CN109902850B - Method, device and storage medium for determining inventory control strategy - Google Patents
Method, device and storage medium for determining inventory control strategy Download PDFInfo
- Publication number
- CN109902850B CN109902850B CN201810972843.6A CN201810972843A CN109902850B CN 109902850 B CN109902850 B CN 109902850B CN 201810972843 A CN201810972843 A CN 201810972843A CN 109902850 B CN109902850 B CN 109902850B
- Authority
- CN
- China
- Prior art keywords
- inventory
- scene
- product
- historical
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method, a device and a storage medium for determining an inventory control strategy, and belongs to the technical field of big data. The method comprises the following steps: generating a plurality of inventory scene simulators according to the inventory data set, wherein each inventory scene simulator corresponds to one inventory scene; according to the inventory data set, an initial inventory control strategy comprising scene parameters is determined, and the scene parameters in the initial inventory control strategy are optimized based on a plurality of inventory scene simulators to obtain a target inventory control strategy. Because the scene parameters in the target inventory control strategy are obtained after optimization based on the multiple inventory scene simulators, the target inventory control strategy can be simultaneously applied to multiple inventory scenes, so that the flexibility of the target inventory control strategy is improved, and the accuracy of adjusting the inventory through the target inventory control strategy can be improved.
Description
Technical Field
The present application relates to the field of big data technologies, and in particular, to a method and an apparatus for determining an inventory control policy, and a storage medium.
Background
At present, in the process of enterprise operation, if the inventory of products is too high, the waste of enterprise funds is easily caused, and thus the inventory holding cost is generated. If the inventory of the product is too low, the inventory may not meet the demand of the market, and the enterprise needs to spend extra funds to make up for the loss caused by the insufficient inventory, thereby generating the stock shortage cost. Therefore, there is a need to develop an inventory control strategy to control the inventory of products in an enterprise to reduce inventory costs. Wherein the inventory cost includes inventory holding cost and inventory backorder cost.
Since the supply period of the product is usually a fixed value, the demand amounts of the product at each time are usually randomly distributed, and the demand amounts of the product at each time are independent from each other, the related art provides an inventory control strategy based on the inventory threshold value, which is updated according to a random period. In the inventory control strategy, the inventory threshold value is changed randomly according to a period, in a certain period, if the existing inventory is smaller than the inventory threshold value in the current period, a first decision is adopted to adjust the inventory so as to reduce the inventory cost, and if the existing inventory is larger than or equal to the inventory threshold value in the current period, a second decision is adopted to adjust the inventory so as to reduce the inventory cost. Wherein, the decision I and the decision II are two preset decisions.
It should be noted that the inventory control strategy is a scenario in which the supply period of the product is constant, but the actual supply period of the product may have uncertainty. For example, the actual supply period of the product may be changed due to the processing progress of the product or the physical distribution condition of the product, in this case, the inventory amount cannot be adjusted by the inventory control policy any more, that is, the inventory control policy is only applicable to a scenario where the supply period of the product is a fixed value, so that the flexibility of the inventory control policy is low, and the accuracy of adjusting the inventory amount by the inventory control policy is not high.
Disclosure of Invention
The application provides a method, a device and a storage medium for determining an inventory control strategy, which can solve the problems that the flexibility of the inventory control strategy provided in the related technology is low and the accuracy of adjusting the inventory is not high. The technical scheme is as follows:
in a first aspect, a method of determining an inventory control policy is provided, the method comprising:
the method comprises the steps of obtaining an inventory data set, wherein the inventory data set comprises data used for describing historical inventory information of at least one product, generating a plurality of inventory scene simulators according to the inventory data set, each inventory scene simulator corresponds to one inventory scene, determining an initial inventory control strategy comprising scene parameters according to the inventory data set, optimizing the scene parameters in the initial inventory control strategy according to the inventory scene simulators to obtain a target inventory control strategy, and the target inventory control strategy is used for adjusting inventory.
In the application, because the scene parameters in the target inventory control strategy are obtained after optimization based on a plurality of inventory scene simulators, the target inventory control strategy can be simultaneously applied to a plurality of inventory scenes, so that the flexibility of the target inventory control strategy is improved, and the accuracy of adjusting the inventory through the target inventory control strategy can be improved.
Optionally, optimizing the scenario parameters in the initial inventory control policy according to a plurality of inventory scenario simulators to obtain a target inventory control policy, including: determining a corresponding relation between a scene parameter value associated with each inventory scene simulator and inventory cost according to each inventory scene simulator and an initial inventory control strategy; and optimizing the scene parameters according to the corresponding relation between the plurality of scene parameter values and the inventory cost which are associated with the plurality of inventory scene simulators one by one to obtain a target inventory control strategy.
In the method, the corresponding relation between the scene parameter value and the inventory cost for each inventory scene simulator can be determined, and then the scene parameters are optimized according to the determined corresponding relations, so that the optimized scene parameters can be adapted to a plurality of inventory scenes simultaneously, and the flexibility of the target inventory control strategy determined by the method is improved
Optionally, determining a correspondence between a scene parameter value associated with each inventory scene simulator and an inventory cost according to each inventory scene simulator and the initial inventory control policy, includes: for any inventory scene simulator A in the inventory scene simulators, generating a plurality of sample data based on the inventory scene simulator A, wherein each sample data is used for describing the demand information or supply time information of a product; and inputting a plurality of pieces of sample data into the initial inventory control strategy to obtain the corresponding relation between the scene parameter value associated with the inventory scene simulator A and the inventory cost.
In one possible implementation, the correspondence between the scene parameter values and the inventory costs for the inventory scene simulator a may be determined in the manner described above.
Optionally, optimizing the scene parameters according to a correspondence between a plurality of scene parameter values and inventory costs associated with the plurality of inventory scene simulators one to obtain a target inventory control policy, including: determining a scene parameter extreme value associated with each inventory scene simulator according to a corresponding relation between a scene parameter value associated with each inventory scene simulator and inventory costs, wherein the inventory cost corresponding to the scene parameter extreme value is the lowest in the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory costs; and determining the average value of a plurality of scene parameter extreme values which are associated with a plurality of inventory scene simulators one by one, and taking the determined average value as the optimized value of the scene parameter to obtain the target inventory control strategy.
In the present application, an average value of a plurality of scene parameter extrema associated with a plurality of inventory scene simulators one to one may be used as a value of a scene parameter after optimization, so that the determined target control strategy can be applied to a plurality of inventory scenes simultaneously.
Optionally, the inventory data set includes historical demand data and historical shelf information describing each of the at least one product. Accordingly, generating a plurality of inventory scenario simulators from an inventory data set includes: for any product B in at least one product, determining a historical demand distribution curve of the product B according to historical demand data of the product B in the inventory data set, and generating a plurality of first-class inventory scene simulators for the product B according to the historical demand distribution curve of the product B, wherein each first-class inventory scene simulator corresponds to one demand scene. Determining a historical delivery time distribution curve of the product B according to historical delivery time information of the product B in the inventory data set, and generating a plurality of second-class inventory scene simulators for the product B according to the historical delivery time distribution curve of the product B, wherein each second-class inventory scene simulator corresponds to one delivery time scene. Wherein the plurality of inventory scenario simulators comprise: a plurality of first type inventory scenario simulators and a plurality of second type inventory scenario simulators for each product in at least one product.
In the present application, in order to enable the determined target control strategy to simultaneously adapt to the uncertainty of the supply period and the uncertainty of the demand, the generated multiple inventory scenario simulators include the multiple first-type inventory scenario simulators and the multiple second-type inventory scenario simulators.
Optionally, generating a plurality of first-class inventory scenario simulators for the product B according to the historical demand distribution curve of the product B includes: determining the mean value and the standard deviation of the historical demand distribution curve of the product B; adjusting the mean value and/or standard deviation of the historical demand distribution curve of the product B to obtain a plurality of demand distribution curves; and generating a first-class inventory scene simulator according to each demand distribution curve to obtain a plurality of first-class inventory scene simulators for the product B.
Specifically, a plurality of first-type inventory scenario simulators for product B may be derived from the historical demand distribution curve of product B based on the manner described above.
Optionally, generating a plurality of second-class inventory scenario simulators for the product B according to the historical delivery time distribution curve of the product B includes: determining the mean value and the standard deviation of the historical supply time distribution curve of the product B; adjusting the mean value and/or standard deviation of the historical supply time distribution curve of the product B to obtain a plurality of supply time distribution curves; and generating a second-class inventory scene simulator according to each supply time distribution curve to obtain a plurality of second-class inventory scene simulators for the product B.
Similarly, a plurality of second-class inventory scenario simulators for the product B can be derived according to the historical supply time distribution curve of the product B based on the above manner.
Optionally, determining an initial inventory control policy including scenario parameters according to the inventory data set includes: acquiring an initial model for inventory control; determining historical demand characteristics and historical supply time characteristics of each product in at least one product according to the inventory data set; and updating the initial model according to the historical demand characteristics and the historical supply time characteristics of each product in at least one product to obtain an initial inventory control strategy comprising scene parameters.
In the present application, the initial inventory control policy is obtained by updating the initial model according to the inventory data set.
In a second aspect, there is provided an apparatus for determining an inventory control strategy, the apparatus for determining an inventory control strategy having the functionality to implement the method acts of determining an inventory control strategy as described in the first aspect above. The apparatus for determining an inventory control policy comprises at least one module configured to implement the method for determining an inventory control policy provided in the first aspect.
In a third aspect, an apparatus for determining an inventory control policy is provided, where the apparatus for determining an inventory control policy includes a processor and a memory, and the memory is used to store a program for supporting the apparatus for determining an inventory control policy to execute the method for determining an inventory control policy provided in the first aspect, and to store data for implementing the method for determining an inventory control policy provided in the first aspect. The processor is configured to execute programs stored in the memory. The operating means of the memory device may further comprise a communication bus for establishing a connection between the processor and the memory.
In a fourth aspect, there is provided a computer-readable storage medium having stored therein instructions which, when run on a computer, cause the computer to perform the method of determining an inventory control policy of the first aspect.
In a fifth aspect, there is provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the method of determining an inventory control policy of the first aspect.
The technical effects obtained by the above second, third, fourth and fifth aspects are similar to the technical effects obtained by the corresponding technical means in the first aspect, and are not described herein again.
Drawings
FIG. 1 is a schematic diagram of a system for determining an inventory control policy according to an embodiment of the present disclosure;
FIG. 2 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure;
FIG. 3 is a flow chart of a method for determining an inventory control policy provided by an embodiment of the present application;
FIG. 4 is a diagram illustrating simulation results provided by an embodiment of the present application;
FIG. 5 is a diagram illustrating another simulation result provided by an embodiment of the present application;
fig. 6 is a block diagram of an apparatus for determining an inventory control policy according to an embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
Before explaining the method for determining the inventory control policy provided in the embodiment of the present application, an application scenario of the embodiment of the present application is briefly introduced.
Currently, when a person related to a company makes a product purchase, it is generally necessary to determine a current amount of the purchase according to an optimal stock amount, which is a stock amount at which stock cost can be minimized. The inventory control strategy is a model for determining the optimal inventory amount, and related personnel can determine the optimal inventory amount according to the inventory control strategy. The method for determining the inventory control strategy provided by the embodiment of the application can be applied to an application scene that the optimal inventory amount needs to be determined when products are purchased, and of course, the method for determining the inventory control strategy provided by the embodiment of the application can also be applied to other scenes that the optimal inventory amount needs to be determined, and is not limited herein.
Fig. 1 is a schematic diagram of a system for determining an inventory control policy according to an embodiment of the present disclosure, and as shown in fig. 1, the system 100 includes an inventory data input module 101, an inventory scenario simulator generation module 102, an inventory scenario simulator sampling module 103, an algorithm selection module 104, an algorithm modification module 105, a meta-learning training module 106, and a result output module 107.
The inventory data input module 101 is respectively connected with the inventory scene simulator generation module 102 and the algorithm transformation module 105, the algorithm selection module 104 is connected with the algorithm transformation module 105, the inventory scene simulator generation module 102 is connected with the inventory scene simulator sampling module 103, the inventory scene simulator sampling module 103 is connected with the meta-learning training module 106, the algorithm transformation module 105 is also connected with the meta-learning training module 106, and the meta-learning training module 106 is connected with the result output module 107.
The inventory data input module 101 is configured to obtain an inventory data set, and input the obtained inventory data set into the inventory scenario simulator generation module 102 and the algorithm modification module 105, respectively. The inventory scenario simulator generation module 102 is configured to generate a plurality of inventory scenario simulators from the inventory data set. The algorithm selection module 104 is configured to select one algorithm from the plurality of algorithms and input the selected algorithm into the algorithm transformation module 105, and the algorithm transformation module 105 is configured to transform the selected algorithm according to the inventory data set to obtain an initial inventory control policy including the scene parameters, and input the initial inventory control policy into the meta-learning training module 106. The inventory scenario simulator sampling module 103 is configured to determine a training sample for each inventory scenario simulator according to the inventory scenario simulator generated by the inventory scenario simulator generation module 102, and input the determined training sample to the meta-learning training module 106, where the meta-learning training module 106 is configured to optimize an initial inventory control strategy according to the training sample for each inventory scenario simulator, input an optimized target inventory control strategy to the result output module 107, and display the target inventory control strategy to a user by the result output module 107.
Through the cooperation among the above modules, the method for determining the inventory control strategy provided in the application embodiment can be implemented, wherein the detailed functions of the above modules will be explained in the following embodiments, which will not be explained herein. In addition, the name of each module is named only according to the function executed by the module, and in practical applications, other names may be defined for each module according to the function executed by the module, for example, the meta-learning training module may also be defined as a scene parameter optimization module, and the embodiment of the present application is not specifically limited herein.
Fig. 2 is a schematic structural diagram of a computer device according to an embodiment of the present application. The system for determining inventory control policies of FIG. 1 may be implemented by a computer device as shown in FIG. 2. Referring to fig. 2, the computer device comprises at least one processor 201, a communication bus 202, a memory 203 and at least one communication interface 204.
The processor 201 may be a Central Processing Unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of programs according to the present disclosure.
The communication bus 202 may include a path that conveys information between the aforementioned components.
The Memory 203 may be a Read-Only Memory (ROM) or other type of static storage device that can store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that can store information and instructions, an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Compact Disc Read-Only Memory (CD-ROM) or other optical Disc storage, optical Disc storage (including Compact Disc, laser Disc, optical Disc, digital versatile Disc, blu-ray Disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to these. The memory 203 may be self-contained and coupled to the processor 201 via the communication bus 202. The memory 203 may also be integrated with the processor 201.
In particular implementations, a computer device may include multiple processors, such as processor 201 and processor 205 shown in fig. 2, as one embodiment. Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).
In particular implementations, the computer device may also include an output device 206 and an input device 207, as one embodiment. The output device 206 is in communication with the processor 201 and may display information in a variety of ways. For example, the output device 206 may be a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display device, a Cathode Ray Tube (CRT) display device, a projector (projector), or the like. The input device 207 is in communication with the processor 201 and may receive user input in a variety of ways. For example, the input device 207 may be a mouse, a keyboard, a touch screen device, or a sensing device, among others.
The computer device may be a general purpose computer device or a special purpose computer device. In a specific implementation, the computer device may be a desktop computer, a laptop computer, a network server, a Personal Digital Assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiment of the invention does not limit the type of the computer equipment.
The memory 203 is used for storing program codes for executing the scheme of the application, and the processor 201 controls the execution. The processor 201 is used to execute program code stored in the memory 203. One or more software modules (e.g., inventory data input module, inventory scenario simulator generation module, inventory scenario simulator sampling module, algorithm selection module, algorithm transformation module, meta-learning training module, and result output module, etc.) may be included in the program code. The system for determining inventory control policies shown in FIG. 1 may determine data for developing an application by one or more software modules in program code in processor 201 and memory 203.
Fig. 3 is a flowchart of a method for determining an inventory control policy according to an embodiment of the present application, where as shown in fig. 3, the method includes the following steps:
step 301: an inventory data set is obtained, the inventory data set including data describing historical inventory information for at least one product.
Since the method for determining the inventory control policy provided by the embodiment of the present application is to provide an inventory control policy suitable for the inventory condition of a client, an implementation manner for acquiring the inventory data set may be: the method comprises the steps of receiving an inventory data set uploaded by a client and storing the inventory data set so as to provide an inventory control strategy for the client according to data in the inventory data set. The inventory data set includes data of historical inventory information for at least one product, for example, the inventory data set includes historical demand data of each product in the at least one product, historical shelf life information of each product, and a ratio between a historical inventory holding cost and a historical inventory stock shortage cost of each product.
In an embodiment of the application, the historical demand data of each product in the at least one product includes ex-warehouse information of each product at each time before the current time. For example, in one possible implementation, the historical demand data for each product in the at least one product may be represented by the following export matrix Y:
wherein, yijThe ex-warehouse quantity of the product i at the time point of j is represented, i is a positive integer which is greater than or equal to 1 and less than or equal to n, and j is a positive integer which is greater than or equal to 1 and less than or equal to t.
The historical date information of each product comprises the historical delivery time information of the product, for example, the historical date information of the product i can comprise the delivery time information of the product i within one year before the current time. Because a large amount of data is needed in the subsequent training process, when the quantity of the existing actual product i historical supply time information is large, the existing actual product i historical supply time information can be directly used as the product i historical shelf information; when the quantity of the existing actual product i historical supply time information is small, some supply time information can be regenerated according to the existing actual product i historical supply time information, and then the regenerated supply time information and the existing actual product i historical supply time information are jointly used as the product i historical goods date information.
In addition, since different product attributes may differ, the ratio between the historical inventory holding cost and the historical inventory backorder cost may also differ for different products. In a specific application, if actual data of the ratio between the historical inventory holding cost and the historical inventory shortage cost for each product is lacked, the ratio between the historical inventory holding cost and the historical inventory shortage cost for each product may be set to be a preset value, where the preset value may be 40, and the embodiment of the present application is not limited specifically herein.
Step 302: and generating a plurality of inventory scenario simulators according to the inventory data set, wherein each inventory scenario simulator corresponds to one inventory scenario.
In the embodiment of the application, in order to enable the determined inventory control policy to be applicable to various possible inventory scenarios, a plurality of inventory scenario simulators may be generated according to the inventory data set, and then the inventory control policy may be determined according to the plurality of inventory scenario simulators, so that the determined inventory control policy may be simultaneously applicable to a plurality of inventory scenarios corresponding to the plurality of inventory scenario simulators.
In addition, since the inventory data set includes data describing inventory information of at least one product, and different inventory scenarios of different products may correspond to multiple inventory scenarios, in the present application, a corresponding inventory scenario simulator is determined for each product in the inventory data set, and there may be multiple inventory scenario simulators corresponding to each product, and the determined inventory scenario simulators corresponding to each product are taken as multiple inventory scenario simulators in step 302.
Since the implementation manners of determining the corresponding inventory scenario simulator for each product in the inventory data set are basically the same, the implementation manner of determining the inventory scenario simulator corresponding to the product B is defined below by taking the product B as an example, and the implementation manners of determining the inventory scenario simulators corresponding to other products are not described herein again.
Specifically, the implementation manner of determining the inventory scenario simulator corresponding to the product B may be: for any product B in at least one product, determining a historical demand distribution curve of the product B according to historical demand data of the product B in the inventory data set, and generating a plurality of first-type inventory scene simulators for the product B according to the historical demand distribution curve of the product B, wherein each first-type inventory scene simulator corresponds to one demand scene. Determining a historical delivery time distribution curve of the product B according to historical delivery time information of the product B in the inventory data set, and generating a plurality of second-class inventory scene simulators for the product B according to the historical delivery time distribution curve of the product B, wherein each second-class inventory scene simulator corresponds to one delivery time scene. That is, in the embodiment of the present application, for any product, a plurality of different demand scenarios and a plurality of different supply time scenarios may be simulated according to the inventory data set.
At this time, the plurality of inventory scenario simulators in step 302 includes: a plurality of first type inventory scenario simulators and a plurality of second type inventory scenario simulators for each product in at least one product.
The implementation manner of determining the historical demand distribution curve of the product B according to the historical demand data of the product B in the inventory data set is as follows: and analyzing historical demand data of the product B from the inventory data set, mapping the analyzed historical demand data to a coordinate corresponding to time and demand to obtain a plurality of points in the coordinate, and fitting the plurality of points in the coordinate to obtain a historical demand distribution curve of the product B. In addition, the implementation manner of determining the historical delivery time distribution curve of the product B according to the historical delivery date information of the product B in the inventory data set can also refer to the implementation manner, and a description thereof is not repeated here.
It should be noted that, if the data uploaded by the client further includes the historical demand distribution curve and the historical delivery time curve of the product B that has already been manufactured, it is not necessary to determine the historical demand distribution curve and the historical delivery time distribution curve of the product B again, and it is only necessary to directly determine the plurality of first-type inventory scene simulators and the plurality of second-type inventory scene simulators for the product B by using the historical demand distribution curve and the historical delivery time curve of the product B uploaded by the client, respectively.
In addition, in a possible implementation manner, the generating a plurality of first-type inventory scenario simulators for the product B according to the historical demand distribution curve of the product B may be: determining the mean value and the standard deviation of the historical demand distribution curve of the product B; adjusting the mean value and/or standard deviation of the historical demand distribution curve of the product B to obtain a plurality of demand distribution curves; and generating a first-class inventory scene simulator according to each demand distribution curve to obtain a plurality of first-class inventory scene simulators for the product B.
Because each demand distribution curve can describe a demand scenario, and in addition, any curve can be characterized by adopting a corresponding algorithm model, an implementation manner for generating a first-type inventory scenario simulator according to each demand distribution curve can be as follows: and analyzing each demand distribution curve to obtain an algorithm model capable of representing each demand distribution curve, and using the obtained algorithm model as a first-class inventory scene simulator. As such, each of the first type inventory scenario simulators may describe a demand scenario. In addition, since the mean value can represent the amount of demand and the standard deviation can represent the fluctuation of the demand, a plurality of first-class inventory scenario simulators for the product B can simulate various scenarios of different demands of the product B, or simulate scenarios of similar demands but different fluctuations.
In addition, the mean and/or standard deviation of the historical demand distribution curve of the product B may be: only the mean value of the historical demand distribution curve of the product B is adjusted, or only the standard deviation of the historical demand distribution curve of the product B is adjusted, or both the mean value and the standard deviation of the historical demand distribution curve of the product B are adjusted, which is not limited herein.
For example, if the determined average value of the historical demand profile of product B is 200 and the standard deviation is 0.3, the average values of the historical demand profiles of product B can be adjusted to 150, 175, 225, and 250 and the standard deviation is still 0.3, so that 4 additional demand profiles can be obtained. Alternatively, product B may be adjusted to have historical demand profiles with standard deviations of 0.1, 0.15, 0.2, 0.25, 0.35, 0.4, 0.45, and 0.5, and still have a mean of 200, so that another 8 demand profiles may be obtained.
In addition, in a possible implementation manner, the implementation manner of generating the plurality of second-class inventory scenario simulators for the product B according to the historical delivery time distribution curve of the product B may be: determining the mean value and the standard deviation of the historical supply time distribution curve of the product B; adjusting the mean value and/or standard deviation of the historical supply time distribution curve of the product B to obtain a plurality of supply time distribution curves; and generating a second-class inventory scene simulator according to each supply time distribution curve to obtain a plurality of second-class inventory scene simulators for the product B.
Similarly, since each sourcing time profile can describe a sourcing time scenario, the implementation of generating a second type library scenario simulator according to each sourcing time profile may be: and analyzing each supply time distribution curve to obtain an algorithm model capable of representing each supply time distribution curve, and taking the obtained algorithm model as a second-class inventory scene simulator. In this way, each second-class inventory scenario simulator may describe a supply-time scenario for product B. For example, one of the second-type inventory scenario simulators may describe a scenario in which the sourcing period is fixed, and the other second-type inventory scenario simulator may describe a scenario in which the sourcing period is distributed according to gaussian, and so on.
The manner of adjusting the mean and/or standard deviation of the historical supply time distribution curve of the product B is substantially the same as the manner of adjusting the mean and/or standard deviation of the historical demand distribution curve, and will not be described herein.
Step 303: an initial inventory control policy including scenario parameters is determined from the inventory data set.
In a possible implementation manner, step 303 may specifically be: the method comprises the steps of obtaining an initial model aiming at inventory control, determining historical demand characteristics and historical delivery time characteristics of each product in at least one product according to an inventory data set, and updating the initial model according to the historical demand characteristics and the historical delivery time characteristics of each product in at least one product to obtain an initial inventory control strategy comprising scene parameters.
The implementation manner of obtaining the initial model for inventory control may be: and randomly selecting one basic operation research algorithm from the multiple basic operation research algorithms, and taking the selected basic operation research algorithm as an initial model. Optionally, the implementation of obtaining the initial model for inventory control may also be: and displaying a plurality of basic operation research algorithms, and when a selection operation for one of the basic operation research algorithms is detected, using the selected basic operation research algorithm as an initial model, wherein the selection operation is triggered by a user through a preset operation. That is, in this embodiment of the application, the obtained initial model may be a basic operation research algorithm randomly selected by the terminal, or refer to a basic operation research algorithm selected by the user.
In addition, the implementation manner of determining the historical demand characteristics and the historical delivery time characteristics of each product in the at least one product according to the inventory data set may refer to the implementation manner of determining the historical demand distribution curve and the historical delivery time distribution curve of the product B in step 302, which is not described in detail herein.
In addition, the implementation manner of updating the initial model according to the historical demand characteristics and the historical supply time characteristics of each product in the at least one product to obtain the initial inventory control strategy including the scene parameters may be: determining parameters capable of influencing demand distribution or supply time distribution according to historical demand characteristics and historical supply time characteristics of each product in at least one product, and then adding the determined parameters into the initial model to obtain an initial inventory control strategy comprising scene parameters.
For example, in the embodiment of the present application, a cycle-based online convex optimization algorithm is used as the obtained basic operations research algorithm, and the online convex optimization algorithm may be expressed as follows:
wherein G represents total inventory cost, h represents unit cost of stockpiling, p represents unit cost of out-of-stock, BS is optimum inventory amount, ItIs the stock at time t, ItIs a parameter related to the supply time, It' (BS) is the derivative of inventory to BS, DtIs the demand at time t.
Through analyzing the historical demand characteristics and the historical supply time characteristics of each product in at least one product, the gradient parameter alpha can influence the demand distribution or the supply time distribution. Therefore, after the gradient parameter α is taken as a scene parameter and is added to the above formula, an initial inventory control strategy can be obtained, which can be expressed by the following formula:
in the above, the gradient parameter α is taken as an example for explanation, and in specific application, other parameters may be taken as scene parameters according to analysis results of historical demand characteristics and historical supply time characteristics of each product in at least one product, and are not described in detail herein.
In addition, the above description is given by taking the cycle-based online convex optimization algorithm as the obtained basic operation research algorithm as an example, and in a specific application, when another algorithm is selected as the basic operation research algorithm, the basic operation research algorithm may still be modified in the above manner to obtain an initial inventory control strategy, which is not described in detail herein.
Step 304: and optimizing the scene parameters in the initial inventory control strategy according to the plurality of inventory scene simulators to obtain a target inventory control strategy, wherein the target inventory control strategy is used for adjusting the inventory.
Step 304 can be implemented by the following two steps:
(1) and determining a corresponding relation between a scene parameter value associated with each inventory scene simulator and the inventory cost according to each inventory scene simulator and the initial inventory control strategy.
In a possible implementation manner, the step (1) may specifically be: for any inventory scene simulator A in the inventory scene simulators, generating a plurality of sample data based on the inventory scene simulator A, wherein each sample data is used for describing the demand information and the supply time information of a product; and inputting a plurality of pieces of sample data into the initial inventory control strategy to obtain the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory cost.
As can be seen from step 302, each inventory scenario simulator is an algorithm model generated according to a demand distribution curve or a supply time distribution curve of a product, and therefore, a plurality of sample data may be generated according to each inventory scenario simulator, where each sample data may include one demand information or one supply time information.
In addition, since the initial inventory control policy is an algorithm that can determine the inventory cost according to the demand information or the supply time information, for any sample data, when the sample data is input to the initial inventory control model, the inventory cost for the sample data can be obtained, and the obtained inventory cost is expressed in a manner including scene parameters. For example, sample data is represented by x, inventory cost is represented by y, and when the operation is performed on multiple pieces of sample data, the following correspondence relationship can be obtained: y is1＝f(x1,α)、y2＝f(x2,α)、…、yn＝f(xnα). Wherein n isThe number of pieces of sample data, α, is a scene parameter. From this correspondence, a correspondence between the inventory cost y and the scene parameter α value can be determined, labeled as y ═ g (α). When the above operations are performed on each of the plurality of inventory scenario simulators generated in step 302, a correspondence between the inventory-out cost y and the scenario parameter α value for each inventory scenario simulator may be obtained.
(2) And optimizing the scene parameters according to the corresponding relation between the plurality of scene parameter values and the inventory cost which are associated with the plurality of inventory scene simulators one by one to obtain a target inventory control strategy.
In a possible implementation manner, the step (2) may specifically be: determining a scene parameter extreme value associated with each inventory scene simulator according to the corresponding relationship between the scene parameter value associated with each inventory scene simulator and the inventory cost, wherein the inventory cost corresponding to the scene parameter extreme value is the lowest in the corresponding relationship between the scene parameter value associated with each inventory scene simulator and the inventory cost; and determining the average value of a plurality of scene parameter extreme values which are associated with a plurality of inventory scene simulators one by one, taking the determined average value as the optimized value of the scene parameter, and replacing the scene parameter in the initial inventory control strategy with the optimized value to obtain the target inventory control strategy.
For example, if the extreme values of the plurality of scene parameters associated with the plurality of inventory scene simulators one by one are labeled as α 1, α 2, α 3, and α 2, respectively, then the implementation manner of determining the optimized values of the scene parameters may be: and determining the average value of alpha 1, alpha 2, alpha 3 and alpha 2 as the optimized value of the scene parameter.
In the above implementation, the average value of the extreme values of the plurality of scene parameters is used as the optimized value of the scene parameter. Of course, the extreme values of the plurality of scene parameters may be processed in other manners to determine the optimized value of the scene parameter, which is not specifically limited herein.
As can be appreciated in connection with the system of FIG. 1, the inventory data input module is configured to determine the set of inventory data via step 301 described above. The inventory scenario simulator generation module is configured to determine a plurality of inventory scenario simulators based on the inventory data set output by the inventory data input module, via step 302 described above. The algorithm selection module is used for selecting one basic operation research algorithm from a plurality of basic operation research algorithms as an initial model. The algorithm transformation module is used for determining an initial inventory control strategy comprising scene parameters according to the initial model output by the algorithm selection module and the inventory data set output by the inventory data input module. The sampling module of the inventory scene simulator is used for generating a plurality of pieces of sample data according to each inventory scene simulator and inputting the plurality of pieces of sample data corresponding to each inventory scene simulator to the meta-learning training module. And the meta-learning training module is used for determining an optimized value of a scene parameter according to a plurality of pieces of sample data corresponding to each inventory scene simulator, updating an initial inventory control strategy according to the optimized value of the scene parameter to obtain a target inventory control strategy, and inputting the target inventory control strategy to the result output module. And the result output module is used for displaying the target inventory control strategy to the user.
In addition, after the target inventory control strategy is determined, if the target inventory control strategy is determined according to the above loop-based online convex optimization algorithm, the target inventory control strategy also belongs to an online iterative algorithm, so that the target inventory control strategy is continuously optimized subsequently, a specific process of each optimization is substantially the same as the process of determining the target inventory control strategy, but data in an inventory data set adopted in the specific process of each optimization is different, for example, each time, inventory data of the last month is adopted to optimize the target inventory control strategy, and a description thereof is omitted.
In the embodiment of the application, a plurality of inventory scene simulators are generated according to an inventory data set, and each inventory scene simulator corresponds to one inventory scene; and according to the inventory data set, determining an initial inventory control strategy comprising scene parameters, and optimizing the scene parameters in the initial inventory control strategy according to a plurality of inventory scene simulators to obtain a target inventory control strategy. Because the scene parameters in the target inventory control strategy are obtained after optimization based on the multiple inventory scene simulators, the target inventory control strategy can be simultaneously applied to multiple inventory scenes, so that the flexibility of the target inventory control strategy is improved, and the accuracy of adjusting the inventory through the target inventory control strategy can be improved.
In order to further explain the beneficial effects of the method for determining the inventory control strategy provided by the embodiment of the present application, the method for determining the inventory control strategy provided by the embodiment of the present application is experimentally verified, and the experimental verification process is as follows:
for the XX client, the data of historical inventory information describing each product between 12 months in 2013 and 2016 and 12 months in 12 are divided into two parts, one part is used as a training set, and the other part is used as a testing set. The training set is used as the inventory data set and the target inventory control strategy is determined according to the various steps in the embodiment illustrated in FIG. 3 above. Wherein, aiming at each product in the training set, half of the product supply periods can be set as fixed values, and the other half of the product supply periods are randomly distributed. After the target inventory control strategy is determined, the target inventory control strategy and the inventory control strategies provided in the related art are simulated according to the data in the training set, respectively, and a simulation result as shown in fig. 4 is obtained. As shown in fig. 4, the inventory cost of the client is 6427.9 according to the target inventory control strategy, the inventory cost determined according to the inventory control strategy in the related art is 7622.4, and the theoretical optimal inventory cost is 4954.6, and obviously, the inventory cost determined by the target inventory control strategy is closer to the theoretical optimal inventory cost.
After the target inventory control strategy is determined, the target inventory control strategy may also be tested using the data in the test set. Specifically, the target inventory control policy and the inventory control policy provided in the related art are simulated according to the data in the test set, respectively, to obtain a simulation result as shown in fig. 5. As shown in fig. 5, the inventory cost of the client determined according to the target inventory control strategy is 6341.3, the inventory cost determined according to the test set according to the related art is 6912.1, and the theoretical optimal inventory cost is 4835.2, and it is obvious that the inventory cost determined according to the test set through the target inventory control strategy is closer to the theoretical optimal inventory cost. And for the target inventory control strategy, the difference between the inventory cost determined by the test set and the inventory cost determined by the training set is not large, which indicates that the over-fitting or under-fitting condition does not exist in the training process for determining the target inventory control strategy basically.
Fig. 6 is an apparatus for determining an inventory control policy according to an embodiment of the present application, where, as shown in fig. 6, the apparatus 600 includes:
an obtaining module 601, configured to execute step 301 in the embodiment of fig. 3;
a generating module 602, configured to perform step 302 in the embodiment of fig. 3;
a determining module 603, configured to perform step 303 in the embodiment of fig. 3;
an optimization module 604 for performing step 304 in the embodiment of fig. 3.
Optionally, the optimization module 604 comprises:
the first determining unit is used for determining the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory cost according to each inventory scene simulator and the initial inventory control strategy;
and the second determining unit is used for optimizing the scene parameters according to the corresponding relation between the plurality of scene parameter values and the inventory cost, which are associated with the plurality of inventory scene simulators one by one, so as to obtain the target inventory control strategy.
Optionally, the first determining unit is specifically configured to:
for any inventory scene simulator A in the inventory scene simulators, generating a plurality of sample data based on the inventory scene simulator A, wherein each sample data is used for describing the demand information or supply time information of a product;
and inputting a plurality of pieces of sample data into the initial inventory control strategy to obtain the corresponding relation between the scene parameter value associated with the inventory scene simulator A and the inventory cost.
Optionally, the second determining unit is specifically configured to:
determining a scene parameter extreme value associated with each inventory scene simulator according to the corresponding relationship between the scene parameter value associated with each inventory scene simulator and the inventory cost, wherein the inventory cost corresponding to the scene parameter extreme value is the lowest in the corresponding relationship between the scene parameter value associated with each inventory scene simulator and the inventory cost;
and determining the average value of a plurality of scene parameter extreme values which are associated with a plurality of inventory scene simulators one by one, and taking the determined average value as the optimized value of the scene parameter to obtain the target inventory control strategy.
Optionally, the inventory data set includes historical demand data and historical shelf information for describing each of the at least one product;
the generating module 602 includes:
a third determining unit, configured to determine, for any product B in the at least one product, a historical demand distribution curve of the product B according to historical demand data of the product B in the inventory data set, and generate, according to the historical demand distribution curve of the product B, a plurality of first-type inventory scenario simulators for the product B, where each first-type inventory scenario simulator corresponds to one demand scenario;
the fourth determining unit is used for determining a historical delivery time distribution curve of the product B according to the historical delivery date information of the product B in the inventory data set, and generating a plurality of second-class inventory scene simulators for the product B according to the historical delivery time distribution curve of the product B, wherein each second-class inventory scene simulator corresponds to one delivery time scene;
wherein the plurality of inventory scenario simulators comprise: a plurality of first type inventory scenario simulators and a plurality of second type inventory scenario simulators for each product in at least one product.
Optionally, the third determining unit is specifically configured to:
determining the mean value and the standard deviation of the historical demand distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical demand distribution curve of the product B to obtain a plurality of demand distribution curves;
and generating a first-class inventory scene simulator according to each demand distribution curve to obtain a plurality of first-class inventory scene simulators for the product B.
Optionally, the fourth determining unit is specifically configured to:
determining the mean value and the standard deviation of the historical supply time distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical supply time distribution curve of the product B to obtain a plurality of supply time distribution curves;
and generating a second-class inventory scene simulator according to each supply time distribution curve to obtain a plurality of second-class inventory scene simulators for the product B.
Optionally, the determining module 603 is specifically configured to:
acquiring an initial model for inventory control;
determining historical demand characteristics and historical supply time characteristics of each product in at least one product according to the inventory data set;
and updating the initial model according to the historical demand characteristics and the historical supply time characteristics of each product in the at least one product to obtain an initial inventory control strategy comprising scene parameters.
In the embodiment of the application, a plurality of inventory scene simulators are generated according to an inventory data set, and each inventory scene simulator corresponds to one inventory scene; and according to the inventory data set, determining an initial inventory control strategy comprising scene parameters, and optimizing the scene parameters in the initial inventory control strategy according to a plurality of inventory scene simulators to obtain a target inventory control strategy. Because the scene parameters in the target inventory control strategy are obtained after optimization based on the multiple inventory scene simulators, the target inventory control strategy can be simultaneously applied to multiple inventory scenes, so that the flexibility of the target inventory control strategy is improved, and the accuracy of adjusting the inventory through the target inventory control strategy can be improved.
It should be noted that: in the apparatus for determining an inventory control policy according to the foregoing embodiment, when determining an inventory control policy, only the division of the functional modules is illustrated, and in practical applications, the function distribution may be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the apparatus for determining an inventory control policy and the method for determining an inventory control policy provided in the above embodiments belong to the same concept, and specific implementation processes thereof are described in the method embodiments and are not described herein again.
In the above embodiments, the implementation may be wholly or partly realized by software, hardware, firmware, or any combination thereof. When implemented in software, may 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. When loaded and executed on a computer, cause the processes or functions described in accordance with embodiments of the invention, to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., Digital Versatile Disk (DVD)), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above-mentioned embodiments are provided not to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (16)
1. A method of determining an inventory control policy, the method comprising:
obtaining an inventory data set including data describing historical inventory information for at least one product;
generating a plurality of inventory scene simulators according to the inventory data set, wherein each inventory scene simulator corresponds to one inventory scene;
determining an initial inventory control strategy comprising scene parameters according to the inventory data set;
determining a corresponding relation between a scene parameter value associated with each inventory scene simulator and inventory cost according to each inventory scene simulator and the initial inventory control strategy;
and optimizing the scene parameters in the initial inventory control strategy according to the corresponding relation between the plurality of scene parameter values and the inventory cost which are associated with the plurality of inventory scene simulators one by one to obtain a target inventory control strategy, wherein the target inventory control strategy is used for adjusting the inventory.
2. The method of claim 1, wherein determining a correspondence between a scenario parameter value associated with each inventory scenario simulator and an inventory cost as a function of each inventory scenario simulator and the initial inventory control policy comprises:
for any inventory scene simulator A in the inventory scene simulators, generating a plurality of sample data based on the inventory scene simulator A, wherein each sample data is used for describing the demand information or supply time information of a product;
and inputting the plurality of sample data into the initial inventory control strategy to obtain a corresponding relation between a scene parameter value associated with the inventory scene simulator A and inventory cost.
3. The method of claim 1, wherein optimizing the scenario parameters in the initial inventory control strategy according to the correspondence between the inventory costs and the scenario parameter values associated with the inventory scenario simulators one to one, to obtain a target inventory control strategy, comprises:
determining a scene parameter extreme value associated with each inventory scene simulator according to a corresponding relation between a scene parameter value associated with each inventory scene simulator and inventory costs, wherein the inventory cost corresponding to the scene parameter extreme value is the lowest in the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory costs;
and determining an average value of a plurality of scene parameter extreme values which are associated with the plurality of inventory scene simulators one by one, and taking the determined average value as an optimized value of the scene parameter to obtain the target inventory control strategy.
4. The method of claim 1, wherein the set of inventory data includes historical demand data and historical shelf information for describing each of the at least one product;
the generating a plurality of inventory scenario simulators from the inventory data set includes:
for any product B in the at least one product, determining a historical demand distribution curve of the product B according to historical demand data of the product B in the inventory data set, and generating a plurality of first-type inventory scene simulators for the product B according to the historical demand distribution curve of the product B, wherein each first-type inventory scene simulator corresponds to a demand scene;
determining a historical delivery time distribution curve of the product B according to the historical delivery information of the product B in the inventory data set, and generating a plurality of second-type inventory scene simulators for the product B according to the historical delivery time distribution curve of the product B, wherein each second-type inventory scene simulator corresponds to one delivery time scene;
wherein the plurality of inventory scenario simulators comprise: a plurality of first type inventory scenario simulators and a plurality of second type inventory scenario simulators for each product in the at least one product.
5. The method of claim 4, wherein said generating a plurality of first type inventory scenario simulators for said product B based on said product B's historical demand profile comprises:
determining the mean and standard deviation of the historical demand distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical demand distribution curve of the product B to obtain a plurality of demand distribution curves;
and generating a first-class inventory scene simulator according to each demand distribution curve to obtain a plurality of first-class inventory scene simulators for the product B.
6. The method of claim 4, wherein generating a plurality of second-class inventory scenario simulators for the product B according to the historical sourcing time profile of the product B comprises:
determining the mean value and the standard deviation of the historical supply time distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical supply time distribution curve of the product B to obtain a plurality of supply time distribution curves;
and generating a second-class inventory scene simulator according to each supply time distribution curve to obtain a plurality of second-class inventory scene simulators for the product B.
7. The method of any of claims 1 to 6, wherein said determining an initial inventory control policy including scenario parameters from said inventory data set comprises:
acquiring an initial model for inventory control;
determining historical demand characteristics and historical supply time characteristics of each product in the at least one product according to the inventory data set;
and updating the initial model according to the historical demand characteristics and the historical supply time characteristics of each product in the at least one product to obtain an initial inventory control strategy comprising scene parameters.
8. An apparatus for determining an inventory control policy, the apparatus comprising:
an acquisition module to acquire an inventory data set including data describing historical inventory information for at least one product;
the generating module is used for generating a plurality of inventory scene simulators according to the inventory data set, and each inventory scene simulator corresponds to one inventory scene;
a determining module for determining an initial inventory control strategy comprising scene parameters according to the inventory data set;
the optimization module comprises a first determination unit and a second determination unit, wherein the first determination unit is used for determining the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory cost according to each inventory scene simulator and the initial inventory control strategy; the second determining unit is configured to optimize the scene parameters in the initial inventory control policy according to a correspondence between a plurality of scene parameter values and inventory costs associated with the plurality of inventory scene simulators one to one, so as to obtain a target inventory control policy, where the target inventory control policy is used to adjust inventory.
9. The apparatus of claim 8, wherein the first determining unit is specifically configured to:
for any inventory scene simulator A in the inventory scene simulators, generating a plurality of sample data based on the inventory scene simulator A, wherein each sample data is used for describing the demand information or supply time information of a product;
and inputting the plurality of sample data into the initial inventory control strategy to obtain a corresponding relation between a scene parameter value associated with the inventory scene simulator A and inventory cost.
10. The apparatus of claim 8, wherein the second determining unit is specifically configured to:
determining a scene parameter extreme value associated with each inventory scene simulator according to a corresponding relation between a scene parameter value associated with each inventory scene simulator and inventory costs, wherein the inventory cost corresponding to the scene parameter extreme value is the lowest in the corresponding relation between the scene parameter value associated with each inventory scene simulator and the inventory costs;
and determining an average value of a plurality of scene parameter extreme values which are associated with the plurality of inventory scene simulators one by one, and taking the determined average value as an optimized value of the scene parameter to obtain the target inventory control strategy.
11. The apparatus of claim 8, wherein the set of inventory data includes historical demand data and historical shelf information for describing each of the at least one product;
the generation module comprises:
a third determining unit, configured to determine, for any product B in the at least one product, a historical demand distribution curve of the product B according to historical demand data of the product B in the inventory data set, and generate, according to the historical demand distribution curve of the product B, a plurality of first-type inventory scenario simulators for the product B, where each first-type inventory scenario simulator corresponds to one demand scenario;
a fourth determining unit, configured to determine a historical delivery time distribution curve of the product B according to the historical delivery date information of the product B in the inventory data set, and generate a plurality of second-class inventory scene simulators for the product B according to the historical delivery time distribution curve of the product B, where each second-class inventory scene simulator corresponds to a delivery time scene;
wherein the plurality of inventory scenario simulators comprise: a plurality of first type inventory scenario simulators and a plurality of second type inventory scenario simulators for each product in the at least one product.
12. The apparatus of claim 11, wherein the third determining unit is specifically configured to:
determining the mean and standard deviation of the historical demand distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical demand distribution curve of the product B to obtain a plurality of demand distribution curves;
and generating a first-class inventory scene simulator according to each demand distribution curve to obtain a plurality of first-class inventory scene simulators for the product B.
13. The apparatus of claim 11, wherein the fourth determining unit is specifically configured to:
determining the mean value and the standard deviation of the historical supply time distribution curve of the product B;
adjusting the mean value and/or standard deviation of the historical supply time distribution curve of the product B to obtain a plurality of supply time distribution curves;
and generating a second-class inventory scene simulator according to each supply time distribution curve to obtain a plurality of second-class inventory scene simulators for the product B.
14. The apparatus according to any one of claims 8 to 13, wherein the determining module is specifically configured to:
acquiring an initial model for inventory control;
determining historical demand characteristics and historical supply time characteristics of each product in the at least one product according to the inventory data set;
and updating the initial model according to the historical demand characteristics and the historical supply time characteristics of each product in the at least one product to obtain an initial inventory control strategy comprising scene parameters.
15. An apparatus for determining an inventory control policy, the apparatus comprising a memory and a processor;
the memory for storing a program that enables the apparatus to perform the method of any of claims 1-7, the processor configured to execute the program stored in the memory.
16. A computer-readable storage medium having instructions stored therein, which when executed on a computer, cause the computer to perform the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810972843.6A CN109902850B (en) | 2018-08-24 | 2018-08-24 | Method, device and storage medium for determining inventory control strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810972843.6A CN109902850B (en) | 2018-08-24 | 2018-08-24 | Method, device and storage medium for determining inventory control strategy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109902850A CN109902850A (en) | 2019-06-18 |
| CN109902850B true CN109902850B (en) | 2021-08-20 |
Family
ID=66943273
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810972843.6A Active CN109902850B (en) | 2018-08-24 | 2018-08-24 | Method, device and storage medium for determining inventory control strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109902850B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132498A (en) * | 2019-06-25 | 2020-12-25 | 北京沃东天骏信息技术有限公司 | Inventory management method, device, equipment and storage medium |
| CN114118503A (en) * | 2020-08-26 | 2022-03-01 | 上海顺如丰来技术有限公司 | Supply chain inventory optimization method, device, equipment and storage medium |
| CN113762582A (en) * | 2021-01-29 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Data processing method and device for determining goods returning amount based on inventory simulation |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1615487A (en) * | 2002-01-23 | 2005-05-11 | 雀巢技术公司 | Systems and methods for inventory management |
| CN1643731A (en) * | 2002-01-23 | 2005-07-20 | 米德韦斯瓦科公司 | Inventory management system |
| CN102142120A (en) * | 2010-02-01 | 2011-08-03 | 台湾积体电路制造股份有限公司 | Methods and systems for dynamic inventory control |
| CN102385724A (en) * | 2010-08-27 | 2012-03-21 | 上海财经大学 | Spare part assembling demand forecasting information processing method applied to inventory management |
| CN103729693A (en) * | 2013-12-23 | 2014-04-16 | 清华大学 | Maintenance and spare part supply combined optimization method based on deterministic inventory degradation model |
| CN104881764A (en) * | 2015-02-13 | 2015-09-02 | 四川大学 | Large-scale construction project multi-material ratio dynamic purchasing and inventory control method |
| CN106096748A (en) * | 2016-04-28 | 2016-11-09 | 武汉宝钢华中贸易有限公司 | Entrucking forecast model in man-hour based on cluster analysis and decision Tree algorithms |
| US20170323274A1 (en) * | 2016-05-06 | 2017-11-09 | General Electric Company | Controlling aircraft operations and aircraft engine components assignment |
| CN107368917A (en) * | 2017-06-21 | 2017-11-21 | 国网辽宁省电力有限公司 | A kind of Power Material inventory optimization system and method based on KNN algorithms |
-
2018
- 2018-08-24 CN CN201810972843.6A patent/CN109902850B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1615487A (en) * | 2002-01-23 | 2005-05-11 | 雀巢技术公司 | Systems and methods for inventory management |
| CN1643731A (en) * | 2002-01-23 | 2005-07-20 | 米德韦斯瓦科公司 | Inventory management system |
| CN102142120A (en) * | 2010-02-01 | 2011-08-03 | 台湾积体电路制造股份有限公司 | Methods and systems for dynamic inventory control |
| CN105825322A (en) * | 2010-02-01 | 2016-08-03 | 台湾积体电路制造股份有限公司 | Methods and systems for dynamic inventory control |
| CN102385724A (en) * | 2010-08-27 | 2012-03-21 | 上海财经大学 | Spare part assembling demand forecasting information processing method applied to inventory management |
| CN103729693A (en) * | 2013-12-23 | 2014-04-16 | 清华大学 | Maintenance and spare part supply combined optimization method based on deterministic inventory degradation model |
| CN104881764A (en) * | 2015-02-13 | 2015-09-02 | 四川大学 | Large-scale construction project multi-material ratio dynamic purchasing and inventory control method |
| CN106096748A (en) * | 2016-04-28 | 2016-11-09 | 武汉宝钢华中贸易有限公司 | Entrucking forecast model in man-hour based on cluster analysis and decision Tree algorithms |
| US20170323274A1 (en) * | 2016-05-06 | 2017-11-09 | General Electric Company | Controlling aircraft operations and aircraft engine components assignment |
| CN107368917A (en) * | 2017-06-21 | 2017-11-21 | 国网辽宁省电力有限公司 | A kind of Power Material inventory optimization system and method based on KNN algorithms |
Non-Patent Citations (2)
| Title |
|---|
| G家电配送中心库存成本控制策略研;江世英;《会计之友》;20150410;全文 * |
| 不确定条件下外协生产库存***控制策略与参数优化;曹乐;《重庆大学学报》;20121215;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109902850A (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220343172A1 (en) | Dynamic, automated fulfillment of computer-based resource request provisioning using deep reinforcement learning | |
| CN110263938B (en) | Method and apparatus for generating information | |
| CN109902850B (en) | Method, device and storage medium for determining inventory control strategy | |
| CN113869801B (en) | Maturity state evaluation method and device for enterprise digital middleboxes | |
| CN111427971B (en) | Business modeling method, device, system and medium for computer system | |
| US11087255B2 (en) | System and methods for fulfilling an order by determining an optimal set of sources and resources | |
| KR102084389B1 (en) | Company evaluation system and evaluation method therefor | |
| CN105183670A (en) | Data processing method and device used for distributed cache system | |
| CN113743971A (en) | Data processing method and device | |
| CN110309142B (en) | Method and device for rule management | |
| CN111798167B (en) | Warehouse replenishment method and device | |
| CN110738508A (en) | data analysis method and device | |
| US9904744B2 (en) | Probabilistic simulation scenario design by using multiple conditional and nested probability distribution input functions | |
| CN114663015A (en) | Replenishment method and device | |
| US10678800B2 (en) | Recommendation prediction based on preference elicitation | |
| CN110675176B (en) | Method and apparatus for generating an attribute prediction model | |
| EP3864525B1 (en) | System, method, and computer program product for load balancing to process large data sets | |
| US20200057959A1 (en) | Reducing instances of inclusion of data associated with hindsight bias in a training set of data for a machine learning system | |
| CN112528103A (en) | Method and device for recommending objects | |
| CN109614328B (en) | Method and apparatus for processing test data | |
| CN110688295A (en) | Data testing method and device | |
| US20220100953A1 (en) | System and method for autoproducing a form | |
| US20230306280A1 (en) | Systems and methods for reducing problematic correlations between features from machine learning model data | |
| CN113177596B (en) | Block chain address classification method and device | |
| WO2024113966A1 (en) | Article identifier information generation method and apparatus, and device, medium and program product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |