Intelligent warehousing system based on supply chain management

Abstract

The invention discloses an intelligent warehousing system based on supply chain management, which relates to the technical field of warehousing management and comprises a cooperative work system consisting of a data acquisition module, a data processing module, an intelligent prediction module, a decision optimization module, an execution control module and a data sharing module. Storage environment, inventory state and logistics state data are acquired in real time through the data acquisition module, the data processing module cleans and analyzes the data to generate dynamic parameters, and the intelligent prediction module predicts inventory demand changes and supply chain abnormal conditions by using a big data algorithm. The decision optimization module generates an optimal strategy of inventory adjustment and resource allocation, the execution control module accurately controls the warehousing equipment to execute operation, and the data sharing module realizes upstream and downstream real-time data synchronization, so that the dynamic response capability and supply chain collaboration efficiency of the warehousing system are remarkably improved finally, the inventory excess and shortage risk is effectively reduced, and the system performance is improved. And the service capability and the profit level of the enterprise are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to an intelligent warehouse system based on supply chain management. Background Art

[0002] With the continuous development of the field of supply chain management, modern warehousing systems have become a key link in the supply chain. Traditional warehousing management methods usually rely on manual operations and independent equipment control, which is difficult to adapt to multi-category, high-frequency, and dynamic storage needs. At the same time, data islands are common, and it is difficult to achieve real-time sharing and collaboration between inventory information and other links in the supply chain such as orders and logistics. In recent years, intelligent warehousing systems based on artificial intelligence, big data, and the Internet of Things have gradually become a research hotspot, improving warehousing efficiency, reducing costs, and optimizing inventory management through automation, digitization, and intelligence.

[0003] The prior art has the following deficiencies: The current intelligent warehousing system based on supply chain management lacks real-time decision-making and adaptive capabilities when dealing with dynamic demand fluctuations and supply chain anomalies. When sudden anomalies occur in the supply chain (such as logistics delays, insufficient supply from suppliers, or sudden changes in order demand), the warehousing system is usually unable to effectively predict and quickly adjust inventory scheduling strategies. This defect may lead to aggravated overstocking or shortages, and even cause an imbalance in the overall operation of the supply chain, affecting the company's service capabilities and profitability. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent warehousing system based on supply chain management to solve the shortcomings of the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent warehousing system based on supply chain management, comprising a data acquisition module, a data processing module, an intelligent prediction module, a decision optimization module, an execution control module and a data sharing module; Data collection module, used to collect real-time data on storage environment, inventory status and logistics status through IoT devices; A data processing module, connected to the data acquisition module, for classifying, cleaning and analyzing the collected real-time data and generating dynamic parameters related to inventory management; An intelligent prediction module, connected to the data processing module, for predicting inventory demand changes and supply chain anomalies based on big data algorithms and machine learning models; A decision optimization module, connected to the intelligent prediction module, for generating an optimal strategy for inventory adjustment and resource allocation according to the prediction results; An execution control module, connected to the decision optimization module, for controlling the storage equipment to execute corresponding inventory operations according to the generated strategy; The data sharing module is used to achieve real-time synchronization of inventory data between upstream and downstream of the supply chain.

[0006] Preferably, in the intelligent prediction module, after analyzing the inventory demand change trend in the future period of time by a time series analysis algorithm, the inventory demand fluctuation value is calculated, and the calculation method of the inventory demand fluctuation value is: Collect historical inventory demand data to form a time series , perform difference processing on the data: ; In the formula, is the time series after difference; Fitting an ARIMA model to capture the trend and volatility of the time series is expressed as: ; In the formula, c is a constant term, is the autoregression coefficient, is the moving average coefficient, , is white noise; the residual of the ARIMA model Used to measure the difference between actual demand and predicted demand, and calculate the predicted value through the fitted ARIMA model , the expression is: ; Calculate residuals , the expression is: ; Calculate the inventory demand fluctuation value, the expression is: ; In the formula, HLK is the inventory demand fluctuation value, is the residual mean, and n is the length of the time series.

[0007] Preferably, in the intelligent prediction module, the logistics delay situation is analyzed and then the logistics delay abnormal value is calculated. The calculation method of the logistics delay abnormal value is: Collecting logistics delay data , sort the delay data from small to large to facilitate the calculation of quantiles, calculate the first quartile Q1, Q1 is the 25% position point in the data set, calculate the third quartile Q3, Q3 is the 75% position point in the data set, calculate the interquartile range IQR, the interquartile range is the difference between Q3 and Q1: ; Calculate the outlier range, where the lower bound for: ; Upper bound for: , collect the delay data that is smaller than the next term and larger than the previous term, establish a data set, and calculate the standard deviation of the data set as the logistics delay outlier.

[0008] Preferably, in the intelligent prediction module, the inventory demand fluctuation values ​​and logistics delay anomalies are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as inputs of the machine learning model. The machine learning model predicts the comprehensive risk value labels of inventory demand and logistics delays for each group of comprehensive feature vectors as the prediction target, and minimizes the sum of prediction errors for all comprehensive risk value labels of inventory demand and logistics delays as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The comprehensive risk value of inventory demand and logistics delay is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0009] Preferably, the obtained comprehensive risk value of inventory demand and logistics delay is compared with a gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the security analysis value of the encryption system is compared with the first standard threshold and the second standard threshold respectively; If the combined risk value of inventory demand and logistics delay is greater than the second standard threshold, it means that the combined risk of inventory demand and logistics delay is high. At this time, a first-level warning signal is generated and it is marked as a high risk level. If the security analysis value of the encryption system is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the comprehensive risk of inventory demand and logistics delay is medium. At this time, a secondary warning signal is generated and it is marked as a medium risk level. If the security analysis value of the encryption system is less than the first standard threshold, it means that the combined risk of inventory demand and logistics delay is low. At this time, a third-level warning signal is generated and marked as a low risk level.

[0010] Preferably, in the decision optimization module, the optimal strategy for inventory adjustment and resource allocation is generated according to the forecast results, specifically: Cost minimization includes transportation costs, inventory holding costs, and out-of-stock costs. The objective function is expressed as: ; In the formula, is the unit transportation cost from point i to point j, is the amount of cargo transported from point i to point j, is the inventory holding cost, is the inventory quantity, is the out-of-stock cost, the unit loss caused by the failure of product k to meet the demand, is the out-of-stock quantity; Maximizing efficiency means maximizing the total amount of goods that meet demand or minimizing the transportation time: ; is the total quantity of goods that demand point j meets, and the inventory constraint is: ; The total amount of goods shipped out of each storage point shall not exceed its inventory , demand constraints: ; The demand of each demand point j Should be satisfied; Path constraint: The replenishment path should not be repeated and must comply with the path planning rules.

[0011] Preferably, based on the solution of the genetic algorithm, an initial solution is randomly generated to form an initial population; each individual represents an inventory adjustment plan; and the fitness of each individual is calculated according to the objective functions C and E: ; is the weight of the efficiency target; select excellent individuals according to the fitness value to enter the next generation, partially exchange the selected individuals, generate new individuals, and randomly adjust the individual characteristics; stop when the fitness value converges or reaches the maximum number of iterations; the genetic algorithm finally outputs the individual with the highest fitness value as the optimal inventory adjustment plan, including: cargo distribution plan: the amount of cargo allocated from each inventory point to the demand point; replenishment path: the shortest transportation path or the lowest cost path; priority: generate a priority processing order based on the urgency of the demand.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention solves the problem that the existing technology lacks real-time decision-making and adaptive capabilities in dealing with dynamic demand fluctuations and supply chain anomalies by constructing an intelligent warehousing system based on supply chain management. The system obtains the warehousing environment, inventory status and logistics status data in real time through the data acquisition module, generates dynamic parameters after cleaning, classification and analysis by the data processing module, and uses the intelligent prediction module to accurately predict inventory demand changes and logistics anomalies based on big data algorithms and machine learning models, and generates inventory adjustment and resource allocation strategies that minimize costs and maximize efficiency through the decision optimization module. Finally, the execution control module directs the warehousing equipment to complete specific operations. At the same time, the data sharing module realizes real-time synchronization of inventory data between upstream and downstream of the supply chain, effectively improving the system's collaborative ability and response efficiency.

[0013] 2. The present invention calculates inventory demand fluctuation values ​​and logistics delay anomalies through an intelligent prediction module, and predicts comprehensive risk values ​​based on comprehensive feature vectors. Combined with multi-level early warning signals, the system can identify and respond to high-risk situations in advance, reducing the risk of excess or shortage inventory; using genetic algorithms to optimize inventory allocation plans, replenishment paths and priorities, significantly improving resource allocation efficiency and transportation efficiency, and reducing operating costs; through modular design and data sharing mechanisms, the system achieves deep collaboration and transparent management of upstream and downstream supply chains, providing an efficient solution for the intelligent transformation of supply chain management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] For examples, see Figure 1 As shown, the intelligent warehousing system based on supply chain management described in this embodiment includes a data acquisition module, a data processing module, an intelligent prediction module, a decision optimization module, an execution control module and a data sharing module; Data collection module, used to collect real-time data on storage environment, inventory status and logistics status through IoT devices; A data processing module, connected to the data acquisition module, for classifying, cleaning and analyzing the collected real-time data and generating dynamic parameters related to inventory management; An intelligent prediction module, connected to the data processing module, for predicting inventory demand changes and supply chain anomalies based on big data algorithms and machine learning models; A decision optimization module, connected to the intelligent prediction module, for generating an optimal strategy for inventory adjustment and resource allocation according to the prediction results; An execution control module, connected to the decision optimization module, for controlling the storage equipment to execute corresponding inventory operations according to the generated strategy; The data sharing module is used to achieve real-time synchronization of inventory data between upstream and downstream of the supply chain.

[0018] In the data acquisition module, the temperature and humidity sensors installed in the storage environment are used to collect temperature and humidity information in the warehouse in real time to ensure that the storage conditions meet the requirements of specific goods, such as the low temperature environment required for cold chain storage of goods. The unique identification, location, quantity and other information of the goods are collected in real time through the radio frequency identification (RFID) tags and barcode readers on the goods. The RFID reader regularly scans the goods labels through wireless signals to obtain storage location and status data. Logistics trackers (such as GPS positioning devices) are used to record the geographical location, transportation route and arrival time of the goods in transit to ensure that the logistics process data can be synchronized with the warehousing system.

[0019] Various sensors and monitoring devices are connected to the data acquisition module via Wi-Fi or 5G networks. After the devices go online, they automatically register their identities and join the system's IoT network.

[0020] The data collection module obtains data from various sensors and devices at a preset frequency (for example, every minute). Environmental monitoring equipment regularly reads temperature and humidity values, RFID readers regularly scan the label information of goods, and logistics trackers continuously upload the location information and status data of goods.

[0021] The data acquisition module's built-in data cache mechanism initially stores the collected raw data and removes noise and redundant information through a data filtering algorithm. For example, when the RFID reader repeatedly scans the signal of the same tag, only the latest record is retained.

[0022] The data acquisition module transmits the cleaned data to the data processing module through a secure communication protocol (such as MQTT or HTTPs) for subsequent classification analysis and optimization processing.

[0023] When the temperature and humidity sensor detects that the storage environment is out of the preset range (such as the temperature is higher than 25°C), the system will send an alarm to the management staff in real time and trigger the refrigeration equipment to adjust the environment. When the RFID reader scans the movement of goods or changes in inventory, the inventory information in the system is updated immediately. For example, when the goods are moved to the shipping area, the inventory quantity is automatically reduced to avoid delays or errors caused by manual records. When the logistics tracker detects that the goods have arrived at the warehouse, the system automatically marks the batch of goods as "arrived" and triggers the warehousing operation process.

[0024] The data processing module is connected to the data acquisition module and is used to classify, clean and analyze the collected real-time data to generate dynamic parameters related to inventory management. The core components of the data processing module include: Data classification unit: divides the data into storage environment data, inventory status data and logistics status data according to the data source and content type. Data cleaning unit: formats the classified data and removes abnormal or redundant information. Data analysis unit: generates dynamic parameters such as inventory turnover rate, abnormal probability of environmental indicators, inventory fluctuation trend, etc. through real-time analysis. Data storage unit: stores the cleaned and analyzed data and dynamic parameters in the database for use by subsequent modules.

[0025] The data classification unit receives the real-time data stream from the data acquisition module and classifies and processes the data according to its source and attributes: Warehouse environment data: such as the ambient temperature and humidity values ​​provided by the temperature and humidity sensor. Inventory status data: such as the unique identification, location, quantity and other information of the goods obtained by the RFID reader. Logistics status data: such as the real-time location and arrival time of the goods provided by the GPS device. The classification results are marked in the form of tags, such as "ENV" (environmental data), "STOCK" (inventory data), and "LOG" (logistics data).

[0026] The data cleaning unit processes the classified data to ensure the accuracy and consistency of the data: abnormal data is eliminated by preset thresholds. For example, when the value collected by the temperature sensor deviates from the normal range (such as below -50°C or above 80°C), it is marked as abnormal and eliminated. For repeated data collected continuously (such as reading the RFID tag of the same goods multiple times in a short period of time), the latest valid record is retained. The data formats of different data sources are unified into the system standard format. For example, timestamps are uniformly converted to ISO 8601 format.

[0027] The data analysis unit performs real-time calculations and dynamic parameter generation on the cleaned data: based on the inventory inbound and outbound data, it calculates the cargo turnover rate over a period of time, using the formula: turnover rate = total outbound volume / average inventory; it dynamically displays cargo circulation efficiency, providing a basis for optimizing inventory decisions.

[0028] By statistically analyzing the changing trends of historical environmental data, the probability of future environmental anomalies can be analyzed. For example, the temperature and humidity change rate can be used to predict whether the safe storage range may be exceeded. Inventory fluctuation trend analysis: Time series analysis algorithms (such as the ARIMA model) are used to predict the short-term fluctuation trend of inventory and provide a reference for replenishment strategies.

[0029] The data processing module stores the cleaned raw data and the dynamic parameters generated by analysis in relational databases (such as MySQL) and distributed databases (such as MongoDB) respectively: the cleaned data is used for subsequent query and verification. The dynamic parameters are transmitted to the decision optimization module in real time through the API interface to support the optimization calculation of inventory scheduling and resource allocation.

[0030] An intelligent prediction module, connected to the data processing module, for predicting inventory demand changes and supply chain anomalies based on big data algorithms and machine learning models; After analyzing the inventory demand change trend in the future through the time series analysis algorithm, the inventory demand fluctuation value is calculated. The calculation method of the inventory demand fluctuation value is: Collect historical inventory demand data to form a time series , make sure the data are recorded at uniform time intervals (such as daily, weekly or monthly). Check the stationarity of the data (such as whether the mean and variance remain stable over time). Use unit root tests (such as ADF test) to determine whether difference processing is needed. If the data is non-stationary, make it stationary through first-order or multiple-order differences: ; In the formula, is the time series after difference.

[0031] The ARIMA model consists of three core parameters: p is the order of the autoregressive (AR) term, which is the relationship between the current value and its previous p values. d is the number of differencing, which is the number of differencing orders required to make the data stationary. q is the order of the moving average (MA) term, which is the relationship between the current value and its previous q error terms. Use the ACF (Autocorrelation Function) and PACF (Partial Autocorrelation Function) plots to determine p and q. d is determined by the number of differencing times.

[0032] Fitting an ARIMA model to capture the trend and volatility of the time series is expressed as: ; In the formula, c is the constant term (intercept), is the autoregression coefficient, is the moving average coefficient, , is white noise (error term). Use maximum likelihood estimation to fit the model and determine the parameters.

[0033] Residuals from the ARIMA model Used to measure the difference between actual demand and predicted demand, and calculate the predicted value through the fitted ARIMA model , the expression is: ; Calculate residuals , the expression is: ; Calculate the inventory demand fluctuation value, the expression is: ; In the formula, HLK is the inventory demand fluctuation value, is the residual mean, and n is the length of the time series.

[0034] When the inventory demand fluctuation value is large, it means that the inventory demand changes more dramatically and the fluctuation range is obvious. Such a situation usually indicates that the inventory demand is frequently affected by external factors, such as seasonal sales peaks, sudden surges in market demand, unstable supply chains or emergencies. This abnormal situation will place higher demands on inventory management, requiring the system to have a rapid response capability to avoid out-of-stock due to a surge in demand or inventory backlogs due to a sudden drop in demand. In addition, a large fluctuation value may indicate that the demand forecasting model needs to be optimized and its ability to adapt to dynamic demand changes needs to be further improved.

[0035] On the contrary, when the inventory demand fluctuation value is small, it means that the inventory demand changes tend to be stable and the fluctuation range is small. This usually means that the market demand is relatively stable, or the warehousing and supply chain management have achieved a high degree of coordination. In this case, the company can better optimize the inventory structure and reduce the operating costs caused by excessive fluctuations. However, too small a demand fluctuation value may also cover up potential market changes or anomalies, such as hidden demand not being captured. Therefore, even if the fluctuation value is small, companies still need to continue to monitor demand trends, ensure sensitivity to market changes, and respond to potential anomalies in a timely manner.

[0036] After analyzing the logistics delay situation, the logistics delay abnormal value is calculated. The calculation method of the logistics delay abnormal value is: Collecting logistics delay data The unit can be minutes, hours or days, depending on the timeliness of logistics. The data is a non-negative number, such as: {2,3,4,10,12,50,60} (unit: hour).

[0037] Sort the delay data from small to large to facilitate the calculation of quantiles: Lsorted = [2,3,4,10,12,50,60]. Calculate the first quartile Q1, Q1 is the 25% position point in the data set, calculate the third quartile Q3, Q3 is the 75% position point in the data set, and calculate the interquartile range IQR, the interquartile range is the difference between Q3 and Q1: ; Calculate the outlier range, where the lower bound for: ; Upper bound for: ,The delay data that is smaller than the next term and larger than the previous term are collected, and a data set is established. The standard deviation of the data set is calculated as the logistics delay outlier.

[0038] The larger the logistics delay outlier value, the more serious the deviation of the logistics delay from the normal range, and the higher the degree of abnormality. This may indicate that there are significant problems in the logistics process, such as major obstacles in the transportation process (such as traffic accidents or bad weather), supply chain bottlenecks (such as warehouse management errors or improper resource allocation), or even systemic problems (such as logistics network overload or interruption). Large delay outliers often have a greater impact on the company's operational efficiency and customer satisfaction, and may lead to chain reactions such as delayed delivery, order cancellations, or damaged brand reputation.

[0039] On the contrary, when the logistics delay outlier value is smaller, it means that the logistics delay is close to the normal range and the degree of abnormality is relatively mild. Small delay outliers may be caused by occasional non-systematic factors, such as temporary traffic jams or minor scheduling errors. Such anomalies usually have limited impact on the overall logistics operation and are easy to adjust through local optimization or rapid intervention. However, if small outliers appear frequently, it may reflect potential process management or resource allocation problems, which need to be combined with trend analysis for in-depth investigation to avoid small problems accumulating into big problems.

[0040] The inventory demand fluctuation values ​​and logistics delay anomalies are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model takes each group of comprehensive feature vectors to predict the comprehensive risk value label of inventory demand and logistics delay as the prediction target, and takes minimizing the sum of prediction errors of all comprehensive risk value labels of inventory demand and logistics delay as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The comprehensive risk value of inventory demand and logistics delay is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0041] The method for obtaining the comprehensive risk value of inventory demand and logistics delay is to obtain the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: ; In the formula, is the output function of the model, HLK is the inventory demand fluctuation value, GLH is the logistics delay abnormal value, It is the comprehensive risk value of inventory demand and logistics delay.

[0042] The obtained comprehensive risk value of inventory demand and logistics delay is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the security analysis value of the encryption system is compared with the first standard threshold and the second standard threshold respectively; If the combined risk value of inventory demand and logistics delay is greater than the second standard threshold, it means that the combined risk of inventory demand and logistics delay is high. At this time, a first-level warning signal is generated and it is marked as a high-risk level. Relevant personnel need to intervene immediately and take emergency measures, such as replenishing stock, accelerating transportation, or adjusting supply chain nodes.

[0043] If the security analysis value of the encryption system is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the comprehensive risk of inventory demand and logistics delay is medium. At this time, a secondary warning signal is generated, which is marked as a medium risk level and adjusted according to the specific situation.

[0044] If the security analysis value of the encryption system is less than the first standard threshold, it means that the combined risk of inventory demand and logistics delay is low. At this time, a third-level warning signal is generated, marking it as a low-risk level. Normal daily monitoring is carried out to keep the system running.

[0045] It should be noted here that the importance of the first-level warning signal is greater than that of the second-level warning signal, and the importance of the second-level warning signal is greater than that of the third-level warning signal. Relevant personnel can take corresponding handling measures according to different warning signal levels.

[0046] The decision optimization module is connected to the intelligent prediction module and is used to generate the optimal strategy for inventory adjustment and resource allocation according to the prediction results.

[0047] Construct an inventory scheduling optimization model with the goal of minimizing costs and maximizing efficiency. Based on genetic algorithms, calculate inventory adjustment plans, including goods allocation, replenishment paths and priorities, and output the optimal decision plan for execution by the execution control module.

[0048] Cost minimization includes transportation costs, inventory holding costs, and stock-out costs.

[0049] The objective function is expressed as: ; In the formula, is the unit transportation cost from point i to point j, is the amount of cargo transported from point i to point j, is the inventory holding cost (the holding cost per unit of inventory), is the inventory quantity, is the out-of-stock cost, the unit loss caused by the failure of product k to meet the demand, Out of stock quantity.

[0050] Maximizing efficiency means maximizing the total amount of goods that meet demand or minimizing the transportation time: ; is the total quantity of goods that demand point j meets, and the inventory constraint is: ; The total amount of goods shipped out of each storage point shall not exceed its inventory . Requirements constraints: ; The demand of each demand point j It should be satisfied as much as possible. Path constraint: The replenishment path should not be repeated and must comply with the path planning rules.

[0051] The solution is based on genetic algorithm, which is an optimization method based on biological evolution, which continuously improves the solution through selection, crossover and mutation.

[0052] Randomly generate initial solutions (cargo allocation plans, replenishment paths, etc.) to form the initial population.

[0053] Each entity represents a possible inventory adjustment scenario.

[0054] Calculate the fitness of each individual according to the objective functions C and E: ; is the weight of the efficiency target.

[0055] Excellent individuals are selected to enter the next generation based on their fitness values, and roulette or tournament selection strategies are commonly used.

[0056] Perform partial exchange of selected individuals (such as path exchange or cargo distribution adjustment) to generate new individuals.

[0057] Randomly adjust certain characteristics of individuals (such as changing replenishment routes or redistributing goods).

[0058] Stop when the fitness value converges or the maximum number of iterations is reached.

[0059] The genetic algorithm finally outputs the individual with the highest fitness value as the optimal inventory adjustment plan, including: Goods allocation plan: the amount of goods allocated from each inventory point to the demand point. Replenishment path: the shortest transportation path or the lowest cost path. Priority: Generate a priority processing order based on the urgency of demand.

[0060] The execution control module is connected to the decision optimization module and is used to control the storage equipment to perform corresponding inventory operations according to the generated strategy. The output data of the decision optimization module is received, including: Cargo allocation plan: the quantity and destination of goods to be allocated to each inventory point. Replenishment path: the optimal path for transporting goods. Priority plan: the urgency and execution order of each task. Parse the data and convert it into an executable instruction format, for example: Sorting instruction: allocate goods to a specific loading area. Handling instruction: control the automatic guided vehicle (AGV) to carry goods to the loading area. Loading and unloading instruction: load goods into the transport vehicle through the robotic arm.

[0061] According to the parsed task information, specific tasks are assigned to various types of storage equipment: Automatic sorting machine: classifies the goods according to demand and assigns them to the corresponding transportation loading area. Example instruction: Sorting task = {goods ID: G001, destination: loading area 1} destination: loading area 1}; Automatic guided vehicle (AGV): performs cargo handling tasks, such as moving the sorted goods from the storage area to the loading area. Example instruction: Handling task = {goods ID: G001, starting point: storage area A, end point: loading area 1}.

[0062] Move the goods from the loading area to the transport vehicle. Example command: Loading task = {Goods ID: G001, Starting point: Loading area 1, Destination: Transport vehicle 1}. Environment equipment: If the storage environment (such as temperature and humidity) needs to be adjusted, send commands to control the operation of related equipment.

[0063] Monitor the operating status of each device (such as sorting progress, AGV position, robot arm movement, etc.) through IoT sensors. Receive real-time feedback from the device on task execution and update the task completion status. For example: Success feedback: After completing the instruction, the device returns a successful status. Failure feedback: If a failure occurs, the module re-plans the task or sends a maintenance request.

[0064] Based on real-time feedback data, dynamically adjust unfinished tasks or abnormal situations: If an AGV fails (such as a blocked path): reallocate the task to a backup AGV. Adjust path planning to avoid blocked areas. If a loading area is overloaded with goods: increase the AGV frequency and move the goods to the transport area in a timely manner.

[0065] The data sharing module is used to achieve real-time synchronization of inventory data between upstream and downstream of the supply chain.

[0066] Inventory data collection: Data sources include: Inventory status recorded in real time by the warehouse management system (WMS). Dynamic inventory parameters (such as inventory turnover rate, inventory fluctuation value) provided by the data processing module. Goods location information and environmental status provided by IoT devices. Data standardization processing: To ensure data compatibility, the collected inventory data is uniformly converted into a standardized format (such as JSON or XML). Data storage: Distributed databases (such as MongoDB or Cassandra) are used to store inventory data to ensure data reading efficiency under high concurrency. Real-time distribution of inventory data based on message queue technology (such as Kafka, RabbitMQ): The data collection module pushes inventory update events to the message queue. The data sharing module subscribes to the update events of the message queue and sends them to the upstream and downstream nodes of the supply chain in real time.

[0067] Synchronization mode: Push mode: The data sharing module actively pushes updated data to the subscriber (supply chain node). Pull mode: The node actively requests the latest inventory data according to demand. Data encryption: Use TLS protocol to ensure the security of data transmission. Encrypt sensitive data (such as product batches, shipping prices) for storage and transmission. Permission management: Use role-based access control (RBAC) mechanism to ensure that different nodes can only access data within the authorized scope. Example: Suppliers can only view raw material inventory, and retailers can only view end product inventory. Data verification: During data transmission, use hash verification (such as MD5) to verify data integrity to avoid data loss or tampering.

[0068] When inventory changes (such as inbound, outbound, transfer, etc.), the data sharing module pushes the updated data to all subscribed nodes in real time. The historical changes of inventory data are recorded by timestamps, supporting upstream and downstream nodes to query and audit historical data.

[0069] Specific application scenarios of data sharing include: Supplier side: Shared data: Raw material inventory status. Application scenario: When the inventory is lower than the safety stock threshold, the supplier can receive replenishment notifications in real time and prepare stocks in advance.

[0070] Manufacturer side: Shared data: semi-finished product inventory and production progress. Application scenario: Manufacturers dynamically adjust production plans based on inventory changes to avoid excessive backlogs or production interruptions.

[0071] Distributor side: Shared data: terminal product inventory and delivery status. Application scenario: Distributors optimize delivery plans based on inventory distribution to reduce transportation costs.

[0072] Retailer side, shared data: product inventory and sales data. Application scenario: Retailers can view inventory status in real time and adjust promotional activities to avoid out-of-stock or slow-selling products.

[0073] In this embodiment, it includes a data acquisition module, a data processing module, an intelligent prediction module, a decision optimization module, an execution control module and a data sharing module, and each module operates in coordination to achieve closed-loop optimization of intelligent warehousing and supply chain management. The data acquisition module obtains data on the warehousing environment, inventory status and logistics status in real time through the Internet of Things devices; the data processing module classifies, cleans and analyzes the collected data to generate dynamic parameters; the intelligent prediction module predicts changes in inventory demand and supply chain anomalies based on big data algorithms and machine learning models; the decision optimization module generates the optimal strategy for inventory adjustment and resource allocation based on the prediction results; the execution control module directs the warehousing equipment to complete specific operations according to the optimization strategy; the data sharing module ensures that inventory data is synchronized in real time in the upstream and downstream of the supply chain, improving the collaborative efficiency and responsiveness of the supply chain.

[0074] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0075] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0076] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent warehousing system based on supply chain management, characterized by: It includes data acquisition module, data processing module, intelligent prediction module, decision optimization module, execution control module and data sharing module; Data collection module, used to collect real-time data on storage environment, inventory status and logistics status through IoT devices; A data processing module, connected to the data acquisition module, for classifying, cleaning and analyzing the collected real-time data and generating dynamic parameters related to inventory management; An intelligent prediction module, connected to the data processing module, for predicting inventory demand changes and supply chain anomalies based on big data algorithms and machine learning models; A decision optimization module, connected to the intelligent prediction module, for generating an optimal strategy for inventory adjustment and resource allocation according to the prediction results; An execution control module, connected to the decision optimization module, for controlling the storage equipment to execute corresponding inventory operations according to the generated strategy; The data sharing module is used to achieve real-time synchronization of inventory data between upstream and downstream of the supply chain.

2. The intelligent warehousing system based on supply chain management according to claim 1, characterized in that: In the intelligent forecasting module, the inventory demand change trend in the future is analyzed by the time series analysis algorithm, and the inventory demand fluctuation value is calculated. The calculation method of the inventory demand fluctuation value is: Collect historical inventory demand data to form a time series , perform difference processing on the data: ; In the formula, is the time series after difference; Fitting an ARIMA model to capture the trend and volatility of the time series is expressed as: ; In the formula, c is a constant term, is the autoregression coefficient, is the moving average coefficient, , is white noise; the residual of the ARIMA model Used to measure the difference between actual demand and predicted demand, and calculate the predicted value through the fitted ARIMA model , the expression is: ; Calculate residuals , the expression is: ; Calculate the inventory demand fluctuation value, the expression is: ; In the formula, HLK is the inventory demand fluctuation value, is the residual mean, and n is the length of the time series.

3. The intelligent warehousing system based on supply chain management according to claim 2 is characterized by: In the intelligent prediction module, the logistics delay situation is analyzed and the logistics delay abnormal value is calculated. The calculation method of the logistics delay abnormal value is: Collecting logistics delay data , sort the delay data from small to large to facilitate the calculation of quantiles, calculate the first quartile Q1, Q1 is the 25% position point in the data set, calculate the third quartile Q3, Q3 is the 75% position point in the data set, calculate the interquartile range IQR, the interquartile range is the difference between Q3 and Q1: ; Calculate the outlier range, where the lower bound for: ; Upper bound for: , collect the delay data that is smaller than the next term and larger than the previous term, establish a data set, and calculate the standard deviation of the data set as the logistics delay outlier.

4. The intelligent warehousing system based on supply chain management according to claim 3 is characterized by: In the intelligent prediction module, the inventory demand fluctuation value and the logistics delay anomaly value are converted into a comprehensive feature vector, and the comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the comprehensive risk value label of inventory demand and logistics delay as the prediction target, and minimizes the sum of prediction errors of all comprehensive risk value labels of inventory demand and logistics delay as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The comprehensive risk value of inventory demand and logistics delay is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. The intelligent warehousing system based on supply chain management according to claim 4 is characterized by: The obtained comprehensive risk value of inventory demand and logistics delay is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the security analysis value of the encryption system is compared with the first standard threshold and the second standard threshold respectively; If the combined risk value of inventory demand and logistics delay is greater than the second standard threshold, it means that the combined risk of inventory demand and logistics delay is high. At this time, a first-level warning signal is generated and it is marked as a high risk level. If the security analysis value of the encryption system is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the comprehensive risk of inventory demand and logistics delay is medium. At this time, a secondary warning signal is generated and it is marked as a medium risk level. If the security analysis value of the encryption system is less than the first standard threshold, it means that the combined risk of inventory demand and logistics delay is low. At this time, a third-level warning signal is generated and marked as a low risk level.

6. The intelligent warehousing system based on supply chain management according to claim 1, characterized in that: In the decision optimization module, the optimal strategy for inventory adjustment and resource allocation is generated based on the forecast results, specifically: Cost minimization includes transportation costs, inventory holding costs, and out-of-stock costs. The objective function is expressed as: ; In the formula, is the unit transportation cost from point i to point j, is the amount of cargo transported from point i to point j, is the inventory holding cost, is the inventory, is the out-of-stock cost, the unit loss caused by the failure of product k to meet the demand, is the out-of-stock quantity; Maximizing efficiency means maximizing the total amount of goods that meet demand or minimizing the transportation time: ; is the total quantity of goods that demand point j meets, and the inventory constraint is: ; The total amount of goods shipped out of each storage point shall not exceed its inventory , demand constraints: ; The demand of each demand point j Should be satisfied; Path constraint: The replenishment path should not be repeated and must comply with the path planning rules.

7. The intelligent warehousing system based on supply chain management is characterized by: Based on the solution of genetic algorithm, the initial solution is randomly generated to form the initial population; each individual represents an inventory adjustment plan; the fitness of each individual is calculated according to the objective functions C and E: ; is the weight of the efficiency target; select excellent individuals to enter the next generation according to the fitness value, partially exchange the selected individuals, generate new individuals, and randomly adjust the individual characteristics; Stop when the fitness value converges or reaches the maximum number of iterations; The genetic algorithm ultimately outputs the individual with the highest fitness value as the optimal inventory adjustment plan, including: cargo allocation plan: the amount of cargo allocated from each inventory point to the demand point; replenishment path: the shortest transportation path or the lowest cost path; priority: generating a priority processing order based on the urgency of demand.

Images