Intelligent logistics supply chain digital management system based on data analysis

By applying a variety of data analysis technologies in the logistics supply chain management system, the problem of inaccurate demand forecasting and inventory management in traditional systems is solved, and more efficient supply chain collaborative management is achieved, and overall efficiency and responsiveness are improved.

CN119005838AActive Publication Date: 2024-11-22SHENZHEN XINGCHENG TECH CO LTD

Patent Information

Application Number
CN202411463933.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-21
Publication Date
2024-11-22
Estimated Expiration
2044-10-21

AI Technical Summary

Technical Problem

Traditional logistics supply chain management systems rely on manual experience and cannot accurately analyze market data and historical order data, resulting in inaccurate demand forecasts and difficulty in dealing with demand fluctuations, resulting in low inventory management efficiency and supplier choices not to optimize, which affects overall efficiency.

Method used

Adopt a digital management system for intelligent logistics supply chain based on data analysis, combining recurrent neural networks, long and short-term memory networks, Bayesian dynamic linear models, convolutional neural networks, multi-layer perceptrons, fuzzy clustering algorithms, genetic algorithms and adaptive Large-scale neighborhood search algorithm and other technologies realize demand forecasting, inventory optimization, supplier management and logistics path planning, and form a system integration and feedback mechanism.

Benefits of technology

It improves the accuracy of demand forecasts, flexibility of inventory management, optimization of supplier selection and efficient logistics paths, ensuring that the supply chain can respond quickly when demand fluctuates, reduce inventory costs, and improve overall efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005838A_ABST
    Figure CN119005838A_ABST
Patent Text Reader

Abstract

The invention relates to an intelligent logistics supply chain digital management system based on data analysis. The system comprises a demand prediction and analysis unit, an inventory optimization unit, a supplier management unit, a logistics path optimization unit, a system integration and feedback unit and a real-time logistics marketing unit. Historical orders and market data are analyzed through a recurrent neural network and a long-short-term memory network, and accurate prediction of short-term and medium-long term demands is realized. And the inventory optimization unit is combined with a Bayesian dynamic linear model and a convolutional neural network to realize dynamic management of the inventory. The supplier management unit optimizes supplier selection through a multi-layer perceptron and a fuzzy clustering algorithm. The logistics path optimization unit realizes optimal scheduling of logistics paths based on a genetic algorithm and an adaptive large-scale neighborhood search (ALNS) algorithm. According to the system, by integrating a plurality of functional units, the operation efficiency and the response speed of the logistics supply chain are effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of supply chain management, and in particular to a digital management system for a smart logistics supply chain based on data analysis. Background Art

[0002] At present, the main challenge facing the supply chain management system is how to improve the collaborative efficiency of each link through intelligent means, especially when dealing with demand fluctuations, inventory management and supplier relationship management. The traditional logistics supply chain management system relies on manual experience and cannot accurately analyze market data and historical order data, resulting in inaccurate forecast results and difficulty in adapting to the dynamic changes in market demand. The existing inventory management system is difficult to adjust in real time and cannot effectively respond to the dynamic changes in inventory, which may lead to excess or shortage of inventory and increase inventory costs. Supplier selection relies on historical performance and lacks analysis of real-time supplier data, resulting in the inability to respond to demand in a timely manner during periods of high demand fluctuations, affecting the overall efficiency of logistics. Summary of the invention

[0003] The present invention provides a digital management system for intelligent logistics supply chain based on data analysis to solve the above technical problems.

[0004] The present invention provides a digital management system for intelligent logistics supply chain based on data analysis, the system comprising: The demand forecasting and analysis unit analyzes historical orders and market fluctuation data through recursive neural networks and long short-term memory networks, predicts short-term and medium-term demand, and outputs demand forecast data; An inventory optimization unit, combining a Bayesian dynamic linear model and a convolutional neural network algorithm, adjusts the inventory level according to the demand forecast data output by the demand forecast and analysis unit; Supplier management unit: Based on the multi-layer perceptron and fuzzy clustering algorithm, it receives the inventory adjustment data of the inventory optimization unit and selects, classifies and sorts suppliers according to the supplier data; the unit also formulates supplier delivery plans and adjusts order allocation strategies based on the supplier's delivery capabilities and historical performance; The logistics path optimization unit adopts a genetic algorithm and an adaptive large-scale neighborhood search algorithm to perform logistics path planning based on the supplier selection results of the supplier management unit and real-time logistics data; The system integration and feedback unit receives data from the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit, and the logistics path optimization unit through a global data management and feedback mechanism, and integrates and synchronizes the operation of each unit; The real-time logistics marketing unit receives the feedback data from the system integration and feedback unit, and adjusts the logistics marketing strategy based on the logistics data and market trends.

[0005] Furthermore, the demand forecasting and analysis unit includes: Short-term forecasting module: Generates short-term demand forecast results through recurrent neural networks and passes the forecast results to the inventory optimization unit to adjust inventory levels; Medium- and long-term forecasting module: using the long short-term memory network to generate medium- and long-term demand forecasts, and transmitting the output results of the long short-term memory network combined with the short-term forecast results of the recursive neural network algorithm to the inventory optimization unit and the logistics path optimization unit to formulate long-term inventory and logistics scheduling plans; Attention mechanism module: assigning different weights to the prediction data output by the long short-term memory network, and transmitting the weighted prediction data to the supplier management unit and the logistics path optimization unit; Feedback module: collects the operation data of the inventory optimization unit, the logistics path unit and the supplier management unit, and transmits the feedback information to the recursive neural network and the long short-term memory network to adjust the model parameters according to the actual operation.

[0006] Furthermore, the long short-term memory network in the medium- and long-term prediction module of the demand forecasting and analysis unit includes an input gate, a forget gate and an output gate, wherein the input gate is used to select current input information, the forget gate is used to adjust the weight of historical data to the current prediction, and the output gate is used to generate a prediction result.

[0007] Furthermore, the inventory optimization unit includes: Data collection module: used to obtain data from the demand forecasting and analysis unit, supplier management unit, logistics path optimization unit and inventory optimization unit. The data includes historical order data, real-time order volume, market fluctuation data, supplier delivery information and logistics status data. The data is standardized and then transferred to the inventory forecasting module and inventory data module to provide a basis for inventory forecasting and adjustment. Inventory forecasting module: Combining the Bayesian dynamic linear model and the convolutional neural network for inventory forecasting; the Bayesian dynamic linear model is used for short-term inventory forecasting, and the convolutional neural network is used to analyze long-term inventory trends; the forecast results are transmitted to the inventory adjustment module to provide a basis for short-term and long-term inventory regulation.

[0008] Inventory adjustment module: dynamically adjusts inventory levels based on the forecast results of the inventory forecast module; Feedback processing module: collects operation feedback information from the supplier management unit and the logistics path optimization unit, and transmits the operation feedback information to the inventory forecasting module and the inventory adjustment module.

[0009] Furthermore, the convolutional neural network in the inventory optimization unit includes multiple layers of convolutional layers and pooling layers, the convolutional layers are used to process local patterns in the inventory data, and the pooling layers are used to reduce the data dimension and extract long-term trend characteristics of the inventory data.

[0010] Furthermore, the supplier management unit includes: Supplier data acquisition module: used to obtain supplier-related data from various units in the supply chain, including supplier delivery time, order completion rate, inventory demand data and logistics status information; the acquired data is standardized and passed to the supplier analysis module and the supplier classification and sorting module; Supplier analysis module: Based on the multi-layer perceptron algorithm, it analyzes the supplier's historical delivery records, response speed and delivery stability, and generates supplier priorities; according to the analysis results, it adjusts the supplier delivery plan and passes the supplier priority ranking results to the supplier classification and sorting module; Supplier classification and sorting module: Use fuzzy clustering algorithm to classify and sort suppliers; determine supplier classification results based on supplier response speed, delivery stability and inventory requirements and pass them to the inventory optimization unit and logistics path optimization unit; Supplier feedback processing module: used to collect and process operational data from other units in the supply chain, including supplier delivery time, logistics execution status and inventory turnover rate; and pass the feedback data to the supplier analysis module.

[0011] Furthermore, the logistics path optimization unit includes: Logistics data acquisition module: used to acquire logistics-related data, including inventory status, demand forecast information, traffic conditions and transportation vehicle status information; the acquired data is processed and then transmitted to the path planning module and resource scheduling module; Path planning module: Generates the optimal path plan based on genetic algorithm and Dijkstra algorithm; the genetic algorithm is used to generate the initial path solution set, and the Dijkstra algorithm calculates the shortest path plan based on the initial solution and passes the plan to the resource scheduling module; Resource scheduling module: Based on the path plan generated by the path planning module, the adaptive large-scale neighborhood search algorithm is used to schedule logistics resources; Path feedback module: used to collect actual execution data of logistics paths, including transportation time, resource utilization and transportation delays; pass feedback data to the path planning module and resource scheduling module to adjust the path planning and resource scheduling plans.

[0012] Furthermore, the initial path solution set of the genetic algorithm is set to 50 to 100.

[0013] Furthermore, the system integration and feedback unit includes: Global data management module: used to obtain data from the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit and the logistics path optimization unit; standardize the obtained data and transmit it to the feedback optimization module and the intelligent collaboration module; Feedback optimization module: according to the data processed by the global data management module, combined with the inventory level, logistics status and the actual response of the supplier, adjust the operating parameters of the demand forecasting and analysis unit, the inventory optimization unit and the logistics path optimization unit; Intelligent collaboration module: coordinates the operation of the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit and the logistics path optimization unit by analyzing the data provided by the global data management module; Dynamic monitoring module: Identifies potential problems and provides early warning signals by monitoring logistics delays, inventory imbalances and supplier responses.

[0014] Furthermore, the real-time logistics marketing unit includes: Logistics data monitoring module: used to monitor key data in the logistics process, including inventory levels, logistics route status, order completion status and customer feedback information; the monitored data is transmitted to the market trend prediction module and user behavior analysis module; Market trend prediction module: predict future market demand trends based on historical sales data, market fluctuation data and logistics data; pass the prediction results to the user behavior analysis module and marketing feedback module; User behavior analysis module: Generates marketing plans for different customer groups by analyzing customers' order history, purchase behavior and logistics feedback data; transmits the generated marketing plans to the marketing feedback module and inventory optimization unit; Marketing feedback module: used to collect execution data of marketing activities and customer feedback on logistics services; pass feedback data to the market trend prediction module and user behavior analysis module to adjust market prediction and marketing strategies.

[0015] The present invention provides a digital management system for intelligent logistics supply chain based on data analysis, which realizes comprehensive optimization of supply chain management by integrating multiple functional units, and has the following beneficial effects: 1. Improved demand forecast accuracy: Through the combination of recurrent neural networks (RNN) and long short-term memory networks (LSTM), historical order and market fluctuation data can be effectively processed to generate accurate short-term and medium-term demand forecasts. This enables the system to respond quickly when demand fluctuates, ensuring the balance of supply and demand in the supply chain.

[0016] 2. Improved inventory management efficiency: The inventory optimization unit uses Bayesian dynamic linear model and convolutional neural network to achieve dynamic inventory forecasting and adjustment. By combining short-term and long-term inventory forecasts, the system can flexibly adjust inventory levels when market demand changes, avoid inventory backlogs or shortages, and reduce inventory holding costs.

[0017] 3. Supplier selection optimization: The supplier management unit uses a multi-layer perceptron (MLP) and fuzzy clustering algorithm to dynamically analyze suppliers’ historical performance and real-time response data, thereby optimizing supplier selection, classification, and sorting, ensuring that the best suppliers are selected during periods of high demand fluctuations and improving the overall efficiency of the supply chain.

[0018] 4. Logistics Path Optimization: The logistics path optimization unit combines genetic algorithms, adaptive large-scale neighborhood search algorithms (ALNS) and Dijkstra algorithms to achieve optimal planning and resource scheduling of logistics paths. This ensures efficient use of logistics resources and reduces transportation delays and resource waste.

[0019] 5. Enhanced system integration and feedback mechanism: Through the global data management and feedback optimization of the system integration and feedback unit, each unit of the supply chain can be adjusted dynamically in real time. This module combines supplier response, inventory status and logistics execution to ensure the overall optimization and efficient operation of the system.

[0020] 6. Real-time marketing strategy adjustment: The real-time logistics marketing unit can dynamically adjust marketing strategies based on data analysis of logistics and market trends, improve customer experience and order completion rate, while optimizing inventory marketing to ensure a close match between logistics and market demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic diagram of the structure of the system provided by the present invention; Figure 2 It is a structural schematic diagram of the demand prediction and analysis unit in the present invention; Figure 3 It is a structural schematic diagram of the inventory optimization unit in the present invention; Figure 4 It is a structural diagram of the supplier management unit in the present invention; Figure 5It is a structural schematic diagram of the logistics path optimization unit in the present invention; Figure 6 It is a structural diagram of the system integration and feedback unit in the present invention; Figure 7 It is a structural schematic diagram of the real-time logistics marketing unit diagram in the present invention. DETAILED DESCRIPTION

[0023] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Preferred embodiments of the present invention are provided in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0024] Example 1: Please refer to Figure 1 This embodiment provides a digital management system for intelligent logistics supply chain based on data analysis, and the system includes:

[0025] The demand forecasting and analysis unit analyzes historical orders and market fluctuation data through recursive neural networks and long short-term memory networks, predicts short-term and medium-term demand, and outputs demand forecast data; An inventory optimization unit, combining a Bayesian dynamic linear model and a convolutional neural network algorithm, adjusts the inventory level according to the demand forecast data output by the demand forecast and analysis unit; Supplier management unit: Based on the multi-layer perceptron and fuzzy clustering algorithm, it receives the inventory adjustment data of the inventory optimization unit and selects, classifies and sorts suppliers according to the supplier data; the unit also formulates supplier delivery plans and adjusts order allocation strategies based on the supplier's delivery capabilities and historical performance; The logistics path optimization unit adopts a genetic algorithm and an adaptive large-scale neighborhood search algorithm to perform logistics path planning based on the supplier selection results of the supplier management unit and real-time logistics data; The system integration and feedback unit receives data from the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit, and the logistics path optimization unit through a global data management and feedback mechanism, and integrates and synchronizes the operation of each unit; The real-time logistics marketing unit receives the feedback data from the system integration and feedback unit, and adjusts the logistics marketing strategy based on the logistics data and market trends.

[0026] This embodiment uses recursive neural networks (RNN) and long short-term memory networks (LSTM) to accurately predict short-term and medium- and long-term demand, ensuring that the supply chain responds quickly to market changes. Combining the Bayesian dynamic linear model and convolutional neural network, it is possible to dynamically adjust inventory levels, avoid excess or shortages, and improve inventory management efficiency. Based on multi-layer perceptrons and fuzzy clustering algorithms, efficient selection and sorting of suppliers are achieved to ensure stable operation and efficient supply of the supply chain. Through genetic algorithms and adaptive large-scale neighborhood search algorithms, logistics routes are optimized, transportation delays are reduced, and transportation costs are reduced. Global data management and feedback mechanisms ensure the synchronous operation of various units in the system and improve the overall collaborative efficiency of the supply chain. Based on logistics data and market trends, marketing strategies can be adjusted in a timely manner to enhance customer experience and increase order completion rates.

[0027] Example 2: Please refer to Figure 2 This embodiment provides a specific implementation method of a demand forecasting and analysis unit in a digital management system for a smart logistics supply chain based on data analysis. The demand forecasting and analysis unit includes:

[0028] Short-term prediction module: The historical order data is processed through the loop structure of the recurrent neural network (RNN). RNN combines a specific time step setting to gradually input the time series data into the network. The time series data includes historical order data, market volatility data, seasonal factors, and environmental variable data. RNN combines the hidden state of the previous time step with the current input data through its hidden state recursive mechanism to generate the prediction result of the next time step, thereby capturing the short-term dependency in the time series data. This short-term dependency reflects the correlation between historical order data in adjacent time steps. RNN recursively models this correlation through multiple layers of hidden units, which can effectively identify the short-term fluctuation trend of order volume and generate accurate short-term prediction results based on it.

[0029] The setting of the time step is based on the order data characteristics and prediction requirements of the specific application scenario, and defines the historical steps that the RNN refers to when processing time series data. The time step is usually adjusted according to the fluctuation frequency and periodicity of the order data, which can be determined in the following ways: (1) Data feature analysis: Through statistical analysis of historical order data, its fluctuation period and short-term change pattern are identified, and then the appropriate time step is determined. For example, if the order data fluctuates daily, the time step can be set to several days within a working cycle. (2) Experimental verification: Through experimental comparison of the prediction effects of different time steps, the optimal time step that can ensure prediction accuracy while improving training efficiency is selected. Usually, the time step is gradually increased until the prediction accuracy is no longer significantly improved. (3) Dynamic adjustment: In some complex application scenarios, the time step can be dynamically adjusted based on system feedback. When order demand fluctuates frequently, the time step can be shortened to enhance the response to short-term changes; when order fluctuations tend to be stable, the time step can be appropriately increased to improve the prediction accuracy of long-term trends.

[0030] The short-term order forecast results generated by RNN include the order demand, demand fluctuation trend and related market changes in the future. These forecast results are calculated through the output layer of the recursive neural network, the hidden layer state is processed based on the activation function, and the model parameters are optimized through the Adam optimizer to improve the forecast accuracy and model training efficiency.

[0031] 2. Medium- and long-term forecasting module: This module uses the long short-term memory network (LSTM) to analyze historical order data and market fluctuation data and capture long-term dependencies in time series. The input gate of LSTM is used to select current input information through its gating mechanism, the forget gate is used to adjust the weight of historical data on the current forecast, and the output gate is used to generate forecast results. This module handles demand forecasts with longer time spans to cope with long-term fluctuation trends in the supply chain.

[0032] The parameter settings of LSTM include the number of hidden layer units, which is usually set to 128 or 256. The appropriate value is selected according to the data scale. The optimization strategy uses the Adam optimizer to adjust the learning rate to ensure that the model can adapt to changes in different data distributions, thereby improving the prediction accuracy. The input data includes short-term prediction results generated by the recurrent neural network (RNN), historical order data, market trend data and environmental factors, and these data are combined to generate medium- and long-term demand forecast results.

[0033] The medium- and long-term forecast results are used in the inventory optimization unit, logistics path optimization unit, and supplier management unit. Specifically, the forecast results generated by LSTM are applied to the convolutional neural network (CNN) in the inventory optimization unit to help identify long-term trends in inventory data and optimize inventory configuration. In addition, the forecast results also support the adaptive large-scale neighborhood search algorithm (ALNS) in the logistics path optimization unit, providing a basis for long-term logistics resource scheduling and ensuring the rationality of logistics planning and efficient resource allocation. At the same time, the medium- and long-term forecast results provide guidance for the multi-layer perceptron (MLP) and fuzzy clustering algorithm in the supplier management unit to ensure the long-term stability and adaptability of supplier selection and order allocation.

[0034] 3. Attention mechanism module: It is used to perform weighted processing on the medium- and long-term forecast data output by the long short-term memory network (LSTM). By assigning different weights to the input data at different time steps, it dynamically adjusts the model's attention to key nodes in the time series. This module is suitable for capturing seasonal fluctuations and sudden market events, thereby improving the accuracy of demand forecasting and the responsiveness of the system.

[0035] The attention mechanism implements weighting through two main algorithms: (1) additive attention mechanism, which assigns weights to each time step by calculating the similarity between the hidden state of the current time step and the historical data. The parameters of this algorithm include the dimension of the hidden unit, which is usually set to 64 or 128 to adjust the sensitivity of the model to more distant time steps; (2) multiplicative attention mechanism, which calculates weights through dot product operations and is suitable for large-scale data processing. The parameters of the multiplicative mechanism include the learning rate and dot product dimension, where the learning rate is usually set to 0.001 or 0.0005 and the dot product dimension is set to 32 or 64 to control the accuracy of the attention weights and the training speed.

[0036] Different algorithms and parameter settings have a significant impact on model results. The additive attention mechanism performs better in subtle seasonal changes, while the multiplicative attention mechanism is more suitable for processing large-scale historical data. The setting of the learning rate directly affects the convergence speed of the model and its responsiveness to new data. A higher learning rate (such as 0.001) is suitable for a rapidly changing market environment, while a lower learning rate (such as 0.0005) helps improve the stability of the model.

[0037] The prediction results after attention weighting are passed to the inventory optimization unit and the logistics path optimization unit to further improve the decision-making ability of supply chain management. Specifically, the convolutional neural network (CNN) in the inventory optimization unit uses weighted data to identify long-term trends in inventory data to ensure that inventory configuration can be flexibly adjusted when demand fluctuates. The adaptive large-scale neighborhood search algorithm (ALNS) in the logistics path optimization unit prioritizes areas or time periods with large demand fluctuations based on weighted prediction results, thereby reducing the waste of logistics resources and transportation delays.

[0038] Through the dynamic weighted processing of the attention mechanism, the system can significantly improve its ability to adapt to market changes and ensure that each unit in the supply chain can respond accurately at key time points.

[0039] 4. Feedback module: used to collect and process real-time feedback data from the inventory optimization unit, logistics route optimization unit and supplier management unit. These feedback data include operating parameters such as inventory level, logistics transportation delay and supplier response speed. The feedback module transmits this data to each module of the demand forecasting and analysis unit, dynamically adjusts the parameters of the recursive neural network (RNN), long short-term memory network (LSTM) and attention mechanism, so as to optimize the forecasting accuracy and system response efficiency.

[0040] The core functions of the feedback module include data collection, dynamic parameter adjustment, and feedback response. Its parameter settings mainly include the frequency and granularity of feedback data. The frequency of feedback data can be set to hourly or daily, depending on the real-time requirements of the system; the feedback granularity is selected based on the operational complexity of each unit in the supply chain, usually divided into order level or inventory level.

[0041] The feedback module uses two algorithms to process feedback data: (1) Adaptive filtering algorithm, which recursively adjusts model parameters and is suitable for scenarios with fast response. The key parameters of this algorithm are the filter order and adjustment step size. The filter order is usually set to 3 or 5, and the adjustment step size is 0.01 to 0.05 to control the sensitivity of the model to short-term fluctuations; (2) Bayesian optimization algorithm, which is used to dynamically adjust model parameters in complex scenarios, especially when there are large fluctuations in the system. The main parameters of this algorithm are the sampling frequency and the confidence interval of the objective function. The sampling frequency is set to once every 10 minutes, and the confidence interval range is set to 95% to 99% to ensure that the prediction accuracy is maintained during large market changes.

[0042] Different algorithms and parameter settings have a significant impact on the results of the feedback module. For example, the adaptive filtering algorithm can prioritize the adjustment of inventory holding strategies and improve system response speed when order demand fluctuates rapidly, while the Bayesian optimization algorithm is suitable for complex and long-term fluctuation scenarios, ensuring that logistics routes and resource scheduling are optimized when dealing with supplier delivery delays or logistics bottlenecks.

[0043] The optimization results of the feedback module directly affect the inventory optimization unit and the logistics path optimization unit. Specifically, when the logistics path optimization unit reports frequent transportation delays, the feedback module will instruct the attention mechanism to increase the weight of this time period to improve the model's ability to identify abnormal demand. At the same time, the feedback module adjusts the time steps of the RNN and the weight of the forget gate of the LSTM to ensure that market fluctuations can be captured more accurately in future forecasts, optimizing the overall responsiveness and resource utilization of the supply chain.

[0044] Example 3: Please refer to Figure 3 In this embodiment, the inventory optimization unit includes a data acquisition module, an inventory forecasting module, an inventory adjustment module, a feedback processing module, and an inventory data module. The modules work closely together to achieve real-time prediction and dynamic adjustment of inventory status by combining the Bayesian dynamic linear model (BDLM) and the convolutional neural network (CNN) algorithm. The algorithms of the inventory optimization unit form positive feedback with the algorithms of the supplier management unit and the logistics path optimization unit to ensure the efficient operation of the overall supply chain system.

[0045] 1. Data collection module: acquires data from each supply chain unit in real time. Input data includes historical order data, real-time order volume, market fluctuation data, supplier delivery information and logistics status data. After being standardized, these data are passed to the inventory forecasting module and inventory adjustment module. Standardization ensures the consistency of data format and the accuracy of subsequent calculations. The output data is standardized order data, supplier delivery information and logistics status information, which are input into the inventory forecasting module and inventory adjustment module. The data collection module optimizes data processing efficiency by adjusting the data update frequency (such as hourly or daily) and the granularity of data collection (order level, inventory level, logistics level). By introducing clustering algorithms to filter and classify data, redundant data can be reduced and the response speed of the overall system can be improved.

[0046] 2. Inventory forecasting module: Combining Bayesian dynamic linear model (BDLM) and convolutional neural network (CNN) for inventory forecasting. The input data comes from the standardized historical order data, real-time order data, market fluctuation information and logistics status data of the data acquisition module. BDLM is used to predict short-term inventory demand, while CNN is used for long-term inventory trend analysis. The output data includes short-term inventory forecast results (based on BDLM) and medium- and long-term inventory forecast results (based on CNN), which are passed to the inventory adjustment module for dynamic adjustment of inventory levels, and passed to the supplier management unit and logistics path optimization unit to optimize supplier selection and logistics path planning. BDLM uses a recursive filtering algorithm to adjust model parameters in real time to adapt to the rapid changes in short-term inventory demand, thereby improving the system's sensitivity to short-term fluctuations. CNN extracts long-term trends in historical data and identifies long-term inventory change patterns through a multi-layer convolution structure. The number of convolutional layers can be set to 4 to 6 layers, and the pooling layer size can be set to 2x2 or 3x3. The optimized model can more accurately capture the long-term features in inventory data, thereby improving the accuracy of long-term forecasts.

[0047] 3. Inventory adjustment module: Based on the short-term and medium-term forecast results generated by the inventory forecast module, the inventory level is dynamically adjusted to adapt to current demand. The input data comes from the short-term inventory forecast (based on BDLM) and medium-term inventory forecast (based on CNN) of the inventory forecast module, including specific inventory demand forecasts and inventory fluctuation trends. The output data comes from the inventory adjustment plan, including inventory replenishment and reduction plans; and is passed to the supplier delivery plan module in the supplier management unit and the resource scheduling module in the logistics path optimization unit. The inventory adjustment module dynamically adjusts the inventory according to the inventory safety level, adjustment frequency and adjustment threshold parameters. The inventory safety level can be set to 10%-20% and the adjustment frequency can be set to daily or hourly. By setting a dynamic adjustment strategy, the inventory adjustment module can quickly respond to insufficient or excessive inventory problems in the short term and optimize inventory turnover.

[0048] 4. Feedback processing module: collects feedback data from the supplier management unit and the logistics path optimization unit, such as supplier delivery time, logistics execution status and inventory turnover rate. These feedback data are used to adjust the model parameters of the inventory forecasting and inventory adjustment modules in real time to ensure the efficient operation of the supply chain. The input data comes from the feedback information of the supplier management unit and the logistics path optimization unit, including supplier delivery data, logistics path status and inventory execution data. The output data includes the optimized inventory forecast and adjustment strategy, which are passed to the inventory forecasting module and the inventory adjustment module.

[0049] This module uses adaptive filtering algorithm and Bayesian optimization algorithm to process feedback data. The adaptive filtering algorithm is used to dynamically adjust inventory strategies. The filter order can be set to 3 or 5, and the adjustment step can be set to 0.01 to 0.05. The Bayesian optimization algorithm ensures the accuracy of prediction when demand fluctuates greatly by adjusting the feedback frequency and confidence interval (95%-99%). Through the dynamic optimization of the feedback processing module, the system can adaptively adjust the parameters of each module, thereby improving the overall efficiency of supply chain management.

[0050] Example 4: Please refer to Figure 4 This embodiment provides a possible implementation of a supplier management unit, which includes a supplier data acquisition module, a supplier analysis module, a supplier classification and sorting module, and a supplier feedback processing module. By working in collaboration with the inventory optimization unit and the logistics path optimization unit, the supplier selection and order allocation strategies are dynamically optimized to improve the efficiency and response speed of supply chain management. The close collaboration between the supplier management unit and the inventory optimization unit and the logistics path optimization unit improves the overall efficiency and flexibility of the supply chain through a positive feedback mechanism.

[0051] 1. Supplier data acquisition module: This module collects supplier data from various supply chain units in real time, including delivery time, order completion rate, inventory demand data and logistics status information. Through data standardization processing, the consistency and accuracy of input data are ensured, providing basic data support for subsequent modules. Input data includes supplier delivery time, order completion rate, inventory status, and logistics status data. The output data includes standardized supplier data and is passed to the supplier analysis module and the supplier classification and sorting module. The data update frequency can be set to hourly or daily according to actual needs, which is suitable for different supply chain complexities. Through data screening and standardization processing, data quality and processing efficiency are ensured.

[0052] 2. Supplier Analysis Module: The multi-layer perceptron (MLP) algorithm is used to generate supplier priorities by analyzing the supplier's historical delivery records, response speed, delivery stability and other indicators. The input data includes supplier delivery time, historical performance, order completion rate and inventory status. The output data includes the supplier priority sorting results and is passed to the supplier classification and sorting module. The number of hidden layer units of MLP can be set to 128 or 256, and the learning rate can be set to 0.001. More hidden units can improve the model's ability to recognize complex supplier features, and the reasonable adjustment of the learning rate can strike a balance between accuracy and efficiency.

[0053] 3. Supplier classification and sorting module: This module uses a fuzzy clustering algorithm to classify suppliers and sort them according to their response speed, delivery stability, and inventory requirements. The input data comes from the supplier priority sorting results, response speed, delivery stability, etc. of the supplier analysis module. The output data includes the supplier classification results and is passed to the inventory optimization unit and the logistics path optimization unit. The fuzzy clustering algorithm segments suppliers according to the number of clusters (3-5 clusters) and the classification threshold (0.05) to ensure that each supplier can be accurately classified. Through this classification, the order allocation strategy can be optimized and the response speed and resource allocation efficiency of the supply chain can be improved.

[0054] 4. Supplier feedback processing module: This module is used to collect feedback information from the inventory optimization unit and the logistics path optimization unit, and dynamically adjust the supplier selection and sorting strategy. The input data includes supplier delivery time, inventory turnover rate, and logistics execution status. The output data includes the optimized supplier selection and sorting results, which are fed back to the supplier analysis module. The adaptive filtering algorithm is used to adjust the supplier sorting and delivery strategy in real time. The filter order (which can be set to 3 or 5) and the adjustment step size (which can be set to 0.01 to 0.05) ensure the timeliness and accuracy of feedback adjustment. The Bayesian optimization algorithm dynamically adjusts the sampling frequency and confidence interval (95% to 99%) of the feedback data to ensure the efficiency and adaptability of supplier management in dealing with market fluctuations.

[0055] Through the close collaboration of various modules in this embodiment, the supplier management unit can quickly adjust the supplier selection strategy when order demand, inventory level and logistics status change, ensuring the efficient operation of the supply chain. The MLP algorithm and fuzzy clustering algorithm effectively improve the response speed of each link in the supply chain by optimizing supplier classification and sorting, while the adaptive filtering algorithm and Bayesian optimization algorithm ensure that in a complex and changing supply chain environment, the system can adaptively adjust the supplier strategy to achieve optimal resource allocation.

[0056] Example 5: Please refer to Figure 5 The logistics path optimization unit in this embodiment is composed of a logistics data acquisition module, a path planning module, a resource scheduling module, and a path feedback module. Each module works together, combining a genetic algorithm, an adaptive large-scale neighborhood search algorithm (ALNS), and a Dijkstra algorithm to achieve dynamic optimization and efficient scheduling of logistics paths. The algorithms of each module form a positive feedback effect with the algorithms of the inventory optimization unit and the supplier management unit to ensure the improvement of the overall efficiency of the supply chain and resource utilization.

[0057] 1. Logistics data acquisition module: responsible for obtaining logistics-related data from the supply chain system in real time. These data include inventory status, demand forecast information, traffic conditions, and transportation vehicle status. After standardization, they are passed to the path planning and resource scheduling modules. Input data includes inventory status, demand forecast information, traffic conditions, transportation vehicle status data, etc. The output data is standardized logistics status data, which is passed to the path planning module and resource scheduling module.

[0058] The frequency of data collection can be set to hourly or minutely according to demand to ensure the real-time and accuracy of logistics status data. This module improves data quality and processing efficiency through data screening and standardization, ensuring that the path planning and resource scheduling of subsequent modules can make decisions based on accurate data.

[0059] 2. Path planning module: First, a set of initial path solutions is generated using a genetic algorithm. These solutions represent multiple possible paths from the starting point to the destination. The genetic algorithm continuously optimizes the path solution set by simulating the process of biological evolution through crossover and mutation operations. The generated preliminary path solution set is then passed as input to the Dijkstra algorithm, which further calculates the local optimal solution of each path based on these preliminary solutions and outputs the final shortest path.

[0060] The role of the genetic algorithm in the path planning module is to provide the Dijkstra algorithm with an initial path solution set. It generates initial solutions randomly and optimizes the solution set through an evolutionary process. Population initialization: The genetic algorithm first generates a set of random path solution sets (called populations), each of which corresponds to a feasible path from the starting point to the end point. These paths are based on the input data of the logistics data acquisition module, taking into account the current inventory status, traffic conditions, and vehicle availability. Population size: The size of the path solution set is usually set to 50 to 100, and a larger population helps to cover more path selection space. Fitness evaluation: The fitness of each path solution is evaluated based on its path length, traffic conditions, and estimated transportation time. The path solution with higher fitness indicates that it is closer to the optimal solution. The fitness function is mainly calculated by the following parameters: Path length is the total distance of the path. Transportation time: The transportation time estimated based on the current traffic conditions. Resource consumption: Including vehicle fuel consumption, personnel costs, etc. Selection operation: The genetic algorithm selects path solutions with higher fitness from the current population to enter the next generation through roulette or tournament selection to improve the average quality of the path solution set. Crossover operation: The crossover operation exchanges parts of the paths of two parent path solutions to generate a new path solution. The crossover rate is usually set to 0.7 to 0.9. The crossover operation ensures information exchange between different paths and promotes path optimization. Mutation operation: In order to prevent the algorithm from falling into the local optimum, the genetic algorithm performs mutation operations on partial path solutions and changes some nodes in the path. The mutation rate is usually set to 0.01 to 0.05. A lower mutation rate ensures a stable search, while a higher mutation rate helps the search space jump out of the local optimum. Iterative evolution: The genetic algorithm continuously optimizes the path solution set through multiple generations of iterations. It is usually set to 10 to 20 generations until a set of high-quality path solutions is found.

[0061] Based on the preliminary path solution set generated by the genetic algorithm, the Dijkstra algorithm further calculates the local optimal path of each path solution to ensure that the shortest path is selected under the specified traffic conditions and transportation conditions. Path update: The Dijkstra algorithm gradually expands the path by selecting each node in the path from the starting point in sequence, and assigns the shortest path distance to each path node. Each time the Dijkstra algorithm calculates the shortest distance from the starting point to the current node, and records the optimal path. Shortest path calculation: The Dijkstra algorithm calculates the shortest transportation time and path length of each path under actual traffic conditions based on the initial path of each path solution. For example, in urban logistics, the Dijkstra algorithm will give priority to roads with smooth traffic and bypass traffic congestion areas, thereby reducing transportation time. Final path output: The path solution optimized by the Dijkstra algorithm is finally output as the shortest path solution, including the node sequence passed, the total distance and the estimated transportation time. These results are passed to the resource scheduling module for scheduling logistics resources.

[0062] The genetic algorithm in the path planning module generates multiple feasible path solutions through a large-scale global search, avoiding the local optimal trap caused by the initial selection. The Dijkstra algorithm ensures that each path is the shortest path under the current traffic conditions by performing refined local optimal calculations on each path solution. The crossover and mutation operations of the genetic algorithm can quickly respond to the complexity in the logistics network (such as multi-node and multi-path selection problems). The Dijkstra algorithm can dynamically update the path selection, respond to changes in traffic conditions in real time, and ensure the timeliness and efficiency of transportation.

[0063] Resource Scheduling Module: Responsible for dynamically scheduling vehicles and other logistics resources based on the path plan generated by the path planning module and the current state of logistics resources. The resource scheduling module uses the adaptive large-scale neighborhood search algorithm (ALNS) to ensure efficient resource scheduling under complex transportation requirements and resource constraints, reduce logistics bottlenecks and optimize resource utilization.

[0064] 1. Input data includes the optimal path plan from the path planning module, including path planning data of the starting point, destination, nodes passed, total distance and estimated transportation time; transportation resources such as the real-time location, load, availability and current transportation task status of the transportation vehicle.

[0065] 2. Output data: including the optimized resource scheduling plan of specific transportation task allocation, vehicle route selection and scheduling time arrangement, and transmitted to the route feedback module and each execution unit.

[0066] 3. The specific process is as follows: (1) Initial solution generation: Based on the path plan output by the path planning module, an initial scheduling plan is generated. The initial plan makes a preliminary allocation based on the current load, availability, and transportation demand of the transport vehicles.

[0067] (2) Neighborhood search: ALNS searches multiple neighborhoods to explore different scheduling options and find the optimal resource allocation method. The definition of the neighborhood is based on multiple conditions, including: considering the vehicle load, optimizing the distribution of goods by dynamically adjusting the load of each vehicle. The time window neighborhood adjusts the time for the vehicle to arrive at each node according to the supplier's delivery time and order priority. Combined with the path planning results, the transportation path of the vehicle is dynamically optimized to ensure the path neighborhood with the optimal transportation cost and time.

[0068] (3) Local adjustment and global optimization: During the search process, ALNS gradually improves the scheduling plan by adjusting the local neighborhood, such as exchanging tasks between vehicles and adjusting the scheduling order. Through global search, it ensures that the global optimal solution is found under the condition of limited resources.

[0069] (4) Parameter settings: Neighborhood size: The size of the neighborhood determines the breadth of the search space. A larger neighborhood can cover more resource allocation schemes and is suitable for complex transportation tasks; a smaller neighborhood helps improve computational efficiency and is suitable for simpler scheduling tasks.

[0070] Search depth: The search depth determines the number of iterations of the algorithm in each neighborhood. A higher search depth (such as 50-100 times) is suitable for large-scale logistics networks to ensure that the optimal solution is found in complex scenarios.

[0071] Adjust rules: including vehicle load limits, transportation time windows, etc., to ensure that all constraints are met during the scheduling process.

[0072] Through the dynamic scheduling of the adaptive large-scale neighborhood search algorithm (ALNS), the resource scheduling module can optimize the utilization of transportation resources in real time, reduce empty load rate and resource waste. By combining neighborhood search with local adjustment, the system can effectively respond to the dynamic changes in transportation demand and ensure efficient logistics scheduling in a complex supply chain environment.

[0073] 4. Path feedback module: The input data of this module include transportation time (actual transportation time of each transportation node, including the actual driving time of the vehicle from the starting point to each intermediate node and the final destination); resource utilization (load utilization of the vehicle during transportation, the actual amount of goods transported and the empty rate of the vehicle, etc.); transportation delays (monitor whether there are any transportation delays relative to the planned time); other logistics status (including real-time data such as road traffic conditions, vehicle location, inventory status, etc.). Output data includes feedback optimization data (optimized path planning adjustment suggestions, including suggestions for reallocating transportation vehicles, changing transportation time or adjusting transportation routes, and passed to the path planning module and resource scheduling module); improvement plan (specific improvement plans proposed for problems in logistics execution, and fed back to the inventory optimization unit and supplier management unit to adjust inventory strategies and supplier selection plans); The specific process is as follows: (1) Feedback data collection: The route feedback module collects data from all aspects of the transportation execution process, including transportation time, resource utilization, and transportation delays. These data mainly come from on-board monitoring equipment, real-time monitoring information from the logistics management system, and status reports provided by suppliers and logistics.

[0074] (2) Feedback analysis and processing: The module uses adaptive filtering algorithm and Bayesian optimization algorithm to analyze and process the collected feedback data. The adaptive filtering algorithm is mainly used to monitor and quickly adjust the transportation time and resource utilization in real time; the Bayesian optimization algorithm is used to adjust the overall logistics path planning and resource allocation plan according to the transportation delay situation to ensure the optimal logistics execution effect.

[0075] The adaptive filtering algorithm adjusts the dynamic parameters in the transportation task allocation (such as the scheduling of transportation vehicles, transportation sequence, etc.) to promptly correct problems caused by logistics delays or insufficient resources. The filter order (such as 3 to 5 orders) and the adjustment step size (such as 0.01 to 0.05) affect the speed and accuracy of the feedback response.

[0076] The Bayesian optimization algorithm dynamically adjusts the planning strategy in the path planning module by analyzing problems such as transportation delays, especially when traffic is congested or vehicles are unavailable, and can prioritize the most urgent paths to ensure overall logistics efficiency. The sampling frequency and confidence interval (95% to 99%) settings ensure accurate feedback and optimization under different conditions.

[0077] Parameter settings: (1) Feedback frequency: Depending on the real-time nature of logistics needs, the feedback frequency can be set to hourly or real-time feedback to ensure that any delays or waste of resources during transportation can be captured and handled in a timely manner.

[0078] (2) Feedback granularity: The granularity of data collection can be refined to the order level or vehicle level, and the level of data detail can be adjusted according to demand to more accurately optimize logistics routes and resource allocation.

[0079] In summary, this embodiment combines genetic algorithms, adaptive large-scale neighborhood search algorithms (ALNS) and Dijkstra algorithms, and the path planning module can perform global search and combine local optimization to ensure the optimal selection of logistics paths and reduce transportation time and costs. The resource scheduling module uses the adaptive large-scale neighborhood search algorithm (ALNS) to efficiently schedule logistics resources under complex transportation needs and resource constraints, maximize resource utilization, and reduce empty load rate and resource waste. The logistics data acquisition module works closely with the path feedback module to monitor and adjust the logistics execution plan in real time. Based on feedback information such as transportation time, resource utilization and traffic conditions, it dynamically optimizes path planning and resource scheduling to ensure the efficiency and timeliness of the logistics process. The logistics path optimization unit is closely linked with the inventory optimization unit and the supplier management unit to form a positive feedback mechanism to ensure efficient coordination of logistics, inventory and suppliers and improve the operating efficiency of the entire supply chain.

[0080] Example 6: Please refer to Figure 6In one possible implementation, the system integration and feedback unit includes a global data management module, a feedback optimization module, an intelligent collaboration module, and a dynamic monitoring module. The system integration and feedback unit coordinates the various algorithm modules to form an effective feedback mechanism, and works with demand forecasting, inventory optimization, and other units to achieve global optimization. To ensure the overall optimization and dynamic adjustment of the supply chain system, it includes:

[0081] 1. Global data management module: responsible for obtaining and processing relevant data from each unit (demand forecasting and analysis unit, inventory optimization unit, supplier management unit, logistics path optimization unit). Input data includes demand forecast results, inventory levels, logistics execution data, supplier response data, etc. Output data includes standardized global data and is passed to the feedback optimization module and intelligent collaboration module.

[0082] This module cleans, formats and standardizes the input data to ensure that the data of each functional unit is in a unified format and has high accuracy. By using clustering algorithms, the data is initially analyzed to eliminate redundant information and improve data processing efficiency. The frequency of data collection is set to real-time, hourly or daily according to the actual needs of the supply chain to ensure the timeliness and effectiveness of data processing.

[0083] 2. Feedback optimization module: According to the data processed by the global data management module, combined with the inventory level, logistics status and the actual response of the supplier, the operating parameters of the demand forecasting and analysis unit, the inventory optimization unit and the logistics path optimization unit are adjusted; by analyzing the operating data of each unit, the operating parameters are dynamically adjusted to ensure the overall optimization of the supply chain system. The input data includes real-time feedback data such as inventory fluctuations, logistics delays, and supplier delivery status. The output data includes the optimized operating parameters, which are passed to the inventory optimization unit, logistics path optimization unit and supplier management unit.

[0084] This module combines the Bayesian optimization algorithm to dynamically adjust the model parameters of each unit based on feedback data. For example, when the logistics path optimization unit reports a transportation delay, the feedback optimization module dynamically adjusts the path planning and resource scheduling strategy based on the degree of delay. By adjusting the Bayesian Dynamic Linear Model (BDLM) parameters in the inventory optimization unit and the Adaptive Large Neighborhood Search Algorithm (ALNS) in the logistics path optimization unit, the system can respond quickly to market changes and optimize the overall efficiency of the supply chain.

[0085] 3. Intelligent collaboration module: By analyzing the data provided by the global data management module, it coordinates the operation of the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit and the logistics path optimization unit; the input data includes demand forecasting results, logistics execution status, supplier response data, etc. The output data includes the optimized global scheduling plan, which is transmitted to the inventory optimization unit, the logistics path optimization unit and the supplier management unit.

[0086] This module uses a multi-objective genetic algorithm for global optimization, taking into account multiple optimization objectives such as inventory holding costs, logistics transportation time, and supplier response speed. By dynamically adjusting parameters such as population size, crossover rate, and mutation rate, it ensures that each unit can be optimized synchronously according to global demand. The intelligent collaboration module also works with the convolutional neural network (CNN) of the inventory optimization unit to ensure a balance between long-term and short-term demand forecasts.

[0087] 4. Dynamic monitoring module: The dynamic monitoring module monitors the operation status of the supply chain in real time, identifies abnormal situations such as logistics delays, inventory imbalances, and supplier response delays, provides early warning signals, and prompts timely adjustments. Input data includes logistics delays, inventory imbalances, supplier response, etc. Output data includes early warning signals and optimization suggestions, and is passed to the feedback optimization module and the intelligent collaboration module. This module uses an adaptive neural network algorithm for trend forecasting, predicts potential bottlenecks of the system based on historical data and real-time data, and ensures the stability of supply chain operation through an early warning mechanism. The warning frequency can be set to minutes or hours according to demand to ensure that the system can respond quickly to abnormal situations.

[0088] Example 7: Please refer to Figure 7 The real-time logistics marketing unit provided in this embodiment includes a logistics data monitoring module, a market trend prediction module, a user behavior analysis module and a marketing feedback module.

[0089] 1. Logistics data monitoring module: responsible for real-time collection and monitoring of key data in the logistics link of the supply chain system. Input data includes inventory level, logistics path status, order completion status and customer feedback information. Through data cleaning and standardization processing, the accuracy and timeliness of the data are ensured.

[0090] The monitored logistics data include: Inventory level: reflects the current inventory status and its matching degree with market demand; Logistics route status: based on the feedback from the logistics route optimization unit, monitors the execution of each transportation route, including transportation time, delays, etc. Order completion: real-time tracking of order execution progress and delivery status. Customer feedback: collects customers' evaluation of logistics services and their satisfaction with orders.

[0091] The output data is standardized logistics status data, which is transmitted to the market trend prediction module and user behavior analysis module to ensure that the marketing strategy can be adjusted according to the real-time status of logistics execution.

[0092] 2. Market trend forecasting module: Based on historical sales data, market fluctuation data and customer behavior data, combined with the real-time logistics data provided by the logistics data monitoring module, it dynamically predicts future market demand trends.

[0093] Input data includes historical sales data, market fluctuation data, logistics status data, customer feedback data, etc.

[0094] The output data includes predicted market trends and demand changes, and is passed to the user behavior analysis module and the marketing feedback module for the formulation of dynamic marketing strategies. This module uses the long short-term memory network (LSTM) to analyze the short-term fluctuations and long-term trends of market demand, combined with changes in logistics status, to ensure that the market forecast results have high timeliness and accuracy. The time step setting of LSTM can be adjusted according to the cycle of market fluctuations, for example, set to 7 days, 30 days or 90 days to reflect seasonal fluctuations or long-term market changes.

[0095] 3. User behavior analysis module: Generates personalized marketing plans for different customer groups by analyzing customer order history, purchase behavior, and logistics feedback data. Input data includes market demand data, customer order history, and logistics feedback data output by the market trend prediction module. Output data includes personalized marketing suggestions, which are passed to the marketing feedback module and inventory optimization unit.

[0096] The user behavior analysis module uses a multi-layer perceptron (MLP) algorithm to generate customized marketing strategies for different customer groups, combining factors such as customer historical purchase preferences and logistics service quality feedback. The number of hidden layer units and learning rate of MLP are optimized according to the complexity and data volume of the customer group to ensure that the model can accurately capture customer behavior patterns and improve the relevance and effectiveness of marketing strategies.

[0097] 4. Marketing feedback module: The marketing feedback module is responsible for collecting feedback data on the execution of marketing activities, adjusting marketing strategies in real time and feeding them back to the market trend prediction module and user behavior analysis module. Input data includes marketing activity results, customer feedback on logistics services, order completion status, etc. Output data includes optimized marketing strategies and market forecast data, which are fed back to the market trend prediction module and user behavior analysis module. The feedback frequency is set to hourly or daily according to the execution of marketing activities to ensure that marketing strategies can be adjusted according to the latest market feedback. Through the feedback mechanism, the marketing feedback module transmits real-time customer data to the market trend prediction module to ensure that market forecasts and user behavior analysis can adjust strategies according to actual conditions.

[0098] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A digital management system for intelligent logistics supply chain based on data analysis, characterized in that: The system comprises: The demand forecasting and analysis unit analyzes historical orders and market fluctuation data through recursive neural networks and long short-term memory networks, predicts short-term and medium-term demand, and outputs demand forecast data; An inventory optimization unit, combining a Bayesian dynamic linear model and a convolutional neural network algorithm, adjusts the inventory level according to the demand forecast data output by the demand forecast and analysis unit; Supplier management unit: Based on the multi-layer perceptron and fuzzy clustering algorithm, it receives the inventory adjustment data of the inventory optimization unit and selects, classifies and sorts suppliers according to the supplier data; the unit also formulates supplier delivery plans and adjusts order allocation strategies based on the supplier's delivery capabilities and historical performance; The logistics path optimization unit adopts a genetic algorithm and an adaptive large-scale neighborhood search algorithm to perform logistics path planning based on the supplier selection results of the supplier management unit and real-time logistics data; The system integration and feedback unit receives data from the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit, and the logistics path optimization unit through a global data management and feedback mechanism, and integrates and synchronizes the operation of each unit; The real-time logistics marketing unit receives the feedback data from the system integration and feedback unit, and adjusts the logistics marketing strategy based on the logistics data and market trends.

2. According to the data analysis-based intelligent logistics supply chain digital management system of claim 1, it is characterized in that: The demand forecasting and analysis unit comprises: Short-term forecasting module: Generates short-term demand forecast results through recurrent neural networks and passes the forecast results to the inventory optimization unit to adjust inventory levels; Medium- and long-term forecasting module: using the long short-term memory network to generate medium- and long-term demand forecasts, and transmitting the output results of the long short-term memory network combined with the short-term forecast results of the recursive neural network algorithm to the inventory optimization unit and the logistics path optimization unit to formulate long-term inventory and logistics scheduling plans; Attention mechanism module: assigning different weights to the prediction data output by the long short-term memory network, and transmitting the weighted prediction data to the supplier management unit and the logistics path optimization unit; Feedback module: collects the operation data of the inventory optimization unit, the logistics path unit and the supplier management unit, and transmits the feedback information to the recursive neural network and the long short-term memory network to adjust the model parameters according to the actual operation.

3. According to the data analysis-based intelligent logistics supply chain digital management system of claim 2, it is characterized in that: The long short-term memory network in the medium- and long-term prediction module of the demand prediction and analysis unit includes an input gate, a forget gate and an output gate. The input gate is used to select current input information, the forget gate is used to adjust the weight of historical data to the current prediction, and the output gate is used to generate a prediction result.

4. The digital management system for intelligent logistics supply chain based on data analysis according to claim 1 is characterized in that: The inventory optimization unit comprises: Data collection module: used to obtain data from the demand forecasting and analysis unit, supplier management unit, logistics path optimization unit and inventory optimization unit. The data includes historical order data, real-time order volume, market fluctuation data, supplier delivery information and logistics status data. The data is standardized and then transferred to the inventory forecasting module and inventory data module to provide a basis for inventory forecasting and adjustment. Inventory forecasting module: Combining the Bayesian dynamic linear model and the convolutional neural network for inventory forecasting; the Bayesian dynamic linear model is used for short-term inventory forecasting, and the convolutional neural network is used to analyze long-term inventory trends; the forecast results are transmitted to the inventory adjustment module to provide a basis for short-term and long-term inventory regulation; Inventory adjustment module: dynamically adjusts inventory levels based on the forecast results of the inventory forecast module; Feedback processing module: collects operation feedback information from the supplier management unit and the logistics path optimization unit, and transmits the operation feedback information to the inventory forecasting module and the inventory adjustment module.

5. According to the data analysis-based intelligent logistics supply chain digital management system of claim 4, it is characterized in that: The convolutional neural network in the inventory optimization unit includes multiple layers of convolutional layers and pooling layers, wherein the convolutional layers are used to process local patterns in the inventory data, and the pooling layers are used to reduce the data dimension and extract the long-term trend characteristics of the inventory data.

6. The digital management system for intelligent logistics supply chain based on data analysis according to claim 1 is characterized in that: The supplier management unit includes: Supplier data acquisition module: used to obtain supplier-related data from various units in the supply chain, including supplier delivery time, order completion rate, inventory demand data and logistics status information; the acquired data is standardized and passed to the supplier analysis module and the supplier classification and sorting module; Supplier analysis module: Based on the multi-layer perceptron algorithm, it analyzes the supplier's historical delivery records, response speed and delivery stability, and generates supplier priorities; according to the analysis results, it adjusts the supplier delivery plan and passes the supplier priority ranking results to the supplier classification and sorting module; Supplier classification and sorting module: Use fuzzy clustering algorithm to classify and sort suppliers; determine supplier classification results based on supplier response speed, delivery stability and inventory requirements and pass them to the inventory optimization unit and logistics path optimization unit; Supplier feedback processing module: used to collect and process operational data from other units in the supply chain, including supplier delivery time, logistics execution status and inventory turnover rate; and pass the feedback data to the supplier analysis module.

7. The digital management system for intelligent logistics supply chain based on data analysis according to claim 1 is characterized in that: The logistics path optimization unit comprises: Logistics data acquisition module: used to acquire logistics-related data, including inventory status, demand forecast information, traffic conditions and transportation vehicle status information; the acquired data is processed and then transmitted to the path planning module and resource scheduling module; Path planning module: Generates the optimal path plan based on genetic algorithm and Dijkstra algorithm; the genetic algorithm is used to generate the initial path solution set, and the Dijkstra algorithm calculates the shortest path plan based on the initial solution and passes the plan to the resource scheduling module; Resource scheduling module: Based on the path plan generated by the path planning module, the adaptive large-scale neighborhood search algorithm is used to schedule logistics resources; Path feedback module: used to collect actual execution data of logistics paths, including transportation time, resource utilization and transportation delays; pass feedback data to the path planning module and resource scheduling module to adjust the path planning and resource scheduling plans.

8. The digital management system for intelligent logistics supply chain based on data analysis according to claim 7 is characterized in that: The initial path solution set of the genetic algorithm is set to 50 to 100.

9. The digital management system for intelligent logistics supply chain based on data analysis according to claim 1 is characterized in that: The system integration and feedback unit includes: Global data management module: used to obtain data from the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit and the logistics path optimization unit; standardize the obtained data and transmit it to the feedback optimization module and the intelligent collaboration module; Feedback optimization module: according to the data processed by the global data management module, combined with the inventory level, logistics status and the actual response of the supplier, adjust the operating parameters of the demand forecasting and analysis unit, the inventory optimization unit and the logistics path optimization unit; Intelligent collaboration module: coordinates the operation of the demand forecasting and analysis unit, the inventory optimization unit, the supplier management unit and the logistics path optimization unit by analyzing the data provided by the global data management module; Dynamic monitoring module: Identifies potential problems and provides early warning signals by monitoring logistics delays, inventory imbalances and supplier responses.

10. The intelligent logistics supply chain digital management system based on data analysis according to claim 1 is characterized in that: The real-time logistics marketing unit includes: Logistics data monitoring module: used to monitor key data in the logistics process, including inventory levels, logistics route status, order completion status and customer feedback information; the monitored data is transmitted to the market trend prediction module and user behavior analysis module; Market trend prediction module: predict future market demand trends based on historical sales data, market fluctuation data and logistics data; pass the prediction results to the user behavior analysis module and marketing feedback module; User behavior analysis module: Generates marketing plans for different customer groups by analyzing customers' order history, purchase behavior and logistics feedback data; transmits the generated marketing plans to the marketing feedback module and inventory optimization unit; Marketing feedback module: used to collect execution data of marketing activities and customer feedback on logistics services; pass feedback data to the market trend prediction module and user behavior analysis module to adjust market prediction and marketing strategies.

Citation Information

Patent Citations

  • RF order demand prediction method based on Bayesian optimization

    CN113807900A

  • Intelligent supply chain management system based on big data

    CN116629577A

  • Medical instrument logistics distribution method and system based on big data management

    CN117709829A

  • Catering supply chain safety management system and method based on block chain

    CN117993724A

Cited By

  • Intelligent supply chain scheduling optimization method and system based on autonomous controllable large model

    CN119228080A

  • Consignment sale system management system and method thereof

    CN119228273A

  • Intelligent supply chain and logistics optimization management method and system

    CN119250687A

  • Cloud baking raw material supply chain optimization management system

    CN119312991A

  • Business-based visualization method and system

    CN119379151A