CN109902850A - Determine the method, apparatus and storage medium of Strategy of Inventory Control - Google Patents

Determine the method, apparatus and storage medium of Strategy of Inventory Control Download PDF

Info

Publication number
CN109902850A
CN109902850A CN201810972843.6A CN201810972843A CN109902850A CN 109902850 A CN109902850 A CN 109902850A CN 201810972843 A CN201810972843 A CN 201810972843A CN 109902850 A CN109902850 A CN 109902850A
Authority
CN
China
Prior art keywords
inventory
product
scene simulation
simulation device
scenario parameters
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.)
Granted
Application number
CN201810972843.6A
Other languages
Chinese (zh)
Other versions
CN109902850B (en
Inventor
陈菂
陈天笑
周峰暐
曾嘉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201810972843.6A priority Critical patent/CN109902850B/en
Publication of CN109902850A publication Critical patent/CN109902850A/en
Application granted granted Critical
Publication of CN109902850B publication Critical patent/CN109902850B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

This application discloses the method, apparatus and storage medium of a kind of determining Strategy of Inventory Control, belong to big data technical field.The described method includes: generating multiple inventory's scene simulation devices, the corresponding inventory's scene of each inventory's scene simulation device according to inventory data set；According to inventory data set, determines the initial inventory control strategy including scenario parameters, the scenario parameters in initial inventory control strategy are optimized based on multiple inventory's scene simulation devices, obtain base stock control strategy.Since the scenario parameters in base stock control strategy are obtained later based on the optimization of multiple inventory's scene simulation devices, therefore, base stock control strategy can be applied to multiple inventory's scenes simultaneously, the flexibility of base stock control strategy is improved, while the accuracy for adjusting quantity in stock by base stock control strategy can also be improved.

Description

Determine the method, apparatus and storage medium of Strategy of Inventory Control
Technical field
This application involves big data technical field, in particular to the method, apparatus of a kind of determining Strategy of Inventory Control and deposit Storage media.
Background technique
Currently, the quantity in stock such as fruit product is excessively high during enterprise operation, it is easy to cause the waste of business capital, To generate stockholding cost.Quantity in stock such as fruit product is too low, and quantity in stock may be unsatisfactory for the demand in market, and enterprise needs Additional fund is spent to make up the loss of low stock bring, to generate O/S cost.Therefore, it is necessary to study one kind Strategy of Inventory Control controls the quantity in stock of the product in enterprise, to reduce inventory cost.Wherein, inventory cost includes Stockholding cost and O/S cost.
It since the delivery cycle of product is usually definite value, and is usually at various moments random point to the demand of product Cloth, and each moment to also mutually indepedent between the demand of product, therefore the relevant technologies provide one kind by random period more The new Strategy of Inventory Control based on quantity in stock threshold value.In the Strategy of Inventory Control, quantity in stock threshold value is sent out at random according to the period Changing, within some period, if existing quantity in stock be less than current period in quantity in stock threshold value, using decision one come Quantity in stock is adjusted, to reduce inventory cost drop, if existing quantity in stock is greater than or equal to quantity in stock threshold value in current period, Quantity in stock is adjusted using decision two, is dropped to reducing inventory cost.Wherein, decision one and decision are second is that pre-set two Decision.
It should be noted that above-mentioned Strategy of Inventory Control be for product delivery cycle be definite value scene, still, produce The practical delivery cycle of product may be that there are probabilistic.For example, probably due to the processing progress of product or the logistics of product Situation causes the practical delivery cycle of product to change, in this case, it is impossible to again by above-mentioned Strategy of Inventory Control come Quantity in stock is adjusted, that is to say, the delivery cycle that above-mentioned Strategy of Inventory Control is suitable only for product is the scene of definite value, is caused The flexibility for stating Strategy of Inventory Control is lower, and is also not very by the accuracy that above-mentioned Strategy of Inventory Control adjusts quantity in stock It is high.
Summary of the invention
This application provides the method, apparatus and storage medium of a kind of determining Strategy of Inventory Control, can solve related skill The flexibility of the Strategy of Inventory Control provided in art is lower, and the accuracy also not high problem of adjustment quantity in stock.The skill Art scheme is as follows:
In a first aspect, a kind of method of determining Strategy of Inventory Control is provided, this method comprises:
Inventory data set is obtained, inventory data set includes stored for describing the history library of at least one product Data generate multiple inventory's scene simulation devices, the corresponding inventory field of each inventory's scene simulation device according to inventory data set Scape determines the initial inventory control strategy including scenario parameters, according to multiple inventory's scene simulation devices according to inventory data set Scenario parameters in initial inventory control strategy are optimized, base stock control strategy, base stock control strategy are obtained For being adjusted to quantity in stock.
In this application, since the scenario parameters in base stock control strategy are excellent based on multiple inventory's scene simulation devices It is obtained after changing, therefore, base stock control strategy can be applied to multiple inventory's scenes simultaneously, improve base stock control The flexibility of strategy is made, while the accuracy for adjusting quantity in stock by base stock control strategy can also be improved.
Optionally, the scenario parameters in initial inventory control strategy are optimized according to multiple inventory's scene simulation devices, Obtain base stock control strategy, comprising: determining and each according to each inventory's scene simulation device and initial inventory control strategy Corresponding relationship between the associated scenario parameters value of inventory's scene simulation device and inventory cost；According to multiple inventory's scene simulations The device corresponding relationship between associated multiple scenario parameters values and inventory cost one by one, optimizes scenario parameters, obtains mesh Mark Strategy of Inventory Control.
In this application, it can first determine between scenario parameters value and inventory cost for each inventory's scene simulation device Corresponding relationship, then scenario parameters are optimized according to determining multiple corresponding relationships so that optimization after scene ginseng Number can be adapted to multiple inventory's scenes simultaneously, to improve the flexibility for the base stock control strategy that the application determines
Optionally, determining imitative with each inventory's scene according to each inventory's scene simulation device and initial inventory control strategy Corresponding relationship between the associated scenario parameters value of true device and inventory cost, comprising: for appointing in multiple inventory's scene simulation devices One inventory scene simulation device A generates a plurality of sample data based on inventory's scene simulation device A, and every sample data is for describing one The demand information or supply time information of a product；A plurality of sample data is input to initial inventory control strategy, is obtained and library Deposit the corresponding relationship between the associated scenario parameters value of scene simulation device A and inventory cost.
In one possible implementation, the scene for inventory's scene simulation device A can be determined through the above way Corresponding relationship between parameter value and inventory cost.
Optionally, according to multiple inventory's scene simulation devices one by one between associated multiple scenario parameters values and inventory cost Corresponding relationship, scenario parameters are optimized, base stock control strategy is obtained, comprising: according to imitative with each inventory's scene Corresponding relationship between the associated scenario parameters value of true device and inventory cost, it is determining with each associated field of inventory's scene simulation device Scape parameter extreme value, wherein in the corresponding pass between the associated scenario parameters value of each inventory's scene simulation device and inventory cost In system, the corresponding inventory cost of scenario parameters extreme value is minimum；Determining associated multiple fields one by one with multiple inventory's scene simulation devices The average value of scape parameter extreme value obtains base stock control strategy using determining average value as the optimal value of scenario parameters.
It in this application, can be by associated multiple scenario parameters extreme values are averaged one by one with multiple inventory's scene simulation devices It is worth the value as the scenario parameters after optimization, so that the target control strategy determined can be applied to multiple inventory fields simultaneously Scape.
Optionally, inventory data set include for describe the historic demand data of each product at least one product and History delivery date information.Correspondingly, multiple inventory's scene simulation devices are generated according to inventory data set, comprising: at least one Any product B in product determines the historic demand distribution of product B according to the historic demand data of product B in inventory data set Curve, and according to the historic demand distribution curve of product B, the multiple first kind inventory scene simulation devices for being directed to product B are generated, often The corresponding demand scene of a first kind inventory scene simulation device.According to the history delivery date information of product B in inventory data set, It determines the history supply time distribution curve of product B, and according to the history supply time distribution curve of product B, generates for production Multiple second class inventory scene simulation devices of product B, the corresponding supply time scene of each second class inventory scene simulation device. Wherein, multiple inventory's scene simulation devices include: imitative for multiple first kind inventory scenes of each product at least one product True device and multiple second class inventory scene simulation devices.
In this application, in order to enable the target control strategy determined can adapt to the uncertainty of delivery cycle simultaneously With the uncertainty of demand, multiple inventory's scene simulation devices of generation include above-mentioned multiple first kind inventory scene simulation devices and more A second class inventory scene simulation device.
Optionally, according to the historic demand distribution curve of product B, the multiple first kind inventory scenes for being directed to product B are generated Emulator, comprising: determine the mean value and standard deviation of the historic demand distribution curve of product B；Adjust the historic demand distribution of product B The mean value and/or standard deviation of curve obtain a plurality of demand distribution curve；A first kind is generated according to every demand distribution curve Inventory's scene simulation device obtains multiple first kind inventory scene simulation devices for product B.
Specifically, it can be derived according to the historic demand distribution curve of product B for the more of product B based on aforesaid way A first kind inventory scene simulation device.
Optionally, according to the history supply time distribution curve of product B, the multiple second class inventories for being directed to product B are generated Scene simulation device, comprising: determine the mean value and standard deviation of the history supply time distribution curve of product B；Adjust the history of product B The mean value and/or standard deviation of supply time distribution curve obtain a plurality of supply time distribution curve；According to every supply time point Cloth curve generates second class inventory's scene simulation device, obtains multiple second class inventory scene simulation devices for product B.
Similarly, aforesaid way is also based on to be derived according to the history supply time distribution curve of product B for production Multiple second class inventory scene simulation devices of product B.
Optionally, according to inventory data set, the initial inventory control strategy including scenario parameters is determined, comprising: obtain For the initial model of storage controlling；According to inventory data set, the historic demand of each product at least one product is determined Feature and history supply time feature；According to the historic demand feature of product each at least one product and history supply time Feature is updated initial model, obtain include scenario parameters initial inventory control strategy.
In this application, initial inventory control strategy is to be updated to obtain to initial model according to inventory data set 's.
Second aspect provides a kind of device of determining Strategy of Inventory Control, the device of the determining Strategy of Inventory Control Has the function of realizing the method behavior that Strategy of Inventory Control is determined in above-mentioned first aspect.The determining Strategy of Inventory Control Device includes at least one module, at least one module is for realizing determining storage controlling plan provided by above-mentioned first aspect Method slightly.
The third aspect provides a kind of device of determining Strategy of Inventory Control, the device of the determining Strategy of Inventory Control Structure in include processor and memory, the memory be used for store support determine Strategy of Inventory Control device execution The program that the method for Strategy of Inventory Control is determined provided by first aspect, and storage are stated for realizing above-mentioned first aspect institute Data involved in the method for the determination Strategy of Inventory Control of offer.The processor is configured to for executing the memory The program of middle storage.The operating device of the storage equipment can also include communication bus, which is used for the processor Connection is established between memory.
Fourth aspect provides a kind of computer readable storage medium, is stored in the computer readable storage medium Instruction, when run on a computer, so that computer executes determination Strategy of Inventory Control described in above-mentioned first aspect Method.
5th aspect, provides a kind of computer program product comprising instruction, when run on a computer, so that The method that computer executes determination Strategy of Inventory Control described in above-mentioned first aspect.
In above-mentioned second aspect, the third aspect, fourth aspect and the 5th aspect technical effect obtained and first aspect The technical effect that corresponding technological means obtains is approximate, repeats no more herein.
Detailed description of the invention
Fig. 1 is a kind of system schematic of determining Strategy of Inventory Control provided by the embodiments of the present application；
Fig. 2 is a kind of structural schematic diagram of computer equipment provided by the embodiments of the present application；
Fig. 3 is a kind of method flow diagram of determining Strategy of Inventory Control provided by the embodiments of the present application；
Fig. 4 is a kind of simulation result schematic diagram provided by the embodiments of the present application；
Fig. 5 is another simulation result schematic diagram provided by the embodiments of the present application；
Fig. 6 is that the embodiment of the present application provides a kind of device block diagram of determining Strategy of Inventory Control.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with attached drawing to the application embodiment party Formula is described in further detail.
Before the method to determining Strategy of Inventory Control provided by the embodiments of the present application is explained, first to this Shen Please the application scenarios of embodiment simply introduced.
Currently, when the related personnel of company carries out product purchasing, it usually needs determined according to optimal inventory current The amount of stocking up, wherein optimal inventory refers to quantity in stock when inventory cost can be made to reach minimum.And Strategy of Inventory Control is A kind of model of determining optimal inventory, related personnel can determine optimal inventory according to Strategy of Inventory Control.The application The method for the determination Strategy of Inventory Control that embodiment provides can be applied to when carrying out product purchasing it needs to be determined that optimal inventory Application scenarios in, certainly, the method for determining Strategy of Inventory Control provided by the embodiments of the present application also can be applied at other It needs to be determined that no longer being limited herein in the scene of optimal inventory.
Fig. 1 is a kind of system schematic of determining Strategy of Inventory Control provided by the embodiments of the present application, as shown in Figure 1, should System 100 includes inventory data input module 101, inventory's scene simulation device generation module 102, the sampling of inventory's scene simulation device Module 103, algorithms selection module 104, algorithm transformation module 105, meta learning training module 106 and result output module 107.
Wherein, module is transformed with inventory's scene simulation device generation module 102 and algorithm respectively in inventory data input module 101 105 connections, algorithms selection module 104 are connect with algorithm transformation module 105, inventory's scene simulation device generation module 102 and inventory The connection of scene simulation device sampling module 103, inventory's scene simulation device sampling module 103 are connect with meta learning training module 106, are calculated Method transformation module 105 is also connect with meta learning training module 106, and meta learning training module 106 and result output module 107 connect It connects.
Inventory data input module 101 is for obtaining inventory data set, and the inventory data set difference that will acquire is defeated Enter into inventory's scene simulation device generation module 102 and algorithm transformation module 105.Inventory's scene simulation device generation module 102 is used According to the multiple inventory's scene simulation devices of inventory data set generation.Algorithms selection module 104 from polyalgorithm for selecting One algorithm, and the algorithm of selection is input in algorithm transformation module 105, algorithm is transformed module 105 and is used for according to inventory The algorithm of selection is transformed according to set, obtain include scenario parameters initial inventory control strategy, and by initial inventory control Strategy processed is input to meta learning training module 106.Inventory's scene simulation device sampling module 103 is used for according to inventory's scene simulation device Inventory's scene simulation device that generation module 102 generates determines the training sample for being directed to each inventory's scene simulation device, and will determine Training sample be input to meta learning training module 106, meta learning training module 106 is used for according to imitative for each inventory's scene The training sample of true device optimizes initial inventory control strategy, and the base stock control strategy after optimization is input to Base stock control strategy is showed user by result output module 107 by as a result output module 107.
By the cooperation between above-mentioned modules, the determination Strategy of Inventory Control that application embodiment provides may be implemented Method, wherein the detailed functions of above-mentioned modules will be unfolded to illustrate in the following embodiments, and not illustrate first herein.On in addition, The title for stating each module is only the name of the function according to performed by the module, can also be according to each in practical application Function performed by module is the module definition other names, for example, above-mentioned meta learning training module can also be defined as scene Parameter optimization module, the embodiment of the present application are not specifically limited herein.
Fig. 2 is a kind of structural schematic diagram of computer equipment provided by the embodiments of the present application.Storage controlling plan is determined in Fig. 1 System slightly can be realized by computer equipment shown in Fig. 2.Referring to fig. 2, which includes at least one Manage device 201, communication bus 202, memory 203 and at least one communication interface 204.
Processor 201 can be general central processor (Central Processing Unit, CPU), micro process Device, application-specific integrated circuit (application-specific integrated circuit, ASIC) or one or more For controlling the integrated circuit of application scheme program execution.
Communication bus 202 may include an access, and information is transmitted between said modules.
Memory 203 can be read-only memory (read-only memory, ROM) or can store static information and instruction Other types of static storage device, random access memory (random access memory, RAM) or letter can be stored The other types of dynamic memory of breath and instruction, is also possible to Electrically Erasable Programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read- Only Memory, CD-ROM) or other optical disc storages, optical disc storage (including compression optical disc, laser disc, optical disc, digital universal Optical disc, Blu-ray Disc etc.), magnetic disk storage medium or other magnetic storage apparatus or can be used in carrying or store to have referring to Enable or data structure form desired program code and can by any other medium of computer access, but not limited to this. Memory 203, which can be, to be individually present, and is connected by communication bus 202 with processor 201.Memory 203 can also and be located Reason device 201 integrates.
Communication interface 204, using the device of any transceiver one kind, for other equipment or communication, such as Ethernet, wireless access network (RAN), WLAN (Wireless Local Area Networks, WLAN) etc..
In the concrete realization, as one embodiment, computer equipment may include multiple processors, such as institute in Fig. 2 The processor 201 and processor 205 shown.Each of these processors can be monokaryon (single-CPU) processing Device is also possible to multicore (multi-CPU) processor.Here processor can refer to one or more equipment, circuit, And/or the processing core for handling data (such as computer program instructions).
In the concrete realization, as one embodiment, computer equipment can also include output equipment 206 and input equipment 207.Output equipment 206 and processor 201 communicate, and can show information in many ways.For example, output equipment 206 can be with It is liquid crystal display (liquid crystal display, LCD), light emitting diode (light emitting diode, LED) Show that equipment, cathode-ray tube (cathode ray tube, CRT) show equipment or projector (projector) etc..Input is set It is communicated for 207 and processor 201, the input of user can be received in many ways.For example, input equipment 207 can be mouse Mark, keyboard, touch panel device or sensing equipment etc..
Above-mentioned computer equipment can be a general purpose computing device either dedicated computing machine equipment.Having During body is realized, computer equipment can be desktop computer, portable computer, network server, palm PC (Personal Digital Assistant, PDA), cell phone, tablet computer, wireless terminal device, communication equipment or embedded set It is standby.The unlimited type for determining computer equipment of the embodiment of the present invention.
Wherein, memory 203 is used to store the program code for executing application scheme, and is held by processor 201 to control Row.Processor 201 is for executing the program code stored in memory 203.It may include one or more soft in program code Part module (such as: inventory data input module, inventory's scene simulation device sampling module, is calculated inventory's scene simulation device generation module Method selecting module, algorithm transformation module, meta learning training module and result output module etc.).Inventory's control is determined shown in Fig. 1 Make strategy system can by one or more software modules in the program code in processor 201 and memory 203, To determine the data for development and application.
Fig. 3 is a kind of method flow diagram of determining Strategy of Inventory Control provided by the embodiments of the present application, as shown in figure 3, should Method includes the following steps:
Step 301: obtaining inventory data set, inventory data set includes the history library for describing at least one product Stored data.
Since the purpose of the method for determining Strategy of Inventory Control provided by the embodiments of the present application is to provide a kind of fit for client The Strategy of Inventory Control of itself Inventory Performance is closed, therefore, the implementation for obtaining inventory data set can be with are as follows: receives on client The inventory data set of biography, and the inventory data set is stored, in order to be client according to the data in the inventory data set One Strategy of Inventory Control is provided.Wherein, inventory data set includes the stored number of history library at least one product According to for example, including historic demand data, the history of each product of each product at least one product in inventory data set The data such as the ratio between delivery date information and the history stockholding cost and history inventory's shortage cost of each product.
In the embodiment of the present application, the historic demand data of each product include that each product is being worked as at least one product The outbound information at each moment before the preceding time.For example, in one possible implementation, it is each at least one product The historic demand data of product can be indicated by following outbound matrix Y:
Wherein, yijIndicate outbound amount of the product i at this time point of j, i is more than or equal to 1 and to be less than or equal to n just Integer, j are the positive integer more than or equal to 1 and less than or equal to t.
The history delivery date information of each product includes the history supply time information of the product, such as the history goods of product i Phase information may include supply time information of the product i within the previous year of current time.Wherein, due in subsequent training process A large amount of data are needed, therefore, when the quantity of existing true product i history supply time information is more, Ke Yizhi It connects using existing true product i history supply time information as product i history delivery date information；When existing true product It, can be according to existing true product i history supply time information again when the small number of i history supply time information Some supply time information are generated, then the supply time information regenerated and existing true product i history are supplied Temporal information is collectively as product i history delivery date information.
In addition, since different product attribute may be different, be directed to different product, history stockholding cost with go through Ratio between history O/S cost may also be different.When in concrete application, if lacking the history library for each product The real data of the ratio between cost of carry and history inventory's shortage cost is deposited, the history inventory of each product can will be directed to Ratio between cost of carry and history inventory's shortage cost is disposed as preset value, wherein preset value can be 40, the application Embodiment is not specifically limited herein.
Step 302: multiple inventory's scene simulation devices being generated according to inventory data set, each inventory's scene simulation device is corresponding One inventory's scene.
In application embodiment, in order to enable the Strategy of Inventory Control determined can adapt to various possible inventory fields Scape first can generate multiple inventory's scene simulation devices according to inventory data set, then according to multiple inventory's scene simulation devices come Strategy of Inventory Control is determined, so that the Strategy of Inventory Control determined can adapt to multiple inventory's scene simulation device simultaneously and correspond to Multiple inventory's scenes.
In addition, due to including the data for describing the inventory information of at least one product in inventory data set, without Inventory's scene of same product is different, and identical product may correspond to multiple inventory's scenes, therefore, in this application, for Each product determines corresponding inventory's scene simulation device, and the corresponding inventory's scene simulation of each product in inventory data set Device can will determine obtained inventory's scene simulation device corresponding with each product as multiple libraries in step 302 to be multiple Deposit scene simulation device.
Due to determining that the implementation of corresponding inventory's scene simulation device is basic for each product in inventory data set It is identical, illustrate the implementation for determining the corresponding inventory's scene simulation device of product B by taking product B as an example below, determines other products The implementation of corresponding inventory's scene simulation device not reinflated explanation herein.
Specifically, it is determined that the implementation of the corresponding inventory's scene simulation device of product B can be with are as follows: at least one product In any product B determine that the historic demand distribution of product B is bent according to the historic demand data of the product B in inventory data set Line, and according to the historic demand distribution curve of product B, the multiple first kind inventory scene simulation devices for being directed to product B are generated, each First kind inventory's scene simulation device corresponds to a demand scene.According to the history delivery date information of product B in inventory data set, really The history supply time distribution curve of fixed output quota product B, and according to the history supply time distribution curve of product B, it generates and is directed to product B Multiple second class inventory scene simulation devices, the corresponding supply time scene of each second class inventory scene simulation device.Namely It is in the embodiment of the present application, for any product, multiple and different demand scenes can be gone out according to inventory data Ensemble simulation With multiple and different supply time scenes.
At this point, multiple inventory's scene simulation devices in step 302 include: at least one product for the more of each product A first kind inventory scene simulation device and multiple second class inventory scene simulation devices.
Wherein, according to the historic demand data of the product B in inventory data set, the historic demand distribution of product B is determined The implementation of curve are as follows: the historic demand data of product B, the historic demand that will be analyzed are analyzed from inventory data set Data are mapped in time coordinate corresponding with demand, obtain multiple points in coordinate, and multiple points in coordinate are intended Close the historic demand distribution curve that product B can be obtained.In addition, according to the history delivery date information of product B in inventory data set, Determine that the implementation of the history supply time distribution curve of product B can also be with reference to the implementation, herein not reinflated theory It is bright.
It should be noted that if the data that client uploads further include the historic demand point of the product B to have completed Cloth curve and history supply time curve, then without the historic demand distribution curve of determining product B and history supply time again Distribution curve, the historic demand distribution curve and history supply time curve for being directly utilized respectively the product B of client's upload determine Multiple first kind inventory scene simulation devices and multiple second class inventory scene simulation devices for product B.
In addition, in one possible implementation, according to the historic demand distribution curve of product B, generating and being directed to product B The implementations of multiple first kind inventory scene simulation devices can be with are as follows: determine the mean value of the historic demand distribution curve of product B And standard deviation；The mean value and/or standard deviation for adjusting the historic demand distribution curve of product B, obtain a plurality of demand distribution curve；Root First kind inventory's scene simulation device is generated according to every demand distribution curve, obtains multiple first kind inventories for product B Scene simulation device.
Since every demand distribution curve can describe a demand scene, in addition, any bar curve can use Corresponding algorithm model characterizes, and therefore, generates first kind inventory's scene simulation device according to every demand distribution curve Implementation can be with are as follows: analyzes every demand distribution curve, obtains the algorithm that can characterize every demand distribution curve Model, using obtained algorithm model as first kind inventory's scene simulation device.In this case, each first kind inventory scene Emulator can describe a demand scene.In addition, standard deviation can characterize since mean value can characterize the number of demand The fluctuation of demand, therefore can be with the various differences of analog equipment B for multiple first kind inventory scene simulation devices of product B The scene of demand, or the simulation scene that demand is similar but fluctuation is different.
In addition, the mean value and/or standard deviation of the historic demand distribution curve of adjustment product B, it can be with are as follows: only adjustment product B Historic demand distribution curve mean value, alternatively, only adjust product B historic demand distribution curve standard deviation, alternatively, simultaneously The mean value and standard deviation of the historic demand distribution curve of product B are adjusted, it is not limited here.
For example, the mean value of the historic demand distribution curve of the product B determined is 200, standard deviation 0.3, then can be with The mean value for adjusting the historic demand distribution curve of product B is 150,175,225 and 250, standard deviation is still 0.3, can be obtained in this way To other 4 demand distribution curves.Alternatively, the standard deviation of the historic demand distribution curve of adjustable product B be 0.1,0.15, 0.2,0.25,0.35,0.4,0.45 and 0.5, mean value is still 200, other 8 demand distribution curves available in this way.
In addition, in one possible implementation, according to the history supply time distribution curve of product B, generation is directed to The implementation of multiple second class inventory scene simulation devices of product B can be with are as follows: determines the history supply time distribution of product B The mean value and standard deviation of curve；The mean value and/or standard deviation for adjusting the history supply time distribution curve of product B, obtain a plurality of Supply time distribution curve；Second class inventory's scene simulation device is generated according to every supply time distribution curve, obtains needle To multiple second class inventory scene simulation devices of product B.
Similarly, since every supply time distribution curve can describe a supply time scene, according to every The implementation that supply time distribution curve generates second class inventory's scene simulation device can be with are as follows: to every supply time Distribution curve is analyzed, and the algorithm model that can characterize every supply time distribution curve, the algorithm model that will be obtained are obtained As second class inventory's scene simulation device.In this case, each second class inventory's scene simulation device can describe product B A supply time scene.For example, one of them second class inventory's scene simulation device can describe the fixed field of delivery cycle Scape, another the second class inventory's scene simulation device can describe delivery cycle according to the scene etc. of Gaussian Profile.
Wherein, the mean value of the history supply time distribution curve of product B and/or the mode of standard deviation and adjustment history are adjusted The mean value of demand distribution curve and/or the mode of standard deviation are essentially identical, herein not reinflated explanation.
Step 303: according to inventory data set, determining the initial inventory control strategy including scenario parameters.
In one possible implementation, step 303 is specifically as follows: the initial model for being directed to storage controlling is obtained, According to inventory data set, the historic demand feature and history supply time feature of each product at least one product are determined, According to the historic demand feature and history supply time feature of product each at least one product, initial model is carried out more Newly, obtain include scenario parameters initial inventory control strategy.
Wherein, the implementation obtained for the initial model of storage controlling can be with are as follows: from multiple basic operational research algorithms The basic operational research algorithm of middle random selection one, and using the basic operational research algorithm selected as initial model.Optionally, it obtains It can be with for the implementation of the initial model of storage controlling are as follows: the multiple basic operational research algorithms of display are directed to when detecting When the selection operation of one of basis operational research algorithm, using the basic operational research algorithm selected as initial model, wherein choosing It selects operation and is triggered by user by predetermined registration operation.It that is to say, in the embodiment of the present application, the initial model of acquisition can be terminal Randomly selected one basic operational research algorithm, or refer to the basic operational research algorithm of user's selection.
In addition, determining the historic demand feature and history of each product at least one product according to inventory data set The implementation of supply time feature, can be with reference to the historic demand distribution curve and the history supply of material for determining product B in step 302 The implementation of time distribution curve, no longer elaborates herein.
In addition, according to the historic demand feature and history supply time feature of product each at least one product, to first Beginning model is updated, and obtains including that the implementation of the initial inventory control strategy of scenario parameters can be with are as follows: according at least one When the historic demand feature and history supply time feature of each product are determined to the distribution of influence demand or the supply of material in a product Between the parameter that is distributed, then determining parameter is added in initial model, obtain include scenario parameters initial inventory control Strategy.
For example, in the embodiment of the present application, using the online convex optimized algorithm based on circulation as the basic operational research obtained Algorithm, the online convex optimized algorithm can be indicated using following formula are as follows:
Wherein, G indicates total inventory cost, and h indicates the unit cost of heap goods, and p indicates unit cost out of stock, and BS is most Excellent quantity in stock, ItRefer to the inventory of t moment, ItIt is a parameter relevant to the supply time, It' (BS) is that inventory leads BS Number, DtIt is the demand of t moment.
It is analyzed by historic demand feature to each product at least one product and history supply time feature It was found that gradient parameter α can influence demand distribution or supply time distribution.Therefore, can join gradient parameter α as scene Number, and after gradient parameter α is added to above-mentioned formula, available initial inventory control strategy, the initial inventory controls plan It can slightly be indicated using following formula:
It is above-mentioned be using by gradient parameter α as being illustrated for scenario parameters, can be according to at least when concrete application The analysis of the historic demand feature and history supply time feature of each product is as a result, using other parameters as field in one product Scape parameter, no longer elaborates herein.
In addition, it is above-mentioned be using the online convex optimized algorithm based on circulation as obtain basic operational research algorithm for carry out Illustrate, in concrete application, when selecting other algorithms as basic operational research algorithm, still basis can be transported in the manner described above It raises algorithm to be transformed, obtains initial inventory control strategy, no longer elaborate herein.
Step 304: the scenario parameters in initial inventory control strategy are optimized according to multiple inventory's scene simulation devices, Base stock control strategy is obtained, base stock control strategy is for being adjusted quantity in stock.
Wherein, step 304 can be divided into following two step to realize:
(1) according to each inventory's scene simulation device and initial inventory control strategy, determining and each inventory's scene simulation device Corresponding relationship between associated scenario parameters value and inventory cost.
In one possible implementation, step (1) is specifically as follows: for appointing in multiple inventory's scene simulation devices One inventory scene simulation device A generates a plurality of sample data based on inventory's scene simulation device A, and every sample data is for describing one The demand information and supply time information of a product；A plurality of sample data is input to initial inventory control strategy, obtain with often Corresponding relationship between the associated scenario parameters value of a inventory's scene simulation device and inventory cost.
Wherein, by step 302 it is found that each inventory's scene simulation device is demand distribution curve or the confession according to product Therefore a plurality of sample data can be generated according to each inventory's scene simulation device in the algorithm model that ETCD estimated time of commencing discharging distribution curve generates, Every sample data may include a demand information or a supply time information.
In addition, since initial inventory control strategy is that one kind information or supply time information can determine inventory according to demand The algorithm of cost therefore can when the sample data is input to initial inventory Controlling model for any bar sample data To obtain the inventory cost for the sample data, and obtained inventory cost is using including stating by the way of scenario parameters 's.For example, sample data is indicated using x, inventory cost is identified using y, when being performed both by the operation to a plurality of sample data, Available following corresponding relationship: y1=f (x1,α)、y2=f (x2,α)、…、yn=f (xn,α).Wherein, n is a plurality of sample number According to number, α is scenario parameters.According to the corresponding relationship, pair between inventory cost y and scenario parameters α value can be determined It should be related to, be labeled as y=g (α).It, can be with when being performed both by aforesaid operations to multiple inventory's scene simulation devices that step 302 generates Obtain the corresponding relationship gone out between inventory cost y and scenario parameters α value for each inventory's scene simulation device.
(2) according to multiple inventory's scene simulation devices pair between associated multiple scenario parameters values and inventory cost one by one It should be related to, scenario parameters are optimized, obtain base stock control strategy.
In one possible implementation, step (2) is specifically as follows: being associated with according to each inventory's scene simulation device Scenario parameters value and inventory cost between corresponding relationship, it is determining with the associated scenario parameters pole of each inventory's scene simulation device It is worth, in the corresponding relationship between the associated scenario parameters value of each inventory's scene simulation device and inventory cost, scenario parameters The corresponding inventory cost of extreme value is minimum；Determining associated multiple scenario parameters extreme values one by one with multiple inventory's scene simulation devices it is flat Mean value can join the scene in initial inventory control strategy at this time using determining average value as the optimal value of scenario parameters Number replaces with the optimal value, and base stock control strategy can be obtained.
For example, with multiple inventory's scene simulation devices, associated multiple scenario parameters extreme values are respectively labeled as α 1, α 2, α 3 one by one And α 2, then it is determined that the implementation of the optimal value of scenario parameters can be with are as follows: determine the average value of α 1, α 2, α 3 and α 2 For the optimal value of scenario parameters.
It is using the average value of multiple scenario parameters extreme values as the optimal value of scenario parameters in above-mentioned implementation.When So, multiple scenario parameters extreme values can also be handled using other modes, to determine the optimal value of scenario parameters, herein not It is specific again to limit.
In conjunction with the system in Fig. 1 it is found that inventory data input module determines inventory data collection for through the above steps 301 It closes.Inventory's scene simulation device generation module is for through the above steps 302 inventory based on the output of inventory data input module Multiple inventory's scene simulation devices are determined according to set.Algorithms selection module is used to select a base from multiple basic operational research algorithms Plinth operational research algorithm is as initial model.The initial model and inventory that algorithm transformation module is used to be exported according to algorithms selection module The inventory data set of data input module output determines the initial inventory control strategy including scenario parameters.Inventory's scene simulation Device sampling module is used to generate a plurality of sample data according to each inventory's scene simulation device, and by each inventory's scene simulation device pair The a plurality of sample data answered is input to meta learning training module.Meta learning training module is used for according to each inventory's scene simulation device Corresponding a plurality of sample data determines the optimal value of scenario parameters, and updates initial inventory control according to the optimal value of scenario parameters Strategy obtains base stock control strategy, and base stock control strategy is input to result output module.As a result output module For showing base stock control strategy to user.
In addition, after determining base stock control strategy, if base stock control strategy is based on according to above-mentioned What the online convex optimized algorithm of circulation determined, then base stock control strategy also belongs to iterative algorithm on a kind of line, in this way, after It is continuous also base stock control strategy persistently to be optimized, the detailed process of every suboptimization and above-mentioned determining base stock control The process of strategy is essentially identical, the data in inventory data set that the detailed process of only every suboptimization uses it is different and , not reinflated herein for example, optimized using nearest one month inventory data to base stock control strategy every time Explanation.
In the embodiment of the present application, multiple inventory's scene simulation devices, each inventory's scene are generated according to inventory data set Emulator corresponds to inventory's scene；According to inventory data set, the initial inventory control strategy including scenario parameters, root are determined The scenario parameters in initial inventory control strategy are optimized according to multiple inventory's scene simulation devices, obtain base stock control plan Slightly.Since the scenario parameters in base stock control strategy are obtained later based on the optimization of multiple inventory's scene simulation devices, because This, base stock control strategy can be applied to multiple inventory's scenes simultaneously, the flexibility of base stock control strategy is improved, The accuracy that quantity in stock is adjusted by base stock control strategy can also be improved simultaneously.
In order to further illustrate the beneficial effect of the method for determining Strategy of Inventory Control provided by the embodiments of the present application, to this The method for the determination Strategy of Inventory Control that application embodiment provides has carried out experimental verification, and experimental verification process is as follows:
For XX client, by the history for being used to describe each product between the December of the client in December, 2013~2016 year The data of inventory information are divided into two parts, and a part is used as training set, and a part is used as test set.Using training set as inventory Base stock control strategy is determined according to set, and according to each step in above-mentioned embodiment shown in Fig. 3.Wherein, for instruction Practice each product concentrated, the delivery cycle that the product of half can be set is fixed value, and the delivery cycle of the other half product is Random distribution.After determining base stock control strategy, base stock is controlled according to the data in training set respectively Strategy and the Strategy of Inventory Control provided in the related technology are emulated, and simulation result as shown in Figure 4 is obtained.As shown in figure 4, It is 6427.9 according to the inventory cost that base stock control strategy obtains client, and according to Strategy of Inventory Control in the related technology The inventory cost determined is 7622.4, and theoretical Optimal Inventory cost is 4954.6, it is clear that controls plan by base stock The inventory cost slightly determined will be closer to theoretical Optimal Inventory cost.
After determining base stock control strategy, the base stock can also be controlled using the data in test set Strategy is tested.Specifically, it is provided respectively according to the data in test set to base stock control strategy and in the related technology Strategy of Inventory Control emulated, obtain simulation result as figure 5 illustrates.As shown in figure 5, true according to base stock control strategy The inventory cost of fixed client is 6341.3, and the inventory cost determined according to test set according to the relevant technologies is 6912.1, And theoretical Optimal Inventory cost is 4835.2, it is clear that the inventory determined according to test set by base stock control strategy Cost will be closer to theoretical Optimal Inventory cost.Also, for base stock control strategy, inventory that test set is determined at The difference between inventory cost that sheet and training set are determined is little, illustrates the training process base for determining base stock control strategy The case where over-fitting or poor fitting is not present in this.
Fig. 6 is a kind of device of determining Strategy of Inventory Control provided by the embodiments of the present application, as shown in fig. 6, the device 600 Include:
Module 601 is obtained, for executing the step 301 in Fig. 3 embodiment；
Generation module 602, for executing the step 302 in Fig. 3 embodiment；
Determining module 603, for executing the step 303 in Fig. 3 embodiment；
Optimization module 604, for executing the step 304 in Fig. 3 embodiment.
Optionally, optimization module 604 includes:
First determination unit is used for according to each inventory's scene simulation device and initial inventory control strategy, determining and each Corresponding relationship between the associated scenario parameters value of inventory's scene simulation device and inventory cost；
Second determination unit, for according to multiple inventory's scene simulation devices associated multiple scenario parameters values and library one by one The corresponding relationship being saved as between this, optimizes scenario parameters, obtains base stock control strategy.
Optionally, the first determination unit is specifically used for:
For inventory's scene simulation device A any in multiple inventory's scene simulation devices, generated based on inventory's scene simulation device A more Sample data, every sample data are used to describe the demand information or supply time information of a product；
A plurality of sample data is input to initial inventory control strategy, is obtained and the associated scene of inventory scene simulation device A Corresponding relationship between parameter value and inventory cost.
Optionally, the second determination unit is specifically used for:
According to the corresponding relationship between the associated scenario parameters value of each inventory's scene simulation device and inventory cost, determine With the associated scenario parameters extreme value of each inventory's scene simulation device, with the associated scenario parameters value of each inventory's scene simulation device In corresponding relationship between inventory cost, the corresponding inventory cost of scenario parameters extreme value is minimum；
The determining average value of associated multiple scenario parameters extreme values one by one with multiple inventory's scene simulation devices, will be determining flat Optimal value of the mean value as scenario parameters, obtains base stock control strategy.
Optionally, inventory data set include for describe the historic demand data of each product at least one product and History delivery date information；
Generation module 602 includes:
Third determination unit is used for for product B any at least one product, according to being produced in inventory data set The historic demand data of product B, determine the historic demand distribution curve of product B, and according to the historic demand distribution curve of product B, Generate the multiple first kind inventory scene simulation devices for being directed to product B, the corresponding demand of each first kind inventory scene simulation device Scene；
4th determination unit determines going through for product B for the history delivery date information according to product B in inventory data set History supply time distribution curve, and according to the history supply time distribution curve of product B, it generates and is directed to multiple the second of product B Class inventory's scene simulation device, the corresponding supply time scene of each second class inventory scene simulation device；
Wherein, multiple inventory's scene simulation devices include: multiple first class libraries that each product is directed at least one product Deposit scene simulation device and multiple second class inventory scene simulation devices.
Optionally, third determination unit is specifically used for:
Determine the mean value and standard deviation of the historic demand distribution curve of product B；
The mean value and/or standard deviation for adjusting the historic demand distribution curve of product B, obtain a plurality of demand distribution curve；
First kind inventory's scene simulation device is generated according to every demand distribution curve, is obtained for the multiple of product B First kind inventory's scene simulation device.
Optionally, the 4th determination unit is specifically used for:
Determine the mean value and standard deviation of the history supply time distribution curve of product B；
The mean value and/or standard deviation for adjusting the history supply time distribution curve of product B obtain a plurality of supply time distribution Curve；
Second class inventory's scene simulation device is generated according to every supply time distribution curve, is obtained for product B's Multiple second class inventory scene simulation devices.
Optionally it is determined that module 603, is specifically used for:
Obtain the initial model for being directed to storage controlling；
According to inventory data set, when determining that the historic demand feature of each product at least one product and history are supplied Between feature；
According to the historic demand feature and history supply time feature of each product at least one product, to introductory die Type is updated, obtain include scenario parameters initial inventory control strategy.
In the embodiment of the present application, multiple inventory's scene simulation devices, each inventory's scene are generated according to inventory data set Emulator corresponds to inventory's scene；According to inventory data set, the initial inventory control strategy including scenario parameters, root are determined The scenario parameters in initial inventory control strategy are optimized according to multiple inventory's scene simulation devices, obtain base stock control plan Slightly.Since the scenario parameters in base stock control strategy are obtained later based on the optimization of multiple inventory's scene simulation devices, because This, base stock control strategy can be applied to multiple inventory's scenes simultaneously, the flexibility of base stock control strategy is improved, The accuracy that quantity in stock is adjusted by base stock control strategy can also be improved simultaneously.
It should be understood that the device of determining Strategy of Inventory Control provided by the above embodiment is determining Strategy of Inventory Control When, only the example of the division of the above functional modules, in practical application, it can according to need and divide above-mentioned function With being completed by different functional modules, i.e., the internal structure of equipment is divided into different functional modules, to complete above description All or part of function.In addition, the device of determining Strategy of Inventory Control provided by the above embodiment and determining storage controlling The embodiment of the method for strategy belongs to same design, and specific implementation process is detailed in embodiment of the method, and which is not described herein again.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or its any combination real It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program Product includes one or more computer instructions.It is all or part of when loading on computers and executing the computer instruction Ground is generated according to process or function described in the embodiment of the present invention.The computer can be general purpose computer, special purpose computer, Computer network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or Person is transmitted from a computer readable storage medium to another computer readable storage medium, for example, the computer instruction Can from a web-site, computer, server or data center by it is wired (such as: coaxial cable, optical fiber, data use Family line (Digital Subscriber Line, DSL)) or wireless (such as: infrared, wireless, microwave etc.) mode to another net Website, computer, server or data center are transmitted.The computer readable storage medium can be computer can Any usable medium of access either includes the data storage such as one or more usable mediums integrated server, data center Equipment.The usable medium can be magnetic medium (such as: floppy disk, hard disk, tape), optical medium (such as: digital versatile disc (Digital Versatile Disc, DVD)) or semiconductor medium (such as: solid state hard disk (Solid State Disk, SSD)) etc..
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The above is embodiment provided by the present application, all in spirit herein and original not to limit the application Within then, any modification, equivalent replacement, improvement and so on be should be included within the scope of protection of this application.

Claims (18)

1. a kind of method of determining Strategy of Inventory Control, which is characterized in that the described method includes:
Inventory data set is obtained, the inventory data set includes stored for describing the history library of at least one product Data；
Multiple inventory's scene simulation devices, the corresponding inventory of each inventory's scene simulation device are generated according to the inventory data set Scene；
According to the inventory data set, the initial inventory control strategy including scenario parameters is determined；
The scenario parameters in the initial inventory control strategy are optimized according to the multiple inventory's scene simulation device, are obtained Base stock control strategy, the base stock control strategy is for being adjusted quantity in stock.
2. the method as described in claim 1, which is characterized in that it is described according to the multiple inventory's scene simulation device to it is described just Scenario parameters in beginning Strategy of Inventory Control optimize, and obtain the base stock control strategy, comprising:
According to each inventory's scene simulation device and the initial inventory control strategy, determination is associated with each inventory's scene simulation device Scenario parameters value and inventory cost between corresponding relationship；
According to corresponding between associated multiple scenario parameters values and inventory cost one by one with the multiple inventory's scene simulation device Relationship optimizes the scenario parameters, obtains the base stock control strategy.
3. method according to claim 2, which is characterized in that described according to each inventory's scene simulation device and the initial library Deposit control strategy, the determining corresponding relationship between the associated scenario parameters value of each inventory's scene simulation device and inventory cost, Include:
It is raw based on inventory's scene simulation device A for inventory's scene simulation device A any in the multiple inventory's scene simulation device At a plurality of sample data, every sample data is used to describe the demand information or supply time information of a product；
The a plurality of sample data is input to the initial inventory control strategy, obtains closing with inventory's scene simulation device A Corresponding relationship between the scenario parameters value and inventory cost of connection.
4. method as claimed in claim 2 or claim 3, which is characterized in that the basis and the multiple inventory's scene simulation device one Corresponding relationship between one associated multiple scenario parameters values and inventory cost, optimizes the scenario parameters, obtains institute State base stock control strategy, comprising:
It is determining and every according to the corresponding relationship between the associated scenario parameters value of each inventory's scene simulation device and inventory cost A associated scenario parameters extreme value of inventory's scene simulation device, wherein with each associated scenario parameters of inventory's scene simulation device It is worth the corresponding relationship between inventory cost, the corresponding inventory cost of the scenario parameters extreme value is minimum；
The determining average value of associated multiple scenario parameters extreme values one by one with the multiple inventory's scene simulation device, will be determining flat Optimal value of the mean value as the scenario parameters obtains the base stock control strategy.
5. the method as described in Claims 1-4 is any, which is characterized in that the inventory data set includes for describing State the historic demand data and history delivery date information of each product at least one product；
It is described that multiple inventory's scene simulation devices are generated according to the inventory data set, comprising:
For any product B at least one described product, according to the historic demand of product B described in the inventory data set Data determine the historic demand distribution curve of the product B, and according to the historic demand distribution curve of the product B, generate needle To multiple first kind inventory scene simulation devices of the product B, the corresponding demand field of each first kind inventory scene simulation device Scape；
According to the history delivery date information of product B described in the inventory data set, the history supply time of the product B is determined Distribution curve, and according to the history supply time distribution curve of the product B, generate multiple second classes for being directed to the product B Inventory's scene simulation device, the corresponding supply time scene of each second class inventory scene simulation device；
Wherein, the multiple inventory's scene simulation device includes: to be directed to multiple the first of each product at least one described product Class inventory's scene simulation device and multiple second class inventory scene simulation devices.
6. method as claimed in claim 5, which is characterized in that the historic demand distribution curve according to the product B, it is raw At the multiple first kind inventory scene simulation devices for being directed to the product B, comprising:
Determine the mean value and standard deviation of the historic demand distribution curve of the product B；
The mean value and/or standard deviation for adjusting the historic demand distribution curve of the product B obtain a plurality of demand distribution curve；
First kind inventory's scene simulation device is generated according to every demand distribution curve, is obtained for the multiple of the product B First kind inventory's scene simulation device.
7. method as claimed in claim 5, which is characterized in that described to be distributed song according to the history supply time of the product B Line generates the multiple second class inventory scene simulation devices for being directed to the product B, comprising:
Determine the mean value and standard deviation of the history supply time distribution curve of the product B；
The mean value and/or standard deviation for adjusting the history supply time distribution curve of the product B obtain a plurality of supply time distribution Curve；
Second class inventory's scene simulation device is generated according to every supply time distribution curve, is obtained for the product B's Multiple second class inventory scene simulation devices.
8. the method as described in claim 1 to 7 is any, which is characterized in that it is described according to the inventory data set, determine packet Include the initial inventory control strategy of scenario parameters, comprising:
Obtain the initial model for being directed to storage controlling；
According to the inventory data set, determine that the historic demand feature and history of each product at least one described product supply ETCD estimated time of commencing discharging feature；
According to the historic demand feature and history supply time feature of each product at least one described product, to described initial Model is updated, obtain include scenario parameters initial inventory control strategy.
9. a kind of device of determining Strategy of Inventory Control, which is characterized in that described device includes:
Module is obtained, for obtaining inventory data set, the inventory data set includes for describing at least one product The stored data of history library；
Generation module, for generating multiple inventory's scene simulation devices, each inventory's scene simulation according to the inventory data set Device corresponds to inventory's scene；
Determining module, for determining the initial inventory control strategy including scenario parameters according to the inventory data set；
Optimization module, for according to the multiple inventory's scene simulation device to the scenario parameters in the initial inventory control strategy It optimizes, obtains base stock control strategy, the base stock control strategy is for being adjusted quantity in stock.
10. device as claimed in claim 9, which is characterized in that the optimization module includes:
First determination unit is used for according to each inventory's scene simulation device and the initial inventory control strategy, determining and each Corresponding relationship between the associated scenario parameters value of inventory's scene simulation device and inventory cost；
Second determination unit, for according to the multiple inventory's scene simulation device associated multiple scenario parameters values and library one by one The corresponding relationship being saved as between this, optimizes the scenario parameters, obtains the base stock control strategy.
11. device as claimed in claim 10, which is characterized in that first determination unit is specifically used for:
It is raw based on inventory's scene simulation device A for inventory's scene simulation device A any in the multiple inventory's scene simulation device At a plurality of sample data, every sample data is used to describe the demand information or supply time information of a product；
The a plurality of sample data is input to the initial inventory control strategy, obtains closing with inventory's scene simulation device A Corresponding relationship between the scenario parameters value and inventory cost of connection.
12. device as described in claim 10 or 11, which is characterized in that second determination unit is specifically used for:
It is determining and every according to the corresponding relationship between the associated scenario parameters value of each inventory's scene simulation device and inventory cost A associated scenario parameters extreme value of inventory's scene simulation device, with the associated scenario parameters value of each inventory's scene simulation device and library It is saved as in the corresponding relationship between this, the corresponding inventory cost of the scenario parameters extreme value is minimum；
The determining average value of associated multiple scenario parameters extreme values one by one with the multiple inventory's scene simulation device, will be determining flat Optimal value of the mean value as the scenario parameters obtains the base stock control strategy.
13. the device as described in claim 9 to 12 is any, which is characterized in that the inventory data set includes for describing The historic demand data of each product and history delivery date information at least one described product；
The generation module includes:
Third determination unit is used for for any product B at least one described product, according in the inventory data set The historic demand data of the product B determine the historic demand distribution curve of the product B, and according to the history of the product B Demand distribution curve, generates the multiple first kind inventory scene simulation devices for being directed to the product B, and each first kind inventory scene is imitative The corresponding demand scene of true device；
4th determination unit determines the production for the history delivery date information of the product B according to the inventory data set The history supply time distribution curve of product B, and according to the history supply time distribution curve of the product B, it generates for described Multiple second class inventory scene simulation devices of product B, the corresponding supply time scene of each second class inventory scene simulation device；
Wherein, the multiple inventory's scene simulation device includes: to be directed to multiple the first of each product at least one described product Class inventory's scene simulation device and multiple second class inventory scene simulation devices.
14. device as claimed in claim 13, which is characterized in that the third determination unit is specifically used for:
Determine the mean value and standard deviation of the historic demand distribution curve of the product B；
The mean value and/or standard deviation for adjusting the historic demand distribution curve of the product B obtain a plurality of demand distribution curve；
First kind inventory's scene simulation device is generated according to every demand distribution curve, is obtained for the multiple of the product B First kind inventory's scene simulation device.
15. device as claimed in claim 13, which is characterized in that the 4th determination unit is specifically used for:
Determine the mean value and standard deviation of the history supply time distribution curve of the product B；
The mean value and/or standard deviation for adjusting the history supply time distribution curve of the product B obtain a plurality of supply time distribution Curve；
Second class inventory's scene simulation device is generated according to every supply time distribution curve, is obtained for the product B's Multiple second class inventory scene simulation devices.
16. the device as described in claim 9 to 15 is any, which is characterized in that the determining module is specifically used for:
Obtain the initial model for being directed to storage controlling；
According to the inventory data set, determine that the historic demand feature and history of each product at least one described product supply ETCD estimated time of commencing discharging feature；
According to the historic demand feature and history supply time feature of each product at least one described product, to described initial Model is updated, obtain include scenario parameters initial inventory control strategy.
17. a kind of device of determining Strategy of Inventory Control, which is characterized in that described device includes memory and processor；
The memory is used to store the program for supporting described device perform claim to require the described in any item methods of 1-8, described Processor is configurable for executing the program stored in the memory.
18. a kind of computer readable storage medium, it is stored with instruction in the computer readable storage medium, when it is in computer When upper operation, so that the computer perform claim requires the described in any item methods of 1-8.
CN201810972843.6A 2018-08-24 2018-08-24 Method, device and storage medium for determining inventory control strategy Active CN109902850B (en)

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 true CN109902850A (en) 2019-06-18
CN109902850B 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)

Cited By (4)

* Cited by examiner, † Cited by third party
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
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
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
CN114862323A (en) * 2022-05-28 2022-08-05 平安银行股份有限公司 Method, device and equipment for analyzing inventory reserve and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
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

Patent Citations (10)

* Cited by examiner, † Cited by third party
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)

* Cited by examiner, † Cited by third party
Title
曹乐: "不确定条件下外协生产库存***控制策略与参数优化", 《重庆大学学报》 *
江世英: "G家电配送中心库存成本控制策略研", 《会计之友》 *

Cited By (4)

* Cited by examiner, † Cited by third party
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
CN114862323A (en) * 2022-05-28 2022-08-05 平安银行股份有限公司 Method, device and equipment for analyzing inventory reserve and storage medium

Also Published As

Publication number Publication date
CN109902850B (en) 2021-08-20

Similar Documents

Publication Publication Date Title
CN109902850A (en) Determine the method, apparatus and storage medium of Strategy of Inventory Control
EP3304350B1 (en) Column ordering for input/output optimization in tabular data
WO2022237667A1 (en) Method and apparatus for determining order fulfillment warehouse
CN104731528B (en) A kind of construction method and system of cloud computing block storage service
US20140214632A1 (en) Smart Crowd Sourcing On Product Classification
JPWO2020026741A1 (en) Information processing method, information processing device and information processing program
US20120059861A1 (en) Method and system for creating a relationship structure
CN109388397A (en) Product page generation method, system, computer equipment and storage medium
US20140214841A1 (en) Semantic Product Classification
CN112785144A (en) Model construction method, device and storage medium based on federal learning
US10552399B2 (en) Predicting index fragmentation caused by database statements
Tutumlu et al. A MIP model and a hybrid genetic algorithm for flexible job-shop scheduling problem with job-splitting
US10242069B2 (en) Enhanced template curating
US10915557B2 (en) Product classification data transfer and management
CN117131299A (en) Page layout method and device based on user
US20140214844A1 (en) Multiple classification models in a pipeline
CN107621980A (en) A kind of virtual machine migration method, cluster control system and control device
JP7244777B2 (en) Generation method, generation device, program, information processing method, and information processing device
US10089674B1 (en) Ordering a set of data associated with an item
CN114862098A (en) Resource allocation method and device
US10915336B1 (en) Optimizing content engagement with imposed content constraints
CN108345941A (en) A kind of parameter regulation means and device
CN112579246A (en) Virtual machine migration processing method and device
CN108197193A (en) A kind of display methods, system and the relevant apparatus of multiple view data
KR102593134B1 (en) Method for resizing embedding table for user devices and apparatus for same

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