Dynamic evaluation system and method for collaborative development of digital logistics and rural economy

Abstract

The invention relates to the field of digital logistics and rural economy, and provides a dynamic evaluation system and method for collaborative development of digital logistics and rural economy. The system integrates big data, the Internet of Things, artificial intelligence (AI) and a geographic information system (GIS), and aims to realize comprehensive evaluation and optimization of a rural area logistics system and promote efficient fusion and collaborative development of digital logistics and rural economy. The system completes an evaluation task through the following steps: firstly, collecting and sorting basic data related to rural digital logistics, and establishing a data set; secondly, designing and constructing an evaluation model, and performing model training and optimization based on big data and an AI technology; then, deploying the optimized evaluation model to a cloud platform, and carrying out real-time monitoring and evaluation; and finally, the evaluation result is visually displayed, and the logistics system is dynamically optimized.

Description

Technical Field

[0001] This invention relates to the fields of digital logistics and the rural economy, and more particularly to a dynamic assessment system and method for the coordinated development of digital logistics and the rural economy. Based on technologies such as big data, the Internet of Things, and artificial intelligence, this system aims to comprehensively evaluate and optimize logistics systems in rural economic development, thereby promoting the efficient integration and coordinated development of the rural economy and digital logistics. Background Art

[0002] With the rapid development of information technology, digital logistics has demonstrated tremendous potential for improving logistics efficiency, reducing costs, and promoting economic growth. In rural economic development, in particular, optimizing logistics systems has become a key factor in increasing agricultural productivity, ensuring the efficient circulation of agricultural products, and promoting rural modernization. Through real-time data collection and analysis, digital logistics can precisely dispatch transportation resources, optimize route planning, and increase the transparency and controllability of the logistics process, thereby promoting rural economic development.

[0003] However, despite the increasing application of digital logistics in rural areas, the following problems still exist: First, logistics facilities and resources are unevenly distributed, and transportation infrastructure in rural areas lags behind, resulting in high logistics costs and low distribution efficiency; second, traditional logistics evaluation methods rely heavily on manual experience and cannot achieve real-time, dynamic and accurate evaluation; third, the degree of coordination between rural economic development and digital logistics is low, and the coordinated development of logistics with agriculture, industry and other fields has not yet fully realized its potential.

[0004] Existing logistics evaluation systems mostly focus on urban or regional applications, lacking adaptability and specificity tailored to the specific characteristics of the rural economy. To promote the coordinated development of the rural economy and digital logistics, a system that can dynamically evaluate, monitor, and optimize logistics processes and agricultural production links is urgently needed.

[0005] Based on this, the development of an efficient and accurate dynamic assessment system for the coordinated development of digital logistics and the rural economy is crucial. This system can monitor the operational status of the logistics system within the rural economy in real time, integrate the actual needs and development trends of rural areas, and provide more intelligent support for agricultural production. This system is of great significance for promoting rural economic development and digital transformation. This system not only improves logistics efficiency but also provides scientific decision-making support for policymakers, helping to promote the comprehensive modernization of rural areas. Summary of the Invention

[0006] This paper proposes a dynamic assessment system and method for the coordinated development of digital logistics and the rural economy. This system aims to promote the coordinated development of digital logistics and the rural economy through real-time monitoring, evaluation, and optimization of digital logistics systems in rural areas through intelligent means. By integrating big data analysis, Internet of Things sensors, artificial intelligence algorithms, and geographic information systems (GIS), the system enables comprehensive assessments of logistics processes, agricultural production, market demand, and other aspects. This helps decision-makers better understand the interrelationship between the logistics system and the rural economy, providing a sound basis for decision-making.

[0007] The present invention provides a dynamic evaluation system and method for the coordinated development of digital logistics and rural economy, wherein the method comprises the following steps:

[0008] Step (1) Collect and organize basic data of digital logistics systems in rural areas and construct an evaluation dataset;

[0009] Step (2) Design and establish an evaluation model for the coordinated development of digital logistics and rural economy;

[0010] Step (3) training and optimizing the evaluation model based on big data and artificial intelligence technology;

[0011] Step (4) deploying the optimized evaluation model to the cloud platform to monitor and evaluate the rural digital logistics system in real time;

[0012] Step (5) visualizes the evaluation results and dynamically optimizes the logistics system based on the evaluation feedback.

[0013] Dataset creation involves collecting and organizing multi-dimensional data related to digital logistics in rural areas, including but not limited to agricultural product circulation, logistics distribution routes, market demand, transportation networks, and climate conditions. Data collection utilizes IoT sensors and GPS devices to ensure the timeliness and accuracy of collected transportation data. During data preprocessing, incomplete or abnormal data due to sensor failure or human interference is removed to ensure data accuracy and reliability. To improve the stability of model training and the accuracy of evaluation results, logistics data from different time periods, geographic locations, and weather conditions are selected as the basis for model training, ensuring the broad adaptability of the evaluation results.

[0014] Furthermore, dataset production requires the use of data augmentation techniques to expand the dataset and ensure its diversity and robustness. In addition to conventional data augmentation methods such as data translation, rotation, and scaling, data synthesis is performed by simulating special environmental factors such as weather changes and traffic congestion to simulate the various changes in logistics systems in real-world scenarios. Through these data augmentation processes, the original data is randomly divided into training, validation, and test sets in an 8:1:1 ratio. The training and validation sets are used for model training and evaluation, while the test set is used for final evaluation and validation.

[0015] Furthermore, after pre-processing, the collected data will be processed through standardization or normalization methods to ensure the consistency and comparability of the data. The standardization formula is set as follows:

[0016]

[0017] Where X is the raw data, μ is the mean, σ is the standard deviation, and Xnorm is the normalized data. After data processing, a comprehensive evaluation dataset is constructed, including various key indicators, for subsequent evaluation model training and optimization.

[0018] The evaluation model involves: first, designing an evaluation framework based on big data analysis that comprehensively considers multiple factors, including logistics efficiency, agricultural product supply chain management, market demand, and transportation costs in rural areas. By performing statistical analysis and pattern recognition on this data, a dynamic evaluation model for the coordinated development of digital logistics and the rural economy is constructed. This model dynamically adjusts and provides real-time feedback on the operating status of the rural logistics system. Second, the evaluation model is optimized using artificial intelligence algorithms, employing deep learning and machine learning methods to generate accurate evaluation results through big data analysis. By modeling the logistics demand and costs of different geographical regions and agricultural products, logistics routes and transportation modes are dynamically optimized to improve overall logistics efficiency and the synergy of the rural economy. Finally, the optimized evaluation model is deployed on a cloud platform, and dynamic optimization of the logistics system is implemented through real-time monitoring and data feedback. The cloud platform's visualization capabilities provide a real-time display of the logistics system's operating status, helping decision-makers promptly identify logistics bottlenecks and optimization opportunities. Through this feedback mechanism, logistics strategies can be continuously adjusted to ensure the coordinated development of the digital logistics system and the rural economy.

[0019] Furthermore, first, a basic linear regression model is constructed to evaluate the relationship between logistics efficiency and agricultural production efficiency, and the objective function is set as follows:

[0020] E logistics =α·T logistics +β·C logistics +γ·Dlogistics

[0021] Among them, E logistics is the logistics efficiency, T logistics is the transportation time, C logistics is the logistics cost, D logistics is the impact of delivery delay, and α, β, and γ are unknown parameters.

[0022] Next, machine learning algorithms (decision trees, random forests) are used to comprehensively evaluate these multiple factors to form a nonlinear collaborative evaluation model, in which the relationship between each factor is optimized through training data, and ultimately a comprehensive score is generated that can dynamically evaluate the level of coordinated development of digital logistics and the rural economy.

[0023] Furthermore, big data and artificial intelligence technologies are used to train and optimize the evaluation model. The model is trained using collected historical data and its parameters are adjusted to accurately reflect the dynamic changes in the coordinated development of digital logistics and the rural economy.

[0024] Furthermore, first, 80% of the data in the dataset is used for model training, and 20% is used for model validation. During the model training process, deep learning algorithms such as deep neural networks (DNNs) and long short-term memory networks (LSTMs) can be used to capture complex patterns in the data. The objective function is set as:

[0025]

[0026] in, is the predicted value, y i is the actual value, λ is the regularization parameter, θ j are model parameters, and L(θ) is the loss function.

[0027] By using backpropagation and gradient descent algorithms, the model parameters are optimized to minimize the loss function. The optimized model can make more accurate predictions about future data.

[0028] Furthermore, in order to improve the prediction accuracy of the model, the cross-validation method is used to verify the model. The evaluation formula of the validation set is set as follows:

[0029]

[0030] Among them, MSE is the mean square error, is the model prediction value, y i The model parameters are adjusted through multiple rounds of cross-validation to ensure the generalization ability of the model.

[0031] After model training and optimization are complete, the optimized evaluation model is deployed to the cloud platform, enabling real-time monitoring and evaluation of the operational status of the rural digital logistics system. The cloud platform connects to the rural logistics system through an interface, acquiring real-time information such as logistics routes, transportation status, and market demand. This data is then fed into the optimized evaluation model for analysis.

[0032] According to a second aspect of the present invention, a dynamic evaluation system and method for the coordinated development of digital logistics and rural economy are provided, the system comprising:

[0033] A data collection module, which uses IoT sensors and GPS devices to collect multi-dimensional data related to digital logistics in rural areas, including but not limited to agricultural product circulation, logistics distribution routes, market demand, transportation networks, climate conditions, etc.;

[0034] The data preprocessing module is used to preprocess the collected raw data, remove abnormal data, and perform standardization or normalization to ensure the consistency and comparability of the data;

[0035] The dataset creation module is used to construct a comprehensive evaluation dataset from preprocessed data, including various key indicators, for subsequent training and optimization of the evaluation model;

[0036] The evaluation model module is used to design and establish a dynamic evaluation model for the coordinated development of digital logistics and the rural economy based on big data analysis and artificial intelligence algorithms. The evaluation model can reflect the operating status of the logistics system in rural areas in real time and conduct a comprehensive assessment based on factors such as agricultural production and market demand;

[0037] The cloud platform deployment module is used to deploy the optimized evaluation model to the cloud platform, monitor and evaluate the rural digital logistics system in real time, and support dynamic optimization of the logistics system;

[0038] The visualization display module is used to display the evaluation results through the cloud platform, helping decision makers to obtain logistics bottlenecks and optimization space in real time, and providing a basis for scientific decision-making.

[0039] The data set production module includes a data enhancement module, which expands the data set by simulating special environmental factors, thereby enhancing the diversity of the data and the robustness of the model. Specifically, the data enhancement module generates diverse simulation data by introducing factors such as different weather changes (such as rainfall, snowstorms, haze, etc.) and traffic congestion. For example, when simulating weather changes, the module considers the impact of factors such as temperature, humidity, precipitation and wind speed on logistics routes, agricultural product transportation and production efficiency; in traffic congestion simulation, it sets different degrees of traffic congestion, road closures or construction conditions to reflect the performance of the logistics system under complex traffic conditions. These enhanced data not only enrich the training samples of the model, but also make the evaluation model more adaptable and stable when dealing with various changes in the real environment, further improving the prediction accuracy and decision support capabilities of the system.

[0040] Furthermore, the evaluation model module includes an evaluation framework based on big data analysis. This framework comprehensively considers multiple factors, such as logistics efficiency, agricultural product supply chain management, and transportation costs, to comprehensively assess the level of coordinated development between digital logistics and the rural economy. Specifically, the evaluation framework establishes a dynamic, real-time updated evaluation model by conducting statistical analysis and pattern recognition on a large amount of collected multi-dimensional data. In terms of logistics efficiency, the framework considers key indicators such as transportation timeliness, delivery accuracy, and resource utilization. In terms of agricultural product supply chain management, the framework incorporates the production, storage, and transportation of agricultural products into the evaluation, evaluating the coordination and efficiency of each link. Transportation costs combine factors such as energy consumption, road usage fees, and personnel costs during the transportation process to help evaluate the cost-effectiveness of the overall logistics system. Through the analysis of these comprehensive factors, the evaluation framework can dynamically adjust the evaluation model, reflect the operating status of the logistics system in real time, and provide decision makers with accurate data support and optimization suggestions to ensure that logistics and agricultural development complement each other and advance in a coordinated manner.

[0041] Furthermore, the evaluation model module also includes an artificial intelligence algorithm module, which utilizes advanced machine learning and deep learning methods to train and optimize the evaluation model, thereby continuously improving evaluation accuracy and the system's intelligence level. This module first applies machine learning techniques to a large amount of historical and real-time data input to identify and mine underlying patterns and patterns that influence logistics efficiency, agricultural product supply chains, and transportation costs. By learning from these data patterns, the module is able to build accurate predictive models, helping the evaluation system better understand the interrelationships between different factors and make real-time decisions. Furthermore, the artificial intelligence algorithm module is capable of dynamic adjustments based on the evaluation results. For example, in logistics route planning, the module can optimize transportation routes and modes in real time based on real-time traffic flow, climate conditions, and market demand to avoid congestion, improve delivery timeliness, and reduce energy consumption. These optimization strategies not only improve the overall operational efficiency of the logistics system but also ensure that logistics distribution remains efficient under various environmental conditions.

[0042] Furthermore, the evaluation model module further includes a linear regression model for evaluating the relationship between logistics efficiency and agricultural production efficiency, and the objective function includes a weighted combination of factors such as logistics efficiency, transportation time, logistics cost, and distribution delay.

[0043] The evaluation model module further includes a nonlinear collaborative evaluation model, which optimizes the relationship between different factors through machine learning algorithms (such as decision trees, random forests, etc.), generates a comprehensive score, and dynamically evaluates the level of coordinated development of the logistics system and the rural economy.

[0044] Furthermore, the cloud platform deployment module interfaces with rural logistics systems to acquire a variety of key data in real time, such as logistics routes, transportation status, and market demand. This data is rapidly and reliably acquired from various logistics nodes, sensors, and monitoring systems via efficient data transmission protocols, and aggregated to the cloud platform for centralized processing and analysis. Logistics route data covers the entire process from warehousing to distribution, including transportation routes, vehicle conditions, and cargo location. Transportation status provides real-time details such as vehicle operating status, cargo delivery progress, and transportation delays. Market demand information reflects fluctuations in agricultural product demand by analyzing factors such as consumer orders, inventory levels, and climate change. This real-time data is rapidly transmitted to an optimized evaluation model for analysis, which automatically and dynamically adjusts and optimizes based on current logistics conditions and market demand. For example, the evaluation model adjusts the optimal transportation route based on real-time traffic conditions, optimizes the allocation of transportation resources, and predicts future transportation demand. Furthermore, the system can proactively optimize logistics routes and transportation modes based on special factors such as weather conditions and holidays, thereby mitigating potential logistics bottlenecks and delays.

[0045] The beneficial effects of the present invention are:

[0046] 1. Efficiently Optimize Logistics Systems: By integrating advanced technologies such as big data, the Internet of Things, and artificial intelligence, this invention monitors rural digital logistics systems in real time and accurately optimizes routes, dispatches resources, and adjusts transportation modes based on dynamic changes, effectively improving logistics efficiency and reducing costs. This process not only promotes the efficient operation of the rural economy but also provides real-time decision support at critical stages.

[0047] 2. Promote the coordinated development of the rural economy: This invention systematically evaluates multiple factors, including logistics efficiency, agricultural product supply chain management, market demand, and transportation costs, to form a comprehensive and dynamic evaluation model. This model provides a scientific basis for the modernization and digital transformation of the rural economy. By synergizing with agricultural production, market demand, and other aspects, it promotes the deep integration of the rural economy and digital logistics, enhancing agricultural productivity and the overall competitiveness of rural areas.

[0048] 3. Enhanced Data Diversity and Model Robustness: Through data augmentation techniques (such as simulating weather changes, traffic congestion, and other special environmental factors), this invention generates diverse training datasets, improving the robustness and adaptability of the evaluation model in complex environments. This enhanced data helps the model cope with variations in climate and traffic conditions, thereby providing more accurate predictions and optimization solutions, ensuring the stable operation of the logistics system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is an overall flow chart of a dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to the present invention;

[0050] Figure 2 It is the overall system architecture diagram of the present invention;

[0051] Figure 3 It is a data acquisition and preprocessing flow chart of the present invention;

[0052] Figure 4 It is a model training flow chart of the present invention;

[0053] Figure 5 It is a training process optimization curve diagram of the present invention;

[0054] Figure 6 It is a cross-validation schematic diagram of the present invention;

[0055] Figure 7 It is a diagram of a dynamic evaluation and optimization example of the present invention;

[0056] Figure 8It is a schematic diagram of the system hardware and data interaction of the present invention. DETAILED DESCRIPTION

[0057] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. The examples provided are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments without creative effort are within the scope of protection of the present invention.

[0058] See also Figure 1 The embodiment of the present invention provides a dynamic evaluation system and method for the coordinated development of digital logistics and rural economy, which specifically includes the following steps:

[0059] Step (1) Collect and organize basic data of digital logistics systems in rural areas and construct an evaluation dataset;

[0060] Step (2) Design and establish an evaluation model for the coordinated development of digital logistics and rural economy;

[0061] Step (3) training and optimizing the evaluation model based on big data and artificial intelligence technology;

[0062] Step (4) deploying the optimized evaluation model to the cloud platform to monitor and evaluate the rural digital logistics system in real time;

[0063] Step (5) visualizes the evaluation results and dynamically optimizes the logistics system based on the evaluation feedback.

[0064] Dataset creation involves collecting and organizing multi-dimensional data related to digital logistics in rural areas, including but not limited to agricultural product circulation, logistics distribution routes, market demand, transportation networks, and climate conditions. Data collection utilizes IoT sensors and GPS devices to ensure the timeliness and accuracy of collected transportation data. During data preprocessing, incomplete or abnormal data due to sensor failure or human interference is removed to ensure data accuracy and reliability. To improve the stability of model training and the accuracy of evaluation results, logistics data from different time periods, geographic locations, and weather conditions are selected as the basis for model training, ensuring the broad adaptability of the evaluation results.

[0065] Furthermore, dataset production requires the use of data augmentation techniques to expand the dataset and ensure its diversity and robustness. In addition to conventional data augmentation methods such as data translation, rotation, and scaling, data synthesis is performed by simulating special environmental factors such as weather changes and traffic congestion to simulate the various changes in logistics systems in real-world scenarios. Through these data augmentation processes, the original data is randomly divided into training, validation, and test sets in an 8:1:1 ratio. The training and validation sets are used for model training and evaluation, while the test set is used for final evaluation and validation.

[0066] Furthermore, after pre-processing, the collected data will be processed through standardization or normalization methods to ensure the consistency and comparability of the data. The standardization formula is set as follows:

[0067]

[0068] Where X is the raw data, μ is the mean, σ is the standard deviation, and Xnorm is the normalized data. After data processing, a comprehensive evaluation dataset is constructed, including various key indicators, for subsequent evaluation model training and optimization.

[0069] The evaluation model involves: first, designing an evaluation framework based on big data analysis that comprehensively considers multiple factors, including logistics efficiency, agricultural product supply chain management, market demand, and transportation costs in rural areas. By performing statistical analysis and pattern recognition on this data, a dynamic evaluation model for the coordinated development of digital logistics and the rural economy is constructed. This model dynamically adjusts and provides real-time feedback on the operating status of the rural logistics system. Second, the evaluation model is optimized using artificial intelligence algorithms, employing deep learning and machine learning methods to generate accurate evaluation results through big data analysis. By modeling the logistics demand and costs of different geographical regions and agricultural products, logistics routes and transportation modes are dynamically optimized to improve overall logistics efficiency and the synergy of the rural economy. Finally, the optimized evaluation model is deployed on a cloud platform, and dynamic optimization of the logistics system is implemented through real-time monitoring and data feedback. The cloud platform's visualization capabilities provide a real-time display of the logistics system's operating status, helping decision-makers promptly identify logistics bottlenecks and optimization opportunities. Through this feedback mechanism, logistics strategies can be continuously adjusted to ensure the coordinated development of the digital logistics system and the rural economy.

[0070] Furthermore, first, a basic linear regression model is constructed to evaluate the relationship between logistics efficiency and agricultural production efficiency, and the objective function is set as follows:

[0071] E logistics =α·T logistics +β·C logistics +γ·Dlogitics

[0072] Among them, E logistics is the logistics efficiency, T logistics is the transportation time, C logistics is the logistics cost, D logistics is the impact of delivery delay, and α, β, and γ are unknown parameters.

[0073] Next, machine learning algorithms (decision trees, random forests) are used to comprehensively evaluate these multiple factors to form a nonlinear collaborative evaluation model, in which the relationship between each factor is optimized through training data, and ultimately a comprehensive score is generated that can dynamically evaluate the level of coordinated development of digital logistics and the rural economy.

[0074] Furthermore, big data and artificial intelligence technologies are used to train and optimize the evaluation model. The model is trained using collected historical data and its parameters are adjusted to accurately reflect the dynamic changes in the coordinated development of digital logistics and the rural economy.

[0075] Furthermore, first, 80% of the data in the dataset is used for model training, and 20% is used for model validation. During the model training process, deep learning algorithms such as deep neural networks (DNNs) and long short-term memory networks (LSTMs) can be used to capture complex patterns in the data. The objective function is set as:

[0076]

[0077] in, is the predicted value, y i is the actual value, λ is the regularization parameter, θ j are model parameters, and L(θ) is the loss function.

[0078] By using backpropagation and gradient descent algorithms, the model parameters are optimized to minimize the loss function. The optimized model can make more accurate predictions about future data.

[0079] Furthermore, in order to improve the prediction accuracy of the model, the cross-validation method is used to verify the model. The evaluation formula of the validation set is set as follows:

[0080]

[0081] Among them, MSE is the mean square error, is the model prediction value, y i The model parameters are adjusted through multiple rounds of cross-validation to ensure the generalization ability of the model.

[0082] After model training and optimization are complete, the optimized evaluation model is deployed to the cloud platform, enabling real-time monitoring and evaluation of the operational status of the rural digital logistics system. The cloud platform connects to the rural logistics system through an interface, acquiring real-time information such as logistics routes, transportation status, and market demand. This data is then fed into the optimized evaluation model for analysis.

[0083] Specific embodiment 2, see Figure 2 and Figure 8 The system architecture shown in the figure adopts a three-level hardware deployment scheme: STM32F407 core sensor nodes are arranged in the agricultural product production area. Each node integrates an SHT30 temperature and humidity sensor (measuring range -40 to 125 ° C, accuracy ± 0.3 ° C) and a UBLOX NEO-6M GPS module (positioning accuracy 2.5m CEP). The node spacing is calculated according to the formula ), where S is the production area and n ≥ 5 is the minimum number of nodes. The edge layer uses Jetson Nano as the compute node, equipped with a quad-core ARM Cortex-A57 processor and a 128-core Maxwell GPU, and pre-installed with the TensorFlow Lite 2.8 runtime environment. The cloud is deployed on an Alibaba Cloud ECS instance (ecs.g6ne.4xlarge) running Ubuntu 20.04 LTS and a Docker Swarm cluster.

[0084] See also Figure 3 The data processing flow shown above first collects logistics route data (JSON format, including latitude and longitude coordinates and timestamps), climate data (connected to the China Meteorological Administration's CMACast system via WebSocket), and market data (connected to the Ministry of Agriculture and Rural Affairs' wholesale market database). The collected data is processed as follows:

[0085] 1) Use sliding window filtering (window width 10, Hamming window function) to eliminate noise;

[0086] 2) Implemented data augmentation, including traffic jam simulation (randomly reducing the speed values ​​of 30% of the data points) and weather perturbations (rainfall was multiplied by a random coefficient of 1.2-1.5);

[0087] 3) Standardization processing was performed. The numerical features were normalized using z-score, and the geographic coordinates were converted to UTM projection (zone number 49N).

[0088] See also Figure 4-6A hybrid LSTM-DNN model was constructed. This model uses a bidirectional LSTM layer to capture temporal features, introduces an attention mechanism to achieve feature reweighting, and constructs a deep neural network for multi-dimensional feature fusion. Specifically, the model consists of an input layer processing a 64-dimensional feature vector with a time step of 24. A bidirectional LSTM layer is configured with 128 units and a dropout setting of 0.3 to prevent overfitting. An attention mechanism layer calculates feature importance weights, and a fully connected layer contains 64 neurons. Training utilizes a dynamic learning rate scheduling strategy, with an initial learning rate of 0.001 and a 10% decay every 1000 steps. The Huber loss function is used to enhance training stability, with the delta parameter set to 1.0 to balance mean squared error (MSE) and absolute error (AER). Training is terminated with an early stopping mechanism, terminating when the validation set MSE improves by less than 1e-4 for 15 consecutive epochs. Five-fold cross-validation is used to ensure model stability. Experimental data shows that after 150 epochs, the validation set MSE steadily decreases from an initial 0.58 to 0.12, demonstrating good convergence of the model.

[0089] See also Figure 7 , an evaluation interface was established to display optimization results in real time. Based on a multi-objective planning model, this interface dynamically generates the optimal logistics route by comprehensively considering factors such as transportation time, cost, and energy consumption. In actual application, the system performed exceptionally well in a provincial specialty agricultural product logistics scenario, reducing average transportation time from 4.2 hours to 3.3 hours, an improvement of 21.4%, lowering the cold chain loss rate from 12% to 8.5%, an improvement of 29.2%, and reducing fuel costs per unit distance from 3.2 yuan / km to 2.7 yuan / km, a savings of 15.6%. These performance improvements are primarily due to the system's dynamic evaluation mechanism and adaptive optimization algorithm. In terms of communication architecture, the system adopts a hierarchical design. The perception layer uses the LoRaWAN protocol to achieve low-power wide-area coverage. The edge layer and the cloud platform maintain a reliable connection through the MQTT 3.1.1 protocol. A 60-second heartbeat packet is set to maintain the session. The data frame adopts a lightweight encapsulation format, including a 2-byte frame header, a 4-byte device identifier, a 12-byte payload and a 2-byte checksum to ensure transmission efficiency and reliability. This design not only meets the real-time requirements, but also adapts to the network environment characteristics of rural areas.

[0090] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structures or equivalent process changes made using the description and drawings of the present invention are also included in the patent protection scope of the present invention.

Claims

1. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy, characterized by: The system comprises: A data collection module, which uses IoT sensors and GPS devices to collect multi-dimensional data related to digital logistics in rural areas, including but not limited to agricultural product circulation, logistics distribution routes, market demand, transportation networks, climate conditions, etc.; The data preprocessing module is used to preprocess the collected raw data, remove abnormal data, and perform standardization or normalization to ensure the consistency and comparability of the data; The dataset creation module is used to construct a comprehensive evaluation dataset from preprocessed data, including various key indicators, for subsequent training and optimization of the evaluation model; The evaluation model module is used to design and establish a dynamic evaluation model for the coordinated development of digital logistics and the rural economy based on big data analysis and artificial intelligence algorithms. The evaluation model can reflect the operating status of the logistics system in rural areas in real time and conduct a comprehensive assessment based on factors such as agricultural production and market demand; The cloud platform deployment module is used to deploy the optimized evaluation model to the cloud platform, monitor and evaluate the rural digital logistics system in real time, and support dynamic optimization of the logistics system; The visualization display module is used to display the evaluation results through the cloud platform, helping decision makers to obtain logistics bottlenecks and optimization space in real time, and providing a basis for scientific decision-making.

2. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 1, characterized in that: The dataset production module includes a data enhancement module for expanding the dataset by simulating special environmental factors such as weather changes and traffic congestion to enhance the diversity and robustness of the data.

3. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 1, characterized in that: The evaluation model module includes: The evaluation framework based on big data analysis can comprehensively consider multiple factors such as logistics efficiency, agricultural product supply chain management, and transportation costs, and construct a dynamic evaluation model through statistical analysis and pattern recognition; The artificial intelligence algorithm module is used to train and optimize the evaluation model using machine learning and deep learning methods to improve evaluation accuracy and dynamically adjust logistics routes and transportation modes.

4. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 1, characterized in that: The evaluation model module further includes: A linear regression model is used to evaluate the relationship between logistics efficiency and agricultural production efficiency, where the objective function includes a weighted combination of factors such as logistics efficiency, transportation time, logistics cost, and delivery delay; A nonlinear collaborative evaluation model optimizes the relationship between different factors through machine learning algorithms (such as decision trees, random forests, etc.), generates a comprehensive score, and dynamically evaluates the level of collaborative development between the logistics system and the rural economy.

5. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 1, characterized in that: The cloud platform deployment module connects with the logistics system in rural areas through an interface to obtain real-time information such as logistics routes, transportation status, market demand, etc., and inputs these data into the optimized evaluation model for analysis.

6. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy, characterized by: The method comprises: Step (1) collect and organize basic data of digital logistics systems in rural areas and construct an evaluation dataset; Step (2) Design and establish an evaluation model for the coordinated development of digital logistics and rural economy; Step (3) training and optimizing the evaluation model based on big data and artificial intelligence technology; Step (4) deploying the optimized evaluation model to the cloud platform to monitor and evaluate the rural digital logistics system in real time; Step (5) displays the evaluation results through the cloud platform and dynamically optimizes the logistics system based on the feedback.

7. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 6, characterized in that: The dataset preparation steps include collecting real-time data related to the logistics system through IoT sensors and GPS devices, and preprocessing and standardizing the data.

8. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 6, characterized in that: The evaluation model training step includes using deep learning and machine learning algorithms to train the evaluation model to optimize logistics routes and transportation modes.

9. A dynamic evaluation system and method for the coordinated development of digital logistics and rural economy according to claim 6, characterized in that: The optimization steps include adjusting logistics strategies based on the evaluation results, improving logistics efficiency, and enhancing the level of coordinated development of the rural economy and digital logistics.

Images