Digital AI intelligent analysis system based on big data
Through the digital AI intelligent analysis system of big data, time series prediction and clustering algorithms and mixed integer planning technology are used to integrate enterprise data, solving the problems of data isolation and low optimization efficiency in traditional systems, and achieving the accuracy of inventory and customer relationships Efficient optimization of management and supply chain.
Patent Information
- Application Number
- CN202411217615.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-02
AI Technical Summary
Traditional enterprise intelligent analysis systems lead to data isolation and inability to be effectively integrated, affecting decision-making efficiency, and there are insufficient accurate prediction and analysis in inventory management and customer relationship management, resulting in increased costs and reduced customer satisfaction and inefficient supply chain optimization.
Digital AI intelligent analysis system based on big data, including data acquisition module, inventory optimization module, customer relationship management module and supply chain optimization module, integrates enterprise operations and customers through time series prediction algorithms, clustering algorithms and mixed integer linear planning technology. and supply chain data, build inventory and supply chain optimization models, formulate optimal inventory replenishment strategies and personalized marketing strategies, and optimize resource allocation and transportation paths.
It has achieved comprehensive integration of data, accurately predict sales demand, reduce inventory backlog and out of stock, improve customer satisfaction and supply chain efficiency, generate comprehensive business reports, and support data-driven decision-making.
Smart Images

Figure CN118735415B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a digital AI intelligent analysis system based on big data. Background Art
[0002] With the rapid development of technologies such as big data, artificial intelligence, and the Internet of Things, enterprises have accumulated a large amount of data in operations, customer relationship management, and supply chain management. Traditional data processing and analysis methods can no longer meet the needs of modern enterprises for efficient decision-making and refined management. Through data-driven intelligent analysis systems, enterprises can extract valuable information from massive data to optimize operational efficiency, improve customer satisfaction, and reduce supply chain management costs. This digital AI intelligent analysis system based on big data is becoming increasingly important in various industries, especially in manufacturing, retail, and logistics.
[0003] However, in traditional enterprise intelligent analysis systems, data between various departments of the enterprise are isolated, resulting in ineffective information integration, affecting overall decision-making efficiency. In inventory management, the inability to accurately predict product sales demand may lead to inventory backlogs or out-of-stocks, thereby increasing enterprise costs and operational risks. In addition, traditional customer relationship management methods lack in-depth analysis of customer behavior patterns and demand trends, resulting in inaccurate marketing strategies, affecting customer satisfaction and loyalty. In the supply chain, the optimization of resource allocation and transportation routes is crucial to reducing costs and improving efficiency, but the complex supply chain structure makes traditional optimization methods inefficient. Summary of the invention
[0004] In view of the deficiencies of the prior art, the present invention provides a digital AI intelligent analysis system based on big data, which solves the problems of the above-mentioned background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a digital AI intelligent analysis system based on big data, including the following modules: a data acquisition module, an inventory optimization module, a customer relationship management module, a supply chain optimization module, and a report generation module; the data acquisition module is used to acquire enterprise management data, and the enterprise management data includes enterprise operation data, enterprise customer data, and enterprise supply chain data; the inventory optimization module is used to analyze enterprise operation data through a time series prediction algorithm, predict enterprise product sales demand, build an inventory linear programming model, and obtain the enterprise's optimal inventory replenishment strategy and product safety inventory; the customer relationship management module is used to analyze enterprise customer data through a clustering algorithm, identify customer behavior patterns and demand trends, obtain enterprise customer classification, and formulate personalized marketing strategies; the supply chain optimization module is used to analyze enterprise supply chain data through mixed integer linear programming, establish a supply chain optimization model, and optimize resource allocation and transportation routes in the supply chain; the report generation module is used to summarize the analysis results of the inventory optimization module, the customer relationship management module, and the supply chain optimization module to generate a comprehensive business report; the specific process of building an inventory linear programming model to obtain the enterprise's optimal inventory replenishment strategy and product safety inventory is as follows: set inventory balance constraints to ensure that the inventory at each time point meets the balance constraints, the formula is, ;in, is the product inventory at time point t, is the replenishment quantity at time point t, is the product sales demand at time point t; set the inventory lower limit constraint to ensure that the inventory is not lower than the lower limit, the formula is, , where Saf is the lower limit inventory of the product; set the replenishment quantity limit constraint condition, the formula is, ,in, is the maximum replenishment quantity limit; determine the objective function that minimizes the total cost, the objective function formula is, ;in, represents minimizing the total cost, C is the total cost, is the product procurement cost, is the product storage cost, is the product out-of-stock cost, is the out-of-stock quantity at time point t; according to the inventory balance constraint, inventory lower limit constraint and replenishment quantity limit constraint, the optimal replenishment quantity is determined by solving the model using linear programming or other optimization algorithms and product safety stock ; The specific process of establishing a supply chain optimization model by analyzing enterprise supply chain data through mixed integer linear programming is as follows: define the decision variables for resource allocation and transportation paths in the supply chain, set some of the decision variables as integer variables and the rest as continuous variables; set minimizing the total cost of the supply chain as the objective function, and the supply chain optimization equation is as follows ; Where G represents the total cost of the supply chain, represents the purchase cost of the f-th product, represents the purchase quantity of the fth product, represents the transportation cost of the e-th transportation route, represents the transport volume of the e-th transport route, Indicates The storage cost of a warehouse, Indicates The inventory level in a warehouse.
[0006] Furthermore, the enterprise operation data includes historical sales data, existing product inventory, procurement data, supply data, and market trend data; the enterprise customer data includes customer purchase history data, transaction frequency data, and customer feedback data; the enterprise supply chain data includes supplier information data, production capacity data, storage capacity data, transportation cost data, and demand data.
[0007] Furthermore, the specific process of analyzing enterprise operation data through time series prediction algorithm and predicting enterprise product sales demand is as follows: preprocessing enterprise operation data, including data cleaning, data conversion and data standardization; inputting the preprocessed enterprise operation data into the autoregressive integral moving average model, analyzing the relationship between the current sales demand value and the sales demand value at the previous time point, and obtaining the autoregressive parameters; analyzing the relationship between the current sales demand value and the error term of the autoregressive integral moving average model, and obtaining the moving average parameters; the formula for predicting enterprise product sales demand is as follows, ;in, Represents the predicted future time point The sales demand value, represents the constant term of the model, representing the long-term average level of the data, represents the autoregressive parameter, represents the moving average parameter, that is, the relationship between the current sales demand value and the forecast error at the jth time point in the past. represents the actual sales demand value at time point t+hi, represents the forecast error term at time point t+hi, that is, the difference between the actual observation value and the model predicted value, p represents the autoregressive order, that is, the total number of observations, q represents the moving average order, that is, the total number of error terms, i represents the number of lags in the autoregressive part, that is, the number of time points used in the autoregressive model, j represents the number of lags in the moving average part, that is, the number of error terms used in the moving average model, and h represents the number of forecast periods.
[0008] Furthermore, the specific process of analyzing enterprise customer data through clustering algorithms and identifying customer behavior patterns and demand trends is as follows: clean, standardize and select features of enterprise customer data. Feature selection includes purchase frequency, purchase amount, and recent purchase time, and divide customers into high-frequency purchaser clusters and low-frequency purchaser clusters; randomly select the feature values of K customer samples, assign each customer to the cluster with the nearest centroid, calculate the mean of all customer features in each cluster, and update the position of the feature value mean. The formula is as follows: ,in, represents the k-th feature mean of the b-th cluster, represents the set of customers in the bth cluster, Represents the kth eigenvalue; according to the characteristic means of each cluster, the average purchasing behavior characteristics of customers in different clusters are obtained, customer behavior patterns are identified, clustering results of different time periods are compared, the degree of customer cluster migration is analyzed, and customer demand trends are identified.
[0009] Furthermore, the specific process of obtaining the enterprise customer classification is as follows: each customer is assigned to the cluster with the nearest centroid through the clustering algorithm. The formula for obtaining the enterprise customer classification is as follows: ;in, is the distance between the feature mean of the a-th customer and the b-th cluster, is the kth eigenvalue of the ath customer, represents the mean value of the kth feature of the bth cluster, and m represents the total number of features. According to the distance between the customer and the centroid of the jth cluster, the customers are divided into high-frequency buyers and low-frequency buyers.
[0010] Furthermore, the specific process of optimizing resource allocation and transportation paths in the supply chain is as follows: setting supply chain constraints, which include production capacity constraints, demand satisfaction constraints, transportation capacity constraints, and storage capacity constraints; solving the objective function through a mixed integer linear programming algorithm, and determining the optimal resource allocation and transportation path under the premise of satisfying the constraints, thereby optimizing the resource allocation and transportation path in the supply chain.
[0011] The present invention has the following beneficial effects:
[0012] (1) The data acquisition module of this digital AI intelligent analysis system based on big data integrates enterprise operation data, customer data and supply chain data to break the information island and achieve comprehensive data integration. This integration allows enterprises to obtain all relevant data on a unified platform and improve data utilization efficiency. The inventory optimization module uses time series prediction algorithms to accurately predict sales demand to avoid inventory backlogs and out-of-stock phenomena. By constructing an inventory linear programming model, enterprises can formulate the optimal inventory replenishment strategy to ensure that inventory levels can both meet demand and reduce holding costs. The customer relationship management module analyzes customer data through clustering algorithms and can identify the behavior patterns and demand trends of different customer groups. This identification helps enterprises divide customers into high-frequency and low-frequency buyers, thereby formulating more targeted personalized marketing strategies and improving customer satisfaction and loyalty.
[0013] (2) The supply chain optimization module of this digital AI intelligent analysis system based on big data uses mixed integer linear programming technology to optimize resource allocation and transportation routes, reduce logistics costs, and improve supply chain efficiency. This optimization not only reduces operating costs, but also improves the responsiveness and flexibility of the supply chain. The report generation module summarizes the analysis results of the inventory optimization, customer relationship management, and supply chain optimization modules to generate a comprehensive business report. This report provides enterprises with detailed business insights and decision support, helping management make data-driven strategic decisions.
[0014] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 This is a flow chart of the digital AI intelligent analysis system based on big data of the present invention. DETAILED DESCRIPTION
[0016] The embodiment of the present application solves the problems of enterprises in inventory management, customer relationship optimization and supply chain efficiency improvement through a digital AI intelligent analysis system based on big data.
[0017] The overall idea of the problem in the embodiment of this application is as follows:
[0018] Obtain enterprise management data, which includes enterprise operation data, enterprise customer data and enterprise supply chain data.
[0019] Through time series forecasting algorithms, we analyze enterprise operation data, predict enterprise product sales demand, build inventory linear programming models, and obtain the enterprise's optimal inventory replenishment strategy and product safety stock.
[0020] Analyze corporate customer data through clustering algorithms, identify customer behavior patterns and demand trends, obtain corporate customer classifications, and develop personalized marketing strategies.
[0021] Through mixed integer linear programming, enterprise supply chain data is analyzed, a supply chain optimization model is established, and resource allocation and transportation routes in the supply chain are optimized.
[0022] Summarize the analysis results of the inventory optimization module, customer relationship management module and supply chain optimization module to generate a comprehensive business report.
[0023] See also Figure 1 ,The embodiment of the present invention provides a technical solution: a digital AI intelligent analysis system based on big data, comprising the following modules: a data acquisition module, an inventory optimization module, a customer relationship management module, a supply chain optimization module, and a report generation module; the data acquisition module is used to acquire enterprise management data, which includes enterprise operation data, enterprise customer data, and enterprise supply chain data; the inventory optimization module is used to analyze enterprise operation data through a time series prediction algorithm, predict enterprise product sales demand, build an inventory linear programming model, and acquire the enterprise's optimal inventory replenishment strategy and product safety inventory; the customer relationship management module is used to analyze enterprise customer data through a clustering algorithm, identify customer behavior patterns and demand trends, acquire enterprise customer classification, and formulate personalized marketing strategies; the supply chain optimization module is used to analyze enterprise supply chain data through mixed integer linear programming, establish a supply chain optimization model, and optimize resource allocation and transportation paths in the supply chain; the report generation module is used to summarize the analysis results of the inventory optimization module, the customer relationship management module, and the supply chain optimization module to generate a comprehensive business report; the specific process of building an inventory linear programming model to acquire the enterprise's optimal inventory replenishment strategy and product safety inventory is as follows: set inventory balance constraints to ensure that the inventory at each time point meets the balance constraints, the formula is, ;in, is the product inventory at time point t, is the replenishment quantity at time point t, is the product sales demand at time point t; set the inventory lower limit constraint to ensure that the inventory is not lower than the lower limit, the formula is, , where Saf is the lower limit inventory of the product; set the replenishment quantity limit constraint condition, the formula is, ,in, is the maximum replenishment quantity limit; determine the objective function that minimizes the total cost, the objective function formula is, ;in, represents minimizing the total cost, C is the total cost, is the product procurement cost, is the product storage cost, is the product out-of-stock cost, is the out-of-stock quantity at time point t; according to the inventory balance constraint, inventory lower limit constraint and replenishment quantity limit constraint, the optimal replenishment quantity is determined by solving the model using linear programming or other optimization algorithms and product safety stock ; The specific process of establishing a supply chain optimization model by analyzing enterprise supply chain data through mixed integer linear programming is as follows: define the decision variables for resource allocation and transportation paths in the supply chain, set some of the decision variables as integer variables and the rest as continuous variables; set minimizing the total cost of the supply chain as the objective function, and the supply chain optimization equation is as follows ; Where G represents the total cost of the supply chain, represents the purchase cost of the f-th product, represents the purchase quantity of the fth product, represents the transportation cost of the e-th transportation route, represents the transport volume of the e-th transport route, Indicates The storage cost of a warehouse, Indicates The inventory level in a warehouse.
[0024] In this implementation scheme, the time series prediction algorithm in the inventory optimization module is an algorithm for analyzing historical data to predict future trends. The inventory linear programming model is a mathematical optimization model used to determine the optimal inventory level and replenishment strategy. Linear programming involves maximizing or minimizing the objective function under constraints. The clustering algorithm in the customer relationship management module is an algorithm for dividing data into several groups (clusters) so that the data in the same group are as similar as possible, and the data between different groups are as different as possible. Common clustering algorithms include K-means clustering and hierarchical clustering. K-means clustering classifies customers by iteratively updating the center point of the cluster. Customer classification is to group customers according to their purchasing behavior and other characteristics. This can help companies identify different types of customers, such as high-frequency buyers and low-frequency buyers, so as to develop more targeted marketing strategies. Mixed integer linear programming in the supply chain optimization module represents an optimization technique that combines integer constraints and linear constraints. It is used to solve optimization problems with integer decision variables (such as whether to select a certain supplier) and continuous decision variables (such as order quantity). Resource allocation and transportation path optimization is to optimize the allocation of resources and transportation routes through mathematical models and algorithms, with the aim of reducing costs and improving supply chain efficiency. For example, a mixed integer linear programming model is used to determine the optimal transportation route and warehouse location to reduce logistics costs and improve response speed. Set inventory balance constraints, which ensure that the inventory level at each time point meets the balance requirements. Specifically, the inventory update is determined by the previous inventory, replenishment, and demand. Set an inventory lower limit constraint, which ensures that the inventory level is always not less than the set lower limit inventory level. This can avoid the problem of insufficient inventory and ensure that the inventory level at any time point is above the safety level. Set a replenishment limit constraint, which defines the allowable range of replenishment, ensuring that the replenishment level will not be less than 0 and will not exceed the maximum limit. This avoids excessive or insufficient replenishment. Solve the above model using linear programming or other optimization algorithms to determine the optimal replenishment level and product safety stock. The optimization process will ensure that the total cost is minimized while meeting all constraints. In the supply chain optimization model, decision variables are used to describe the resource allocation and transportation path selection in the supply chain. Among them, some variables are integer variables (for example, the number of products assigned to a specific warehouse), while the rest are continuous variables (for example, the volume of transportation routes). Through this formula, the system can optimize costs in multiple dimensions (procurement, transportation, warehousing), ensuring that each link in the supply chain management of the enterprise can operate at the lowest cost. This optimization helps to improve the profitability of the enterprise. The system can optimize resource allocation based on the cost differences of different products, transportation routes and warehouses. For example, the system can select the lowest-cost supplier and transportation route, and reasonably allocate warehousing resources to reduce unnecessary expenses.By defining integer variables and continuous variables, the system can consider multiple practical constraints in the decision-making process, such as the capacity limit of the transportation route or the inventory limit of the warehouse. This flexibility allows the system to more accurately reflect the complexity of the actual business scenario. By optimizing the product procurement volume, transportation route and storage allocation, the system can reduce resource waste and improve the overall operational efficiency of the supply chain. This not only helps to reduce costs, but also shortens delivery time and improves customer satisfaction. By minimizing total costs, companies can also reduce energy consumption and carbon emissions in supply chain operations, thereby promoting sustainable development. This is of great significance to the long-term strategic development and social responsibility of enterprises.
[0025] Specifically, the enterprise operation data includes historical sales data, existing product inventory, procurement data, supply data, and market trend data; the enterprise customer data includes customer purchase history data, transaction frequency data, and customer feedback data; the enterprise supply chain data includes supplier information data, production capacity data, storage capacity data, transportation cost data, and demand data.
[0026] In this embodiment, historical sales data records past sales, including the sales quantity, sales time, sales location, etc. of each product. These data help analyze sales trends and seasonal fluctuations. The existing product inventory represents the number of products in the company's current inventory. Understanding inventory levels helps optimize replenishment and avoid out-of-stock or overstocking. Procurement data includes information such as the quantity, cost, and procurement time of products purchased from suppliers. It is used to analyze procurement efficiency and cost control. Supply data involves information such as the time, frequency, and quality of supplier delivery. It helps evaluate the reliability of the supply chain and supplier performance. Market trend data represents information about changes in market demand, competitor activities, industry development, etc. These data help predict future demand and adjust market strategies. Customer purchase history data represents records of products, purchase time, and quantity purchased by customers in the past. These data help understand customer purchasing behavior and preferences. Transaction frequency data represents the frequency of customer transactions, including the number of purchases and time intervals. This helps identify high-frequency and low-frequency customers. Customer feedback data represents customer evaluations, suggestions, and complaints about products and services. These data are used to improve product quality and service levels and enhance customer satisfaction. Supplier information data includes basic information, credit ratings, delivery capabilities, etc. of suppliers. It is used to select suitable suppliers and evaluate supply chain risks. Production capacity data indicates the production capacity of a company's production facilities, including production speed, number of production lines, and equipment maintenance status. This helps plan production capacity and meet demand. Storage capacity data indicates the storage capacity of a warehouse, including the warehouse's space, storage conditions, and current occupancy. It is used to optimize inventory management and distribution arrangements. Transportation cost data includes transportation costs, fuel costs, and losses during transportation. These data are used to reduce logistics costs and improve supply chain efficiency. Demand data is used to predict future product demand, including sales forecasts and market demand. These data help plan production and inventory strategies to meet market demand.
[0027] Specifically, the specific process of analyzing enterprise operation data through time series prediction algorithm and predicting enterprise product sales demand is as follows: preprocess the enterprise operation data, which includes data cleaning, data conversion and data standardization; input the preprocessed enterprise operation data into the autoregressive integral moving average model, analyze the relationship between the current sales demand value and the sales demand value at the previous time point, and obtain the autoregressive parameters; analyze the relationship between the current sales demand value and the error term of the autoregressive integral moving average model, and obtain the moving average parameters; the formula for predicting enterprise product sales demand is as follows, ;in, Represents the predicted future time point The sales demand value, represents the constant term of the model, representing the long-term average level of the data, represents the autoregressive parameter, represents the moving average parameter, that is, the relationship between the current sales demand value and the forecast error at the jth time point in the past. represents the actual sales demand value at time point t+hi, represents the forecast error term at time point t+hi, that is, the difference between the actual observation value and the model predicted value, p represents the autoregressive order, that is, the total number of observations, q represents the moving average order, that is, the total number of error terms, i represents the number of lags in the autoregressive part, that is, the number of time points used in the autoregressive model, j represents the number of lags in the moving average part, that is, the number of error terms used in the moving average model, and h represents the number of forecast periods.
[0028] In this implementation scheme, data cleaning includes removing erroneous values, missing values, and outliers from the data to ensure data quality. For example, deleting duplicate records and filling missing values. Data conversion is to convert data into a format suitable for analysis. Numerical data is logarithmically transformed to smooth the data. Data standardization is to scale the data to a uniform range to ensure that each data feature has the same scale. The preprocessed enterprise operation data is input into the ARIMA model for analysis. The autoregressive parameters are used to analyze the relationship between the current sales demand value and the sales demand value at the previous time point. The moving average parameters are used to analyze the relationship between the current sales demand value and the forecast error term at the previous time point.
[0029] Specifically, the specific process of analyzing enterprise customer data through clustering algorithms and identifying customer behavior patterns and demand trends is as follows: clean, standardize and select features of enterprise customer data. Feature selection includes purchase frequency, purchase amount, and recent purchase time, and divide customers into high-frequency purchaser clusters and low-frequency purchaser clusters; randomly select the feature values of K customer samples, assign each customer to the cluster with the nearest centroid, calculate the mean of all customer features in each cluster, and update the position of the feature value mean. The formula is as follows: ,in, represents the k-th feature mean of the b-th cluster, represents the set of customers in the bth cluster, Represents the kth eigenvalue; according to the characteristic means of each cluster, the average purchasing behavior characteristics of customers in different clusters are obtained, customer behavior patterns are identified, clustering results of different time periods are compared, the degree of customer cluster migration is analyzed, and customer demand trends are identified.
[0030] In this implementation, each customer is assigned to the cluster with the nearest cluster centroid (i.e., the characteristic mean of the cluster). Using a clustering algorithm (such as the K-means clustering algorithm), each customer determines the cluster to which it belongs by calculating the distance from the centroid of each cluster. By calculating the distance between the customer and the centroid of each cluster, the customer can be accurately assigned to the closest cluster. This helps ensure that the customer is assigned to the cluster that best represents their behavior patterns and characteristics, thereby improving the accuracy of classification. After identifying high-frequency buyers and low-frequency buyers, more targeted marketing strategies can be formulated. High-frequency buyers may receive more loyalty rewards and exclusive offers, while low-frequency buyers may receive promotions that stimulate purchases. By classifying customers into different clusters, companies can allocate resources more effectively, such as in customer service, advertising, and product promotion. This optimization can improve operational efficiency and increase customer satisfaction. Clustering results help identify customer behavior patterns and demand trends. For example, analyzing the purchasing habits and needs of different clusters can help companies predict market trends and adjust product strategies.
[0031] Specifically, the specific process of obtaining the enterprise customer classification is as follows: each customer is assigned to the cluster with the nearest centroid through the clustering algorithm. The formula for obtaining the enterprise customer classification is as follows: ;in, is the distance between the feature mean of the a-th customer and the b-th cluster, is the kth eigenvalue of the ath customer, represents the mean value of the kth feature of the bth cluster, and m represents the total number of features. According to the distance between the customer and the centroid of the jth cluster, the customers are divided into high-frequency buyers and low-frequency buyers.
[0032] In this embodiment, by calculating the distance between each customer and all cluster centroids, the customer is assigned to the closest cluster. According to the distance between the customer and the cluster centroid, the customer can be divided into a high-frequency buyer class and a low-frequency buyer class. The closer the distance, the closer the customer's behavior pattern is to the cluster. By accurately calculating the distance between the customer and the cluster centroid, the system can accurately assign the customer to the cluster that best matches its characteristics. This classification enables the system to better understand the customer's behavior pattern and needs, thereby performing more accurate market segmentation. Based on the customer classification results, the system can formulate targeted marketing strategies for different customer groups. For example, more loyalty programs can be pushed to high-frequency buyer customers, while promotional activities to attract repurchase can be launched for low-frequency buyer customers. This improves the efficiency and effectiveness of marketing activities. The system can optimize resource allocation based on customer classification, such as adjusting advertising delivery strategies, customizing product recommendation plans, and improving customer service quality. Through accurate customer classification, enterprises can allocate and utilize resources more reasonably and improve operational efficiency. Clustering algorithms can help the system identify changing trends in customer behavior. By comparing customer classification results in different time periods, enterprises can analyze changes in customer needs, adjust product and service strategies in advance, and maintain competitive advantages.
[0033] Specifically, the specific process of optimizing resource allocation and transportation paths in the supply chain is as follows: setting supply chain constraints, which include production capacity constraints, demand satisfaction constraints, transportation capacity constraints, and storage capacity constraints; solving the objective function through a mixed integer linear programming algorithm, and determining the optimal resource allocation and transportation path under the premise of satisfying the constraints, thereby optimizing the resource allocation and transportation path in the supply chain.
[0034] In this implementation scheme, the production capacity constraint is used to ensure that the production volume of each production plant does not exceed its maximum production capacity, that is, the production volume of the plant within a given time must be within its production capacity. The demand satisfaction constraint is used to ensure that the needs of all customers can be met in the downstream links of the supply chain (such as retailers and distributors). This means that the resource allocation and transportation of the supply chain must be able to support the satisfaction of the final demand. The transportation capacity constraint is used to ensure that the actual transportation volume of each transportation route does not exceed its maximum transportation capacity. The load of each transportation route must be within its capacity to ensure transportation efficiency and safety. The storage capacity constraint is used to ensure that the inventory of each warehouse does not exceed its maximum storage capacity to avoid warehouse overload or resource waste. Using the mixed integer linear programming algorithm, the system minimizes the total cost of the supply chain while satisfying all constraints. The objective function usually includes procurement cost, transportation cost and warehousing cost. By solving this optimization problem, the system can determine the optimal allocation method for each resource (such as raw materials, finished products). This includes allocating resources to appropriate production plants, warehouses and transportation routes to minimize costs and maximize efficiency. The system selects the transportation route with the lowest cost and highest efficiency through calculation to transport products from the production plant to the demand site or warehouse.
[0035] In summary, this application has at least the following effects:
[0036] The digital AI intelligent analysis system based on big data includes data acquisition module, inventory optimization module, customer relationship management module, supply chain optimization module and report generation module. Through the time series prediction algorithm, accurate demand forecasting is carried out on the enterprise operation data, and the inventory linear programming model is constructed to achieve the optimal inventory replenishment strategy and the determination of product safety inventory, reduce inventory costs, and improve inventory management efficiency. Customer data is analyzed through clustering algorithms to identify customer behavior patterns and demand trends. By dividing customers into high-frequency and low-frequency buyers, the system can formulate personalized marketing strategies, enhance customer satisfaction and loyalty, and improve market competitiveness. Mixed integer linear programming is used to analyze supply chain data, optimize resource allocation and transportation routes, reduce logistics costs and improve supply chain efficiency, thereby improving the response speed and flexibility of the supply chain.
[0037] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0038] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0039] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0040] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0041] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0042] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A digital AI intelligent analysis system based on big data, characterized by: Includes the following modules: data acquisition module, inventory optimization module, customer relationship management module, supply chain optimization module, report generation module; The data acquisition module is used to acquire enterprise management data, which includes enterprise operation data, enterprise customer data and enterprise supply chain data; The inventory optimization module is used to analyze the enterprise operation data through the time series prediction algorithm, predict the enterprise product sales demand, build an inventory linear programming model, and obtain the enterprise's optimal inventory replenishment strategy and product safety inventory; The customer relationship management module is used to analyze enterprise customer data through clustering algorithms, identify customer behavior patterns and demand trends, obtain enterprise customer classifications, and formulate personalized marketing strategies; The supply chain optimization module is used to analyze enterprise supply chain data through mixed integer linear programming, establish a supply chain optimization model, and optimize resource allocation and transportation paths in the supply chain; The report generation module is used to summarize the analysis results of the inventory optimization module, the customer relationship management module and the supply chain optimization module to generate a comprehensive business report; The specific process of building an inventory linear programming model and obtaining the company's optimal inventory replenishment strategy and product safety inventory is as follows: Set inventory balance constraints to ensure that the inventory quantity at each time point meets the balance constraints. The formula is: ;in, is the product inventory at time point t, is the replenishment quantity at time point t, is the product sales demand at time point t; Set the inventory lower limit constraint to ensure that the inventory quantity is not less than the lower limit. The formula is: , where Saf is the lower limit inventory of the product; Set the replenishment quantity limit constraint, the formula is, ,in, The maximum replenishment quantity limit; Determine the objective function that minimizes the total cost. The objective function formula is, ;in, represents minimizing the total cost, C is the total cost, is the product procurement cost, is the product storage cost, is the product out-of-stock cost, is the out-of-stock quantity at time point t; Determine the optimal replenishment quantity by solving the model using linear programming or other optimization algorithms based on inventory balance constraints, inventory lower limit constraints, and replenishment quantity limit constraints. and product safety stock ; The specific process of analyzing enterprise supply chain data through mixed integer linear programming and establishing a supply chain optimization model is as follows: Define the decision variables for resource allocation and transportation routes in the supply chain, set some of the decision variables as integer variables and the rest as continuous variables; Setting the minimization of the total cost of the supply chain as the objective function, the supply chain optimization equation is as follows ; Where G represents the total cost of the supply chain, represents the purchase cost of the f-th product, represents the purchase quantity of the fth product, represents the transportation cost of the e-th transportation route, represents the transport volume of the e-th transport route, Indicates The storage cost of a warehouse, Indicates The inventory level in a warehouse.
2. The digital AI intelligent analysis system based on big data according to claim 1 is characterized by: The enterprise operation data includes historical sales data, existing product inventory, procurement data, supply data, and market trend data; The enterprise customer data includes customer purchase history data, transaction frequency data and customer feedback data; The enterprise supply chain data includes supplier information data, production capacity data, storage capacity data, transportation cost data, and demand data.
3. The digital AI intelligent analysis system based on big data according to claim 2 is characterized in that: The specific process of analyzing enterprise operation data through time series prediction algorithm and predicting enterprise product sales demand is as follows: Preprocess the enterprise operation data, including data cleaning, data conversion and data standardization; Input the pre-processed enterprise operation data into the autoregressive integrated moving average model, analyze the relationship between the current sales demand value and the sales demand value at the previous time point, and obtain the autoregressive parameters; Analyze the relationship between the current sales demand value and the error term of the autoregressive integrated moving average model to obtain the moving average parameters; The formula for predicting the sales demand of enterprise products is as follows: ;in, Represents the predicted future time point The sales demand value, represents the constant term of the model, representing the long-term average level of the data, represents the autoregressive parameter, Represents the moving average parameter, that is, the current sales demand value and the past The relationship between the prediction errors at different time points is represents the actual sales demand value at time point t+hi, represents the forecast error term at time point t+hi, that is, the difference between the actual observation value and the model predicted value, p represents the autoregressive order, that is, the total number of observations, q represents the moving average order, that is, the total number of error terms, i represents the number of lags in the autoregressive part, that is, the number of time points used in the autoregressive model, j represents the number of lags in the moving average part, that is, the number of error terms used in the moving average model, and h represents the number of forecast periods.
4. The digital AI intelligent analysis system based on big data according to claim 3 is characterized by: The specific process of analyzing enterprise customer data through clustering algorithms and identifying customer behavior patterns and demand trends is as follows: Clean, standardize and select features of enterprise customer data. Feature selection includes purchase frequency, purchase amount, and last purchase time, and divide customers into high-frequency purchaser clusters and low-frequency purchaser clusters. Randomly select the feature values of K customer samples, assign each customer to the cluster with the nearest centroid, calculate the mean of all customer features in each cluster, and update the position of the feature value mean. The formula is as follows: ,in, represents the k-th feature mean of the b-th cluster, represents the set of customers in the bth cluster, represents the kth eigenvalue; Based on the characteristic mean of each cluster, the average purchasing behavior characteristics of customers in different clusters are obtained, customer behavior patterns are identified, clustering results of different time periods are compared, the degree of customer cluster migration is analyzed, and customer demand trends are identified.
5. The digital AI intelligent analysis system based on big data according to claim 4 is characterized in that: The specific process of obtaining enterprise customer classification is as follows: Each customer is assigned to the cluster with the closest centroid through the clustering algorithm. The formula for obtaining the enterprise customer classification is as follows: ;in, is the distance between the feature mean of the a-th customer and the b-th cluster, is the kth eigenvalue of the ath customer, represents the mean of the kth feature of the bth cluster, and m represents the total number of features; According to the distance between the customer and the centroid of the jth cluster, the customers are divided into high-frequency buyers and low-frequency buyers.
6. The digital AI intelligent analysis system based on big data according to claim 5 is characterized by: The specific process of optimizing resource allocation and transportation routes in the supply chain is as follows: Set supply chain constraints, which include production capacity constraints, demand satisfaction constraints, transportation capacity constraints, and storage capacity constraints; The objective function is solved through the mixed integer linear programming algorithm. Under the premise of satisfying the constraints, the optimal resource allocation and transportation path are determined to optimize the resource allocation and transportation path in the supply chain.
Citation Information
Patent Citations
Plan optimization method used for supply chain of chemical enterprise
CN108108994A
Seem flag supply chain sales and inventory prediction algorithm model and application system
CN114219169A
