Safety stock computing system and safety stock computing method

Abstract

A safety stock computing system and a safety stock computing method. A stock management system is designed. The safety stock computing system comprises a processor and a storage apparatus. The storage apparatus is used for storing a stock computing module, a statistical module, and a simulation and comparison module. The processor is electrically connected to the storage apparatus, and is used for executing the stock computing module, the statistical module, and the simulation and comparison module. On the basis of historical data, current data, and reliability factor data, the stock computing module computes a recommended safety stock. On the basis of the reliability factor data and the current data, the statistical module computes a statistical safety stock. On the basis of historical data, the simulation and comparison module simulates the recommended safety stock and the statistical safety stock, so as to generate a suitable safety stock on the basis of a simulation result. The stock computing module comprises a reliability factor analysis unit, the reliability factor analysis unit analyzing the historical data and the current data to produce the reliability factor data.

Description

Safe inventory calculation system and safe inventory calculation method

[0001] The present application claims priority to the Chinese patent application No. 202411390479.4, filed on September 30, 2024, and entitled "Safe inventory calculation system and safe inventory calculation method", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to an inventory management system, in particular to a safe inventory calculation system and a safe inventory calculation method. BACKGROUND

[0003] Generally speaking, the stock strategy of a component enterprise is mainly based on customer orders and stock planning for production and manufacturing. However, stock planning needs to predict the future demand quantity and demand time of customers. Therefore, in order to meet the future needs of customers and avoid the situation that the supply cannot meet the customers, the component enterprise usually produces more components, which not only causes great inventory pressure, but also causes high costs of production funds and space occupation. However, if the inventory quantity is reduced without calculation or planning, it will cause the problem that the customer expectations cannot be fulfilled, and the customer satisfaction will be reduced. SUMMARY

[0004] The present application is directed to a safe inventory calculation system and a safe inventory calculation method, which can generate a variable of a safe inventory quantity according to analysis data, so as to effectively generate a safe inventory quantity with low comprehensive cost.

[0005] According to an embodiment of the present application, the safe inventory calculation system of the present application comprises a storage device and a processor. The storage device is used to store an inventory calculation module, a statistical module, and a simulation and comparison module. The processor is electrically connected to the storage device and is used to execute the inventory calculation module, the statistical module, and the simulation and comparison module. The inventory calculation module calculates a recommended safe inventory according to historical data, current data, and reliable factor data. The statistical module calculates a statistical safe inventory according to the reliable factor data and the current data. The simulation and comparison module simulates the recommended safe inventory and the statistical safe inventory based on the historical data, so as to generate a suitable safe inventory according to the simulation result. The inventory calculation module comprises a reliable factor analysis unit, which analyzes the historical data and the current data to generate the reliable factor data.

[0006] According to an embodiment of the present application, the safety stock calculation method of the present application comprises the following steps: executing an inventory calculation module by a processor to calculate a recommended safety stock according to historical data, current data and reliable factor data; executing a statistics module by the processor to calculate a statistical safety stock according to the reliable factor data and the current data; executing a simulation and comparison module by the processor to simulate the recommended safety stock and the statistical safety stock based on the historical data, and to generate a suitable safety stock according to the simulation result, wherein the inventory calculation module comprises a reliable factor analysis unit, and the reliable factor analysis unit analyzes the historical data and the current data to generate the reliable factor data.

[0007] Based on the above, the safety stock calculation system and the safety stock calculation method of the present application can perform data processing and correlation analysis based on reliable factor setting data and historical and current data to obtain perfect reliable factor data, and can calculate safety stock amounts with low comprehensive cost based on inventory calculation models and statistics, respectively.

[0008] In order to make the above features and advantages of the present application more apparent, specific examples are described below, and are described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0009] Fig. 1 is a schematic diagram of a safety stock calculation system according to an embodiment of the present application;

[0010] Fig. 2 is a flowchart of a safety stock calculation method according to an embodiment of the present application.

[0011] BRIEF DESCRIPTION OF DRAWINGS 100: safety stock calculation system; 110: processor; 120: storage device; 121: inventory calculation module; 122: statistics module; 123: simulation and comparison module; S210, S220, S230: steps. DETAILED DESCRIPTION

[0012] Reference will now be made in detail to exemplary embodiments of the present application, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers are used in different drawings and descriptions to indicate the same or similar parts.

[0013] FIG. 1 is a schematic diagram of a safety inventory calculation system according to an embodiment of the present application. Referring to FIG. 1, a safety inventory calculation system 100 includes a processor 110 and a storage device 120. The processor 110 is electrically connected to the storage device 120. In this embodiment, the safety inventory calculation system 100 can communicate with a database, a server, or an enterprise resource planning system, without being limited thereto. The processor 110 can obtain relevant data from the database, the server, or the enterprise resource planning system. The storage device 120 can store relevant algorithms, programs, and / or software of an inventory calculation module 121, a statistics module 122, and a simulation and comparison module 123, and can store or temporarily store data generated in a data processing process.

[0014] In this embodiment, the processor 110 can be, for example, a system on a chip (SOC), or can include, for example, a central processing unit (CPU) or other programmable general purpose or special purpose microprocessors (Microprocessor), digital signal processors (DSP), programmable controllers, application specific integrated circuits (ASIC), programmable logic devices (PLD), other similar processing devices, or a combination of these devices. In this embodiment, the storage device 120 can be, for example, a dynamic random access memory (DRAM), a flash memory, or a non-volatile random access memory (NVRAM), etc.

[0015] In an embodiment, the secure inventory calculation system 100 can be implemented through an architecture of a cloud server system. A user can perform a user interface (UI) program of an electronic device to connect to the cloud server to perform the above-mentioned parameter and rule setting operations. In this regard, the user can operate the content of the user interface displayed on the display screen of the electronic device, so that the user interface, the application programming interface (API) or the related program can provide corresponding user operation instructions and setting data to the cloud server. The cloud server can be, for example, a software as a service (SaaS) server, and the application programming interface corresponds to a software as a service application, so that the secure inventory calculation system 100 can be set in the software as a service server, and the application programming interface receives and transmits data to the data in the database of the enterprise resource planning (ERP) system. In another embodiment, the secure inventory calculation system 100 can be set in a local server in the enterprise, and then connected to the database of the enterprise resource planning system in the local server and the cloud database to input / output data.

[0016] However, the present application is not limited thereto. Alternatively, in an embodiment, the secure inventory calculation system 100 can be set in a local server in the enterprise, and then connected to the database of the enterprise resource planning system in the local server and the cloud database to input / output data, so as to provide the functions of related task management, task flow management, risk analysis and generation of processing suggestion information through different application programming interfaces. In an embodiment, the inventory calculation module 121, the statistical module 122, the simulation and comparison module 123 and other modules and units can be implemented, for example, in a program language such as JSON (JavaScript Object Notation), Extensible Markup Language (XML) or YAML, but the present application is not limited thereto.

[0017] Figure 2 is a flow chart of a safety stock calculation method according to an embodiment of the present application. Referring to Figure 1 and Figure 2, the safety stock calculation system 100 can perform the following steps S210-S230. In step S210, the processor 110 can execute the inventory calculation module 121 to calculate a recommended safety stock according to historical data, current data and reliability factor data by the inventory calculation module 121. In this embodiment, the reliability factor data is a variable of the safety stock. The inventory calculation module 121 can generate a safety stock (amount) that conforms to the current data according to the correlation between the historical data and the reliability factor data. In step S220, the processor 110 can execute the statistics module 122 to calculate a statistical safety stock according to the reliability factor data and the current data by the statistics module 122.

[0018] In this embodiment, the statistics module 122 calculates a statistical safety stock according to the reliability factor data and the current data based on a statistical formula. Specifically, the statistics module 122 performs a statistical operation on the reliability factor data and the current data based on a preset statistical formula to generate a statistical safety stock. In this embodiment, the preset statistical formula is related to the reliability factor data, the demand standard deviation of historical data, the average replenishment period and the demand standard deviation within the replenishment period. The preset statistical formula is shown in the following formula (1) and formula (2): d ……(2);

[0019] SS is the statistical safety stock, Z is the reliability coefficient, σ is the demand standard deviation within the replenishment period, σ d is the standard deviation of demand history, and L is the average replenishment period. In this embodiment, the current data includes the past prediction and demand of the current user. In this way, the statistics module 122 obtains the demand amount, the prediction amount and the error value of each period according to the current product of the current customer, and then obtains the demand standard deviation σ within the replenishment period. Then, the reliability factor data is related to the demand date, the demand amount, the delivery amount, the shortage amount and the service times corresponding to the compliance rate of each product of each user. In this way, the statistics module 122 obtains the reliability coefficient Z according to the reliability factor data by the distribution function, and then calculates the statistical safety stock.

[0020] At step S230, the processor 110 can execute the simulation and comparison module 123 to simulate the recommended safety stock and the statistical safety stock based on the historical data by the simulation and comparison module 123 to generate the appropriate safety stock according to the simulation results. In this embodiment, the simulation and comparison module 123 generates a number of random values based on the cumulative probability in the historical data, and then simulates the conditions when the statistical safety stock and the recommended safety stock are adopted respectively according to the number of random values. That is, the simulation and comparison module 123 obtains the comprehensive cost of the statistical safety stock based on the number of random values and the simulation results of adopting the statistical safety stock. Similarly, the simulation and comparison module 123 obtains the comprehensive cost of the recommended safety stock based on the number of random values and the simulation results of adopting the recommended safety stock. Then, the simulation and comparison module 123 selects the value with lower cost from the comprehensive cost of the recommended safety stock and the comprehensive cost of the statistical safety stock, and takes it as the appropriate safety stock.

[0021] In this embodiment, the inventory calculation module 121 comprises a reliability factor analysis unit, wherein the reliability factor analysis unit analyzes the historical data and the current data to generate the reliability factor data. Specifically, the reliability factor analysis unit obtains the on-time rate of each transaction according to the data of the demand date, the demand quantity, the delivery quantity, the shortage quantity and the service times of each product of each customer in the historical data. In this way, the reliability factor analysis unit can evaluate the reliability factor data of other times based on the historical data to obtain the perfect reliability factor data.

[0022] In an embodiment, the reliability factor analysis unit is a reliability factor analysis model. The processor 110 pre-processes the reliability factor setting data, the historical data and the current data to generate a sample data set. The reliability factor setting data is related to the quantitative and evaluation criteria of the safety stock. The pre-processing can be missing value processing, field normalization or white noise processing, etc. Then, the processor 110 divides the sample data set into 3:7 to generate a test set and a training set in the training data set, and the processor 110 constructs the reliability factor analysis model according to the training data set.

[0023] In an embodiment, the processor 110 builds the inventory calculation model based on a particle swarm algorithm combined with a back propagation (BP) neural network algorithm. In this way, the trained inventory calculation model can generate a recommended safety stock according to current data. In an embodiment, after the processor 110 pre-processes the data, the following steps are further included: the processor 110 performs field correlation analysis on the pre-processed data to obtain the correlation degree between each field and the safety stock and the field correlation degree between each field and each other field. Then, the processor 110 updates the pre-processed data with the fields having a correlation degree greater than a threshold value, and the processor 110 filters out the fields having a correlation degree greater than the threshold value to generate a sample data set. For example, the safety stock is the dependent variable Y, and the reliable factor data is the independent variable X (i.e., the field), and the processor 110 updates the reliable factor data with the fields having a correlation degree greater than a threshold value (e.g., 0.5) as new reliable factors. Moreover, the processor 110 filters out one of the fields in the reliable factor data having a field correlation degree greater than the threshold value with each other field, thereby improving the independence between the fields (i.e., the reliable factors) to avoid the situation that the weight of a certain field is too large, so as to improve the accuracy of the data.

[0024] In an embodiment, the inventory calculation module 121 is an inventory calculation model. The processor 110 pre-processes the historical data and the reliable factor data to generate a sample data set. Similarly, the processor 110 divides the sample data set into a test set and a training set in the training data set in a ratio of 3 to 7, and then the processor 110 builds the inventory calculation model based on a particle swarm algorithm according to the training data set. In an embodiment, after the processor 110 pre-processes the data, the following steps are further included: the processor 110 performs field correlation analysis on the pre-processed data to obtain the correlation degree between each field and the safety stock, and then the processor 110 updates the pre-processed data based on the fields having a correlation degree greater than a threshold value (e.g., 0.5) to generate a sample data set.

[0025] In the present embodiment, the processor 110 builds the inventory calculation model based on a particle swarm algorithm combined with a BP neural network algorithm provided with a network hidden layer number, wherein the node number of the network hidden layer number is a node number having a minimum mean square error among a plurality of node numbers. For example, a range of a node number is first determined according to an empirical formula of the hidden layer node number, and then the mean square error of the neural network under each node number is calculated to obtain the mean square error of each node number, and then the node number having the minimum mean square error is selected.

[0026] In one embodiment, the simulation and comparison module 123 simulates the recommended safety stock and the statistical safety stock by Monte Carlo simulation. Specifically, the simulation and comparison module 123 simulates the recommended safety stock and the statistical safety stock respectively based on historical data to obtain the overall cost of the recommended safety stock and the overall cost of the statistical safety stock respectively. The overall cost is related to the ordering cost, the warehousing cost and the shortage cost, wherein the shortage cost is the cost of compensation according to the agreement with the customer when the customer demand cannot be fulfilled. The ordering cost is the cost when ordering. In one embodiment, the overall cost is the sum of the ordering cost, the warehousing cost and the shortage cost. Then, the simulation and comparison module 123 selects the safety stock with lower overall cost from the recommended safety stock and the statistical safety stock as the suitable safety stock.

[0027] In summary, the safety stock calculation system and the safety stock calculation method of the present application can analyze and evaluate the reliable factor setting data and the historical and current data to improve the reliable factor data related to the safety stock. Moreover, the neural network model is trained based on the reliable factor data to obtain the recommended safety stock. In this way, the safety stock system can obtain the safety stock with low overall cost from the recommended safety stock and the statistical safety stock calculated based on statistics according to the Monte Carlo simulation to provide the user with high accuracy and low cost safety stock.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the above embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A secure inventory calculation system, characterized by, The application relates to a safety stock calculation system, comprising: a storage device for storing an inventory calculation module, a statistics module and a simulation and comparison module; and a processor electrically connected to the storage device and used for executing the inventory calculation module, the statistics module and the simulation and comparison module, wherein the inventory calculation module calculates a recommended safety stock according to historical data, current data and reliability factor data, wherein the statistics module calculates a statistical safety stock according to the reliability factor data and the current data, wherein the simulation and comparison module simulates the recommended safety stock and the statistical safety stock based on the historical data to generate a suitable safety stock according to a simulation result, wherein the inventory calculation module comprises a reliability factor analysis unit, wherein the reliability factor analysis unit analyzes the historical data and the current data to generate the reliability factor data.

2. The secure inventory calculation system of claim 1, wherein, The statistics module statistically operates the reliability factor data and the current data based on a preset statistical formula to generate the statistical safety stock, wherein the preset statistical formula is related to the reliability factor data, a demand standard deviation of the historical data, an average replenishment period and a demand standard deviation within the replenishment period.

3. The secure inventory calculation system of claim 1, wherein, The reliability factor analysis unit is a reliability factor analysis model, wherein the processor pre-processes reliability factor setting data, the historical data and the current data to generate a sample data set, wherein the processor generates a training data set according to the sample data set, wherein the processor constructs the reliability factor analysis model according to the training data set, wherein the reliability factor setting data is related to a quantitative and evaluation standard of safety stock.

4. The system of claim 3, wherein, The processor constructs the inventory calculation model based on a particle swarm algorithm combined with a BP neural network algorithm.

5. The system of claim 3, wherein, The processor performs field correlation analysis on the pre-processed data to obtain a correlation degree between each field and safety stock and a field correlation degree between each field and each other field, wherein the processor updates the pre-processed data based on the correlation degree being greater than a threshold value and filters out the field with the field correlation degree being greater than the threshold value to generate the sample data set.

6. The secure inventory calculation system of claim 1, wherein, The inventory calculation module is an inventory calculation model, wherein the processor pre-processes the historical data and the reliability factor data to generate a sample data set, wherein the processor generates a training data set according to the sample data set, wherein the processor constructs the inventory calculation model based on a particle swarm algorithm according to the training data set.

7. The system of claim 6, wherein, The processor performs field correlation analysis on the pre-processed data to obtain a correlation degree between each field and safety stock, wherein the processor updates the pre-processed data based on the correlation degree being greater than a threshold value to generate the sample data set.

8. The secure inventory calculation system of claim 6, wherein, The processor constructs the inventory calculation model based on the particle swarm algorithm combined with a BP neural network algorithm provided with a network hidden layer number, wherein the node number of the network hidden layer is a node number with a minimum mean square error among multiple node numbers.

9. The system of claim 1, wherein, The simulation and comparison module simulates the recommended safety stock and the statistical safety stock by Monte Carlo simulation.

10. The secure inventory calculation system of claim 9, wherein, The simulation and comparison module respectively simulates the recommended safety stock and the statistical safety stock based on historical data to respectively obtain a comprehensive cost of the recommended safety stock and a comprehensive cost of the statistical safety stock, Wherein the comprehensive cost is related to order cost, storage cost, and shortage cost, Wherein the simulation and comparison module selects the safety stock with lower comprehensive cost from the recommended safety stock and the statistical safety stock as the suitable safety stock.

11. A method of safe stock calculation, characterized by, Comprise: A processor executes an inventory calculation module to calculate a recommended safety stock according to historical data, current data, and reliability factor data; The processor executes a statistics module to calculate a statistical safety stock according to the reliability factor data and the current data; The processor executes a simulation and comparison module to simulate the recommended safety stock and the statistical safety stock based on the historical data, and then generates a suitable safety stock according to the simulation result; The processor executes a reliability factor analysis unit to analyze the historical data and the current data to generate the reliability factor data, wherein the inventory calculation module comprises the reliability factor analysis unit.

12. The method of claim 11, wherein, The step of calculating the statistical safety stock according to the reliability factor data and the current data comprises: The processor executes the statistics module to statistically operate the reliability factor data and the current data based on a preset statistical formula to generate the statistical safety stock, Wherein the preset statistical formula is related to the reliability factor data, demand standard deviation of the historical data, average replenishment period, and demand standard deviation within the replenishment period.

13. The method of claim 11, wherein, The reliability factor analysis unit is a reliability factor analysis model, Wherein the method further comprises: The processor pre-processes reliability factor setting data, the historical data, and the current data to generate a sample data set; The processor generates a training data set according to the sample data set; and The processor constructs the reliability factor analysis model according to the training data set, wherein the reliability factor setting data is related to the quantitative and evaluation standards of safety stock.

14. The method of claim 13, wherein, The step of constructing the reliability factor analysis model according to the training data set comprises: The processor constructs the inventory calculation model based on a particle swarm algorithm combined with a BP neural network algorithm.

15. The method of claim 13, wherein, After pre-processing the reliability factor setting data, the historical data, and the current data, the following steps are included: The processor performs field correlation analysis on the pre-processed data to obtain the correlation degree between each field and safety stock and the field correlation degree between each field and each other field; and The processor updates the pre-processed data based on the correlation degree being greater than a threshold value, and filters out the field with the field correlation degree greater than the threshold value to generate the sample data set.

16. The method of claim 11, wherein, The inventory calculation module is an inventory calculation model, and the method further comprises: preprocessing, by the processor, the historical data and the reliable factor data to generate a sample data set; generating, by the processor, a training data set based on the sample data set; and constructing, by the processor, the inventory calculation model based on a particle swarm algorithm and the training data set.

17. The method of claim 16, wherein, After preprocessing the historical data and the reliable factor data, the method comprises the following steps: performing, by the processor, field correlation analysis on the preprocessed data to obtain the correlation degree between each field and the safety stock; and updating, by the processor, the preprocessed data based on the correlation degree being greater than a threshold field to generate the sample data set.

18. The method of claim 16, wherein, The step of constructing the inventory calculation model comprises: constructing, by the processor, the inventory calculation model based on the particle swarm algorithm combined with a BP neural network algorithm with a set number of network hidden layers, wherein the number of nodes of the network hidden layers is the number of nodes with the smallest mean square error among a plurality of node numbers.

19. The method of claim 11, wherein, The step of simulating the recommended safety stock and the statistical safety stock based on the historical data comprises: executing, by the processor, the simulation and comparison module to simulate the recommended safety stock and the statistical safety stock by Monte Carlo simulation.

20. The method of claim 19, wherein, The step of simulating the recommended safety stock and the statistical safety stock, and then generating the suitable safety stock based on the simulation result comprises: executing, by the processor, the simulation and comparison module to respectively simulate the recommended safety stock and the statistical safety stock based on historical data, and then respectively obtain the comprehensive cost of the recommended safety stock and the comprehensive cost of the statistical safety stock, wherein the comprehensive cost is related to order cost, storage cost, and shortage cost; and executing, by the processor, the simulation and comparison module to select the one with lower comprehensive cost as the suitable safety stock from the recommended safety stock and the statistical safety stock.

Images