of the data. The data that has been checked for consistency is standardized. Standardization includes unifying data units, data formats, naming conventions, etc. to ensure that data from different sources can be understood and integrated by the system and improve the availability and operability of the data.
[0041] Preferably, step S2 comprises the following steps:
[0042] Step S21: constructing a deep learning model for bulk logistics, wherein the deep learning model for bulk logistics includes a convolutional neural network layer and a long short-term memory layer;
[0043] Step S22: input the standard logistics data into the convolutional neural network layer in the bulk logistics deep learning model to perform logistics space feature recognition processing to generate road section connection data;
[0044] Step S23: using the long short-term memory layer in the bulk logistics deep learning model to perform learning time-dependent processing on the road section connection data to generate time-dependent pre-data;
[0045] Step S24: Perform path optimization analysis on the time-dependent prediction data through a path optimization algorithm to generate road route planning data.
[0046] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:
[0047] Step S21: constructing a deep learning model for bulk logistics, wherein the deep learning model for bulk logistics includes a convolutional neural network layer and a long short-term memory layer;
[0048] In an embodiment of the present invention, a method for constructing a deep learning model for bulk logistics includes the integration of a convolutional neural network layer and a long short-term memory network layer. The convolutional neural network layer is designed to extract and identify key physical space features from standard logistics data, such as the shape, length, and connection properties of road sections. The extraction of these features provides a basis for subsequent time dependency analysis. The long short-term memory network layer focuses on capturing long-term dependencies in time series data, such as changing trends and prediction patterns of traffic flow in different time periods. This combination of deep learning models enables effective prediction of future road section traffic conditions and generates more intelligent and efficient logistics route planning through path optimization algorithms.
[0049] Step S22: input the standard logistics data into the convolutional neural network layer in the bulk logistics deep learning model to perform logistics space feature recognition processing to generate road section connection data;
[0050] In an embodiment of the present invention, standard logistics data is passed into the convolutional neural network (CNN) layer of the bulk logistics deep learning model to automatically identify and process logistics spatial features. Through its specially designed convolution and pooling operations, the convolutional neural network can extract key spatial feature information from the original logistics data, such as the shape, length and connection relationship of the road segment. After processing, these logistics spatial feature data form segment connection data that describes the connection between different road segments. This process provides an efficient and automated means to generate the geographic topology of the logistics network and provides important input data for subsequent segment path optimization and traffic prediction models.
[0051] Step S23: using the long short-term memory layer in the bulk logistics deep learning model to perform learning time-dependent processing on the road section connection data to generate time-dependent pre-data;
[0052] In an embodiment of the present invention, the long short-term memory network (LSTM) layer in the bulk logistics deep learning model is used to learn and process the previously generated road segment connection data in a time-dependent manner. As a deep learning model suitable for sequence data, LSTM can effectively capture and analyze key features that change over time in the logistics network. Through this step, the model can predict the changing trends of road segment connections in future time periods, including but not limited to seasonal changes in traffic flow, periodic fluctuations in road segment usage patterns, and other time-related logistics and transportation characteristics. The generated time-dependent predicted data provides a means of accurately predicting and optimizing logistics network operations, which can be used to improve logistics efficiency, reduce transportation costs, and optimize the decision support capabilities of real-time scheduling systems.
[0053] Step S24: Perform path optimization analysis on the time-dependent prediction data through a path optimization algorithm to generate road route planning data.
[0054] In an embodiment of the present invention, a path optimization algorithm is used to deeply analyze and process the time-dependent pre-data to generate efficient road route planning data. Using the pre-generated time-dependent pre-data, these data reflect the dynamic changes and characteristics of the road section connections in different time periods. Subsequently, one or more path optimization algorithms, such as the shortest path algorithm based on graph theory or the heuristic search algorithm, are used to perform systematic optimization analysis on these data. These algorithms can quickly and effectively generate the most optimized road route planning according to pre-set optimization goals, such as minimizing travel time or optimizing distance.
[0055] The beneficial effects of the present invention are that by constructing a deep learning model for bulk logistics including a convolutional neural network layer and a long short-term memory layer, it is possible to effectively identify and extract the spatial characteristics and time dependencies of logistics, and to achieve accurate analysis and processing of standard logistics data. By using the convolutional neural network layer to perform logistics spatial feature recognition processing, accurate road section connection data can be generated, and by using the long short-term memory layer to perform learning time dependency processing on the road section connection data, it is possible to capture the temporal dynamic changes in the logistics data and generate high-precision time-dependent predicted data. Finally, by combining the path optimization algorithm to perform path optimization analysis on the time-dependent predicted data, it is possible to generate optimized road route planning data, thereby improving the efficiency and accuracy of logistics transportation, reducing transportation costs, and improving the overall logistics service quality.
[0056] Preferably, step S3 comprises the following steps:
[0057] Step S31: obtaining a dashcam video and weather data; confirming the starting point location based on the road route planning data to obtain the starting point location data; extracting key frames from the dashcam video to generate a dashcam key frame image;
[0058] Step S32: performing route fitting analysis on the road route planning data through the vehicle-mounted GPS and the departure point location data to generate an actual transportation route;
[0059] Step S33: using image recognition technology to identify vehicles in key frame images of the driving recorder to generate vehicle record data; performing congestion transportation scheduling on the vehicle record data according to a preset road congestion threshold to generate a road congestion scheduling plan;
[0060] Step S34: Perform severe transportation scheduling on the actual transportation route and the road route planning data based on weather and meteorological data to generate a road weather scheduling plan;
[0061] Step S35: Perform transportation destination difference scheduling based on the actual transportation route and road route planning data to generate a road distance transportation scheduling plan;
[0062] Step S36: Merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation merger plan; perform standard timestamp calibration on the bulk logistics transportation merger plan to generate a bulk logistics transportation prediction optimization scheduling plan.
[0063] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:
[0064] Step S31: obtaining a dashcam video and weather data; confirming the starting point location based on the road route planning data to obtain the starting point location data; extracting key frames from the dashcam video to generate a dashcam key frame image;
[0065] In an embodiment of the present invention, by combining the dashcam video and weather and meteorological data, it is possible to monitor and analyze the road conditions and weather changes during transportation in real time. This method first uses the pre-set road route planning data to accurately confirm the starting point position during transportation, ensuring the accuracy and reliability of subsequent data analysis. Then, key frame images are extracted from the dashcam video, which accurately capture the key moments and important scenes during driving. These key frame images are not just static image snapshots, they form a complete record of the driving process through the continuity of time and space. This provides an important basis for subsequent vehicle identification, traffic event analysis, and road safety assessment. By analyzing these key frame images, various types of vehicles, road signs, and traffic conditions in driving can be identified, thereby achieving comprehensive monitoring and data recording of the transportation process. This method of combining dashcam video and weather and meteorological data not only improves the ability to monitor the transportation process in real time, but also optimizes the efficiency of subsequent data processing and analysis.
[0066] Step S32: performing route fitting analysis on the road route planning data through the vehicle-mounted GPS and the departure point location data to generate an actual transportation route;
[0067] In an embodiment of the present invention, by combining the on-board GPS data and the determined departure point location data, the current position of the transport vehicle and the real-time conditions of the road on which it is located can be captured and accurately recorded in real time. This technology can not only provide accurate positioning information of the vehicle, but also realize dynamic monitoring and analysis of various road conditions during transportation through the road status data obtained in real time. The real-time position of the transport vehicle is located using the on-board GPS data, and combined with the determined departure point location data, the accuracy and reliability of subsequent route planning and data analysis are ensured. After confirming the current position of the vehicle, these data are combined with the pre-planned road route data for detailed route fitting analysis. This process aims to generate the most optimized transportation route based on the road conditions captured in real time to ensure that it meets the actual transportation needs and optimizes transportation efficiency.
[0068] Step S33: using image recognition technology to identify vehicles in key frame images of the driving recorder to generate vehicle record data; performing congestion transportation scheduling on the vehicle record data according to a preset road congestion threshold to generate a road congestion scheduling plan;
[0069] In an embodiment of the present invention, a key frame image is obtained through a driving recorder, and the image recognition technology is used to accurately identify the vehicle in the image to generate detailed vehicle record data. These data include key information such as the location, speed, and model of the vehicle. According to the preset road congestion threshold, the collected vehicle record data is analyzed and compared. Once it is detected that the road congestion reaches or exceeds the preset threshold, the congestion transportation scheduling is initiated. In the congestion transportation scheduling process, the optimal scheduling plan is formulated based on the real-time road congestion situation and the actual location of the vehicle. These plans are aimed at alleviating traffic pressure in congested areas and improving road traffic efficiency and overall transportation benefits by rationally allocating transportation resources.
[0070] Step S34: Perform severe transportation scheduling on the actual transportation route and the road route planning data based on weather and meteorological data to generate a road weather scheduling plan;
[0071] In an embodiment of the present invention, real-time weather and meteorological data are obtained, including but not limited to key parameters such as rainfall, visibility, and road surface temperature. These data can be collected through meteorological sensors, satellite meteorological data sources, or real-time updates provided by the Meteorological Bureau. The acquired weather data is compared and analyzed with the actual transportation route and road route planning data. Through comparative analysis, sections and areas that affect transportation safety and efficiency under severe weather conditions can be identified. Once the system detects that weather conditions have deteriorated or predicts that severe weather will occur in the future, the weather scheduling algorithm is started. In the weather scheduling process, the existing transportation routes are adjusted or new road weather scheduling plans are formulated according to the real-time and predicted weather conditions. These plans include measures such as changing routes, adjusting transportation time periods, and optimizing vehicle speeds, aiming to minimize the impact of weather factors on transportation safety and efficiency.
[0072] Step S35: Perform transportation destination difference scheduling based on the actual transportation route and road route planning data to generate a road distance transportation scheduling plan;
[0073] In an embodiment of the present invention, the starting and ending positions of the transport vehicles are determined by obtaining the actual transport route and the preset road route planning data. The differences between the transport destinations, that is, the distance differences between each destination, are analyzed. These differences can be calculated and evaluated based on factors such as the actual distance, traffic conditions, and road conditions. When performing difference scheduling, the optimal road distance scheduling plan will be formulated based on the difference analysis results. These plans involve adjusting the order of transport vehicles, assigning priorities, or selecting different route segments to maximize the overall transportation efficiency and save transportation costs.
[0074] Step S36: Merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation merger plan; perform standard timestamp calibration on the bulk logistics transportation merger plan to generate a bulk logistics transportation prediction optimization scheduling plan.
[0075] In the embodiment of the present invention, the road congestion scheduling scheme, the road weather scheduling scheme and the road distance transportation scheduling scheme are essentially independent scheduling schemes generated based on different data sets and models. In order to merge the schemes, it is necessary to build a multidimensional data fusion model to unify the data of each scheduling scheme. This process needs to combine the real-time road traffic data, meteorological data, geographic information data and vehicle scheduling data, and through data standardization means, make the data from different sources have consistent metrics and data formats, and ensure that each data point can be operated under a unified dimension during the merging process. In the merging process, a weighted fusion algorithm is introduced, and different scheduling schemes are weighted according to their influence on the actual transportation decision. For example, the road congestion scheme has a larger weight during peak hours, while the road weather scheduling scheme has a more critical influence on the scheduling decision under severe weather conditions. By dynamically adjusting the weights, the adaptive merging of the schemes is achieved. At the same time, dynamic programming or heuristic algorithms are used to comprehensively compare the advantages and disadvantages of different scheduling schemes, so as to generate the optimal bulk logistics transportation merging scheme. Next, the timestamp calibration processing of the merged scheme is aimed at the consistency problem of the data time series. Since different schemes are generated at different times, the timestamps of each scheme must be synchronized. The time series of each scheme can be corrected by using timing calibration methods such as linear interpolation algorithm or Kalman filtering to ensure that the time dimension of each data point remains consistent.
[0076] The beneficial effects of the present invention are that by comprehensively utilizing dashcam videos, weather and meteorological data, and vehicle-mounted GPS data, it is possible to perform multi-dimensional optimization and scheduling analysis on bulk logistics transportation. By performing key frame extraction and image recognition technology on dashcam videos, the vehicle dynamics during transportation can be accurately identified, and congestion scheduling can be performed according to a preset road congestion threshold to generate a road congestion scheduling plan. Combined with weather and meteorological data, the actual transportation route and the planned route are scheduled under severe weather conditions to generate a road weather scheduling plan. In addition, by performing a difference analysis on the actual transportation route and the planned route, a transportation destination difference scheduling is performed to generate a road distance transportation scheduling plan. By merging the road congestion scheduling plan, the road weather scheduling plan, and the road distance transportation scheduling plan, a bulk logistics transportation forecast optimization scheduling plan is generated, thereby improving the overall efficiency and reliability of the transportation process, reducing transportation delays and cost increases caused by factors such as road congestion, severe weather, and route deviations, and ensuring the efficiency and stability of logistics transportation.
[0077] Preferably, step S32 includes the following steps:
[0078] Step S321: obtaining current road information data through the vehicle-mounted GPS;
[0079] Step S322: performing path matching processing on the current road information data and the starting point location data to generate preliminary path matching data;
[0080] Step S323: Perform route fitting analysis on the preliminary path matching data and the road route planning data to generate an actual transportation route.
[0081] The present invention obtains the current road information data through the vehicle-mounted GPS, which can realize accurate monitoring and positioning of the real-time position of the transport vehicle; the current road information data and the starting point location data are matched by path matching, which can accurately determine the starting position and driving trajectory of the vehicle; the preliminary path matching data and the road route planning data are further matched by route fitting analysis, which can generate an accurate transport route that conforms to the actual situation. The comprehensive application of these technical means can effectively improve the overall efficiency and accuracy of logistics transportation, reduce the time waste and resource consumption caused by improper route selection, optimize transportation costs, and improve customer service experience.
[0082] In an embodiment of the present invention, information data of the current road is obtained through the vehicle-mounted GPS device, including the geographical location, road name and driving direction. The acquired current road information data is matched with the predetermined departure point location data, and preliminary path matching data is generated using geographic information system (GIS) technology and path matching algorithm. The preliminary path matching data is used to perform route fitting analysis with the pre-planned road route planning data, and the shortest path algorithm, dynamic time warping (DTW) algorithm, etc. are used for comparison and adjustment to generate an accurate actual transportation route.
[0083] Preferably, step S33 includes the following steps:
[0084] Step S331: Perform vehicle recognition on the key frame image of the recorder through a preset YOLO image recognition model to generate vehicle recording data;
[0085] Step S332: judging the congestion of the vehicle recorded data according to the preset road congestion threshold, and when the vehicle recorded data is greater than or equal to the preset road congestion threshold, marking the current road information data as a congested road section;
[0086] Step S333: Search for non-congested routes based on the congested road sections and road route planning data, and generate a road congestion scheduling plan.
[0087] The present invention uses a preset YOLO image recognition model to perform vehicle identification on the key frame images of the driving recorder and generate vehicle record data. The intelligence and real-time performance of bulk logistics transportation scheduling are improved. Through the image recognition model, the vehicle information in the transportation process can be accurately identified to provide a data basis for subsequent congestion judgment. The judgment mechanism based on the road congestion threshold can timely discover and mark congested sections, which helps transportation dispatchers to respond in real time and take scheduling measures. Most importantly, by searching for non-congested routes on the road and generating congestion scheduling plans, the transportation routes can be optimized, congested sections can be avoided, transportation efficiency can be improved, and transportation costs can be reduced, thereby improving the overall logistics and transportation service level and customer satisfaction.
[0088] In an embodiment of the present invention, a vehicle is identified on the key frame image of the driving recorder by a preset YOLO (You Only Look Once) image recognition model, thereby generating vehicle recording data. YOLO is an efficient real-time target detection algorithm that can quickly and accurately identify vehicle information in an image. According to a preset road congestion threshold, the generated vehicle recording data is judged to be congested. When it is detected that the vehicle recording data is greater than or equal to the preset congestion threshold, the current road information data is marked as a congested section. According to the marked congested section information and the pre-planned road route data, a search and optimization of non-congested routes on the road are performed. The system generates an optimized transportation route suitable for bypassing congested sections, i.e., a road congestion scheduling plan, through a route planning algorithm (shortest path algorithm).
[0089] Preferably, step S34 includes the following steps:
[0090] Step S341: performing severe weather forecasting on the current road information data and road route planning data based on weather and meteorological data to generate severe weather forecast data;
[0091] Step S342: When the bad weather forecast data is greater than or equal to the preset bad weather threshold, a search for good weather conditions is performed on the road route planning data and the actual transportation route to generate a new weather dispatch route;
[0092] Step S343: Perform risk assessment on the new route based on weather scheduling and generate a road weather plan.
[0093] The present invention reduces the impact of bad weather by using alternative routes under bad weather conditions, that is, new weather dispatch routes. The safety of the new route and risk factors of traffic capacity, such as road conditions and traffic conditions, are comprehensively considered to generate the optimal weather dispatch plan to ensure safety and efficiency in bad weather. By predicting bad weather in advance, the route can be adjusted in time to avoid or reduce the impact on the transportation process, ensuring the reliability and safety of logistics transportation. At the same time, through the optimization search and risk assessment of new weather dispatch routes, it is also possible to reduce transportation costs, improve customer satisfaction, and enhance corporate competitiveness.
[0094] In an embodiment of the present invention, weather and meteorological data acquired in real time are used to predict and analyze severe weather conditions of the current road information data and pre-planned road route data. This includes identifying and predicting severe weather conditions that affect transportation (such as rainstorms and snowstorms) and generating corresponding severe weather prediction data. When the generated severe weather prediction data reaches or exceeds the set threshold, the search and optimization of good weather conditions for the road route planning data and the actual transportation route is initiated. The safety, traffic capacity and risk factors of the new route, such as changes in road conditions and traffic congestion, are comprehensively considered to generate the optimal weather scheduling plan to ensure safe and effective transportation under severe weather conditions.
[0095] Preferably, step S35 includes the following steps:
[0096] Step S351: Calculating the path distance difference based on the current road information data and the road route planning data to generate path distance difference data, wherein the path distance difference data includes the first distance scheduling data and the second distance scheduling data;
[0097] Step S352: When it is confirmed that the path distance difference data is the first distance scheduling data, the route planning data is searched for the shortest route to generate a new distance scheduling route; when it is confirmed that the path distance difference data is the second distance scheduling data, no processing is performed;
[0098] Step S353: Evaluate the road distance of the new route based on the distance scheduling and generate a road distance plan.
[0099] By calculating and comparing the distance differences of different routes in real time, the system can quickly optimize the transportation routes, reduce transportation costs and time, and improve transportation efficiency and customer satisfaction. By calculating the path distance difference and selecting the new route for distance scheduling, the route planning of logistics transportation is optimized, thereby improving transportation efficiency and economic benefits. These technologies have brought important technological advances to the logistics industry and helped enterprises maintain their leading position in the fiercely competitive market.
[0100] In an embodiment of the present invention, the path distance difference is calculated using the current road information data and the pre-planned road route data. These data include the first distance scheduling data and the second distance scheduling data, which are used to evaluate the distance difference between the current route and other potential routes. Processing is performed according to the situation of the path distance difference data. If the path distance difference data is confirmed to be the first distance scheduling data, the road route planning data will be searched to find a new route with the shortest distance, that is, a distance scheduling new route. Based on the distance scheduling new route, the road distance is evaluated and analyzed to generate a final road distance solution.
[0101] Preferably, step S4 comprises the following steps:
[0102] Step S41: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation forecast optimization scheduling plan to generate logistics simulation test results;
[0103] Step S42: Use historical transportation data to accurately verify the logistics simulation test results and generate bulk logistics transportation test results;
[0104] Step S43: Compare the logistics and transportation indicators of the bulk logistics and transportation forecast optimization scheduling plan and the logistics simulation test results to generate bulk logistics and transportation indicator results;
[0105] Step S44: Conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation index results to generate a bulk logistics feasibility analysis result.
[0106] As an example of the present invention, refer to Figure 4 As shown, in this example, step S4 includes:
[0107] Step S41: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation forecast optimization scheduling plan to generate logistics simulation test results;
[0108] In an embodiment of the present invention, historical transportation data is acquired and collected through Internet of Things (IoT) devices, sensor networks, and big data technologies. These data may include detailed information such as transportation routes, transportation time, vehicle load, and transportation costs. Next, big data analysis and artificial intelligence algorithms are used to predict and optimize the scheduling of bulk logistics transportation, which involves the use of machine learning models for demand forecasting, vehicle path optimization, and scheduling algorithm optimization. Finally, the optimized scheduling plan is simulated and tested using logistics simulation software AnyLogic. The feasibility and effect of the optimization plan are evaluated through simulation path testing, and logistics simulation test results are generated for further verification and adjustment of the optimization plan.
[0109] Step S42: Use historical transportation data to accurately verify the logistics simulation test results and generate bulk logistics transportation test results;
[0110] In the embodiment of the present invention, the collected historical transportation data are stored and managed using big data technology and database management systems, and then key features and indicators related to the simulation test are extracted through data mining and statistical analysis methods. Next, the simulation test results are compared and verified using the logistics simulation software AnyLogic, and the accuracy and reliability of the simulation model are evaluated by backtesting the historical data through the construction of an accurate simulation model. Finally, the bulk logistics transportation test results are generated, and the test results are displayed and analyzed through the built-in data visualization tool of AnyLogic, thereby verifying the accuracy of the simulation test and ensuring the effectiveness and feasibility of the optimized scheduling solution.
[0111] Step S43: Compare the logistics and transportation indicators of the bulk logistics and transportation forecast optimization scheduling plan and the logistics simulation test results to generate bulk logistics and transportation indicator results;
[0112] In the embodiment of the present invention, big data analysis and data mining technology are used to process and analyze the relevant data of the bulk logistics transportation forecast optimization scheduling plan and the logistics simulation test results, and key logistics transportation indicators such as transportation time, cost, efficiency, vehicle utilization and carbon emissions are extracted. Data comparison algorithms and statistical analysis methods are applied to conduct detailed comparative analysis of these indicators, and the comparison results are presented through charts and Tableau data visualization tools. Finally, the bulk logistics transportation indicator results are generated to ensure the scientificity and effectiveness of the plan.
[0113] Step S44: Conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation index results to generate a bulk logistics feasibility analysis result.
[0114] In the embodiment of the present invention, the test results of bulk logistics transportation and the results of bulk logistics transportation indicators are integrated by data analysis and statistical methods, and the data from different sources are standardized by data fusion technology. Then, the analytic hierarchy process AHP is used to comprehensively evaluate the integrated data, considering the key indicators of transportation time, cost, efficiency, vehicle utilization and carbon emissions. The evaluation results are optimized and analyzed by simulation modeling particle swarm optimization to generate comprehensive logistics feasibility analysis results. Finally, the analysis results are displayed by Tableau data visualization tools to provide decision support.
[0115] The beneficial effects of the present invention are to generate logistics simulation test results through simulation path testing, simulate the real transportation process in a virtual environment, and thus foresee potential problems and optimization space. The simulation test results are verified using historical transportation data to ensure the accuracy and practical applicability of the simulation results, thereby generating more reliable bulk logistics transportation test results. By comparing the bulk logistics transportation prediction optimization scheduling plan and the logistics simulation test results, key logistics transportation index results can be obtained, which provide a scientific basis for evaluating transportation efficiency and optimization effects. Finally, the bulk logistics transportation test results and the logistics transportation index results are comprehensively evaluated, and the feasibility of the logistics scheduling plan can be comprehensively analyzed and confirmed, thereby generating bulk logistics feasibility analysis results. Such a process not only improves the scientificity and reliability of the plan, but also effectively reduces the risks and costs in logistics transportation, improves the overall transportation efficiency and service quality, and provides important technical support and decision-making basis for logistics companies in a highly competitive market environment.
[0116] Preferably, step S44 includes the following steps:
[0117] Step S441: extracting transportation indicators based on the bulk logistics transportation forecast optimization scheduling plan and the logistics simulation test results to obtain original logistics transportation indicators and simulated logistics transportation indicators;
[0118] Step S442: performing an index comparison analysis on the simulated logistics and transportation index and the original logistics and transportation index to generate an index analysis result;
[0119] Step S443: synthesize the indicator results based on the indicator analysis results to generate bulk logistics transportation indicator results.
[0120] The present invention provides a method for comprehensively analyzing the efficiency and economy of logistics transportation. By comparing and analyzing the original indicators with the simulation results, enterprises can more clearly understand the actual effect and potential impact of the optimization plan, so as to make more informed choices in the decision-making process. It not only helps to improve the operational efficiency of enterprises, but also can achieve continuous optimization and improvement of the logistics transportation process through data-driven decision-making.
[0121] In an embodiment of the present invention, various logistics and transportation indicators in the transportation process are extracted based on the existing bulk logistics and transportation forecast optimization and scheduling schemes and the results of the logistics simulation tests. These indicators cover data on transportation time, cost, resource utilization, etc. The simulated logistics and transportation indicators are compared and analyzed in detail with the original logistics and transportation indicators. This step aims to evaluate the performance differences between the forecast optimization and scheduling schemes and the simulation test results on different transportation indicators, including understanding whether the optimization scheme can effectively improve transportation efficiency and cost-effectiveness. The index results are synthesized and comprehensively analyzed. This includes comprehensively considering the advantages and disadvantages of each indicator to generate a comprehensive bulk logistics and transportation indicator result.
[0122] The beneficial effect of the present invention is that a comprehensive process from data acquisition, processing to final scheduling optimization is realized through multiple steps. First, by acquiring logistics and transportation information data and performing preprocessing, the accuracy and consistency of the data are ensured, and standard logistics data is generated; secondly, a deep learning model for bulk logistics is constructed, and the standard logistics data is input into the model for road path planning to generate road route planning data. Next, the dashcam video and weather and meteorological data are obtained, the starting point location is confirmed through the road route planning data, and key frame images are extracted from the dashcam video; the road route planning data is subjected to route fitting analysis through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route. Vehicle identification is performed on the key frame images of the dashcam to generate vehicle recording data, and multi-dimensional scheduling of logistics transportation is performed based on the actual transportation route, weather and meteorological data, and vehicle recording data to generate a bulk logistics transportation forecast optimization scheduling plan. Finally, historical transportation data is obtained to simulate the path test of the bulk logistics transportation forecast optimization scheduling plan and generate logistics simulation test results; the logistics simulation test results are accurately verified by simulation test using historical transportation data to generate bulk logistics transportation test results, and the logistics transportation indicators of the bulk logistics transportation forecast optimization scheduling plan and the logistics simulation test results are compared to generate bulk logistics transportation indicator results, and finally, through comprehensive feasibility evaluation, the bulk logistics feasibility analysis results are generated. This method not only improves the accuracy and real-time performance of logistics transportation path planning through comprehensive analysis of multi-source data and the application of deep learning models, but also ensures the reliability and feasibility of the scheduling plan through simulation testing and historical data verification, significantly improves the efficiency and safety of bulk logistics transportation, effectively reduces transportation costs, and enhances the ability to respond to emergencies, thereby achieving all-round optimization of logistics transportation.
[0123] 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.
[0124] 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 method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis. It is characterized in that The following steps are involved: Step S1: obtaining logistics and transportation information data; performing data preprocessing on the logistics and transportation information data to generate standard logistics data; Step S2: construct a deep learning model for bulk logistics; input standard logistics data into the deep learning model for bulk logistics to perform road path planning and generate road route planning data; Step S3: Obtaining driving recorder video and weather data; Confirm the starting point location based on the road route planning data to obtain the starting point location data; extract key frames from the dashcam video to generate the dashcam key frame image; perform route fitting analysis on the road route planning data through the vehicle-mounted GPS and the starting point location data to generate the actual transportation route; Perform vehicle recognition on the key frame images of the driving recorder to generate vehicle recording data; Carry out congestion transportation scheduling for vehicle record data according to preset road congestion thresholds and generate road congestion scheduling plans; carry out severe transportation scheduling for actual transportation routes and road route planning data based on weather and meteorological data and generate road weather scheduling plans; Carry out transportation destination difference scheduling based on actual transportation routes and road route planning data, and generate road distance transportation scheduling plan; Combine the road congestion scheduling plan, road weather scheduling plan and road distance transportation scheduling plan to generate a bulk logistics transportation forecast optimization scheduling plan; Step S4: Obtain historical transportation data; Conduct simulation path testing on the forecast optimization and dispatching scheme for bulk logistics transportation, and generate logistics simulation test results; Use historical transportation data to accurately verify the logistics simulation test results, and generate bulk logistics transportation test results; Compare the logistics transportation indicators of bulk logistics transportation forecast optimization scheduling plan and logistics simulation test results to generate bulk logistics transportation indicator results; conduct a comprehensive feasibility assessment on bulk logistics transportation test results and bulk logistics transportation indicator results to generate bulk logistics feasibility analysis results.
2. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 1, It is characterized in that Step S1 includes the following steps: Step S11: Obtaining logistics transportation information data; Step S12: performing data cleaning on the logistics transportation information data to generate logistics cleaning data; Step S13: Performing data consistency check on the logistics cleaning data to generate logistics consistent data; Step S14: Standardize the logistics consistency data to generate standard logistics data.
3. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 1, It is characterized in that Step S2 includes the following steps: Step S21: constructing a deep learning model for bulk logistics, wherein the deep learning model for bulk logistics includes a convolutional neural network layer and a long short-term memory layer; Step S22: input the standard logistics data into the convolutional neural network layer in the bulk logistics deep learning model to perform logistics space feature recognition processing to generate road section connection data; Step S23: using the long short-term memory layer in the bulk logistics deep learning model to perform learning time-dependent processing on the road section connection data to generate time-dependent pre-data; Step S24: Perform path optimization analysis on the time-dependent prediction data through a path optimization algorithm to generate road route planning data.
4. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 1, It is characterized in that Step S3 includes the following steps: Step S31: obtaining a dashcam video and weather data; confirming the starting point location based on the road route planning data to obtain the starting point location data; extracting key frames from the dashcam video to generate a dashcam key frame image; Step S32: performing route fitting analysis on the road route planning data through the vehicle-mounted GPS and the departure point location data to generate an actual transportation route; Step S33: using image recognition technology to identify vehicles in key frame images of the driving recorder to generate vehicle record data; performing congestion transportation scheduling on the vehicle record data according to a preset road congestion threshold to generate a road congestion scheduling plan; Step S34: Perform severe transportation scheduling on the actual transportation route and the road route planning data based on weather and meteorological data to generate a road weather scheduling plan; Step S35: Perform transportation destination difference scheduling based on the actual transportation route and road route planning data to generate a road distance transportation scheduling plan; Step S36: Merge the road congestion scheduling plan, the road weather scheduling plan and the road distance transportation scheduling plan to generate a bulk logistics transportation merger plan; perform standard timestamp calibration on the bulk logistics transportation merger plan to generate a bulk logistics transportation prediction optimization scheduling plan.
5. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 4, It is characterized in that Step S32 includes the following steps: Step S321: obtaining current road information data through the vehicle-mounted GPS; Step S322: performing path matching processing on the current road information data and the starting point location data to generate preliminary path matching data; Step S323: Perform route fitting analysis on the preliminary path matching data and the road route planning data to generate an actual transportation route.
6. A method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 4, It is characterized in that Step S33 includes the following steps: Step S331: Perform vehicle recognition on the key frame image of the recorder through a preset YOLO image recognition model to generate vehicle recording data; Step S332: performing congestion judgment on the vehicle recorded data according to a preset road congestion threshold, and when the vehicle recorded data is greater than or equal to the preset road congestion threshold, marking the current road information data as a congested road section; Step S333: Search for non-congested routes based on the congested road sections and road route planning data, and generate a road congestion scheduling plan.
7. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 4, It is characterized in that Step S34 includes the following steps: Step S341: performing severe weather forecasting on the current road information data and road route planning data based on weather and meteorological data to generate severe weather forecast data; Step S342: When the bad weather forecast data is greater than or equal to the preset bad weather threshold, a search for good weather conditions is performed on the road route planning data and the actual transportation route to generate a new weather dispatch route; Step S343: Perform risk assessment on the new route based on weather scheduling and generate a road weather plan.
8. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 4, It is characterized in that Step S35 includes the following steps: Step S351: Calculating the path distance difference based on the current road information data and the road route planning data to generate path distance difference data, wherein the path distance difference data includes the first distance scheduling data and the second distance scheduling data; Step S352: When it is confirmed that the path distance difference data is the first distance scheduling data, the route planning data is searched for the shortest route to generate a new distance scheduling route; when it is confirmed that the path distance difference data is the second distance scheduling data, no processing is performed; Step S353: Evaluate the road distance of the new route based on the distance scheduling and generate a road distance plan.
9. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 1, It is characterized in that Step S4 includes the following steps: Step S41: Acquire historical transportation data; perform simulation path testing on the bulk logistics transportation forecast optimization scheduling plan to generate logistics simulation test results; Step S42: Use historical transportation data to accurately verify the logistics simulation test results and generate bulk logistics transportation test results; Step S43: Compare the logistics and transportation indicators of the bulk logistics and transportation forecast optimization scheduling plan and the logistics simulation test results to generate bulk logistics and transportation indicator results; Step S44: Conduct a comprehensive feasibility assessment on the bulk logistics transportation test results and the bulk logistics transportation index results to generate a bulk logistics feasibility analysis result.
10. The method for optimizing bulk logistics transportation forecasting and scheduling based on data analysis as claimed in claim 9, It is characterized in that Step S43 includes the following steps: Step S431: extracting transportation indicators based on the bulk logistics transportation forecast optimization scheduling plan and the logistics simulation test results to obtain original logistics transportation indicators and simulated logistics transportation indicators; Step S432: performing an index comparison analysis on the simulated logistics and transportation index and the original logistics and transportation index to generate an index analysis result; Step S433: synthesize the index results based on the index analysis results to generate bulk logistics transportation index results.
Citation Information
Cited By
Logistics data analysis processing method and system
CN119204602A
A Logistics Data Analysis and Processing Method and System
CN119204602B
