CN112132498A - Inventory management method, device, equipment and storage medium - Google Patents
Inventory management method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112132498A CN112132498A CN201910557464.5A CN201910557464A CN112132498A CN 112132498 A CN112132498 A CN 112132498A CN 201910557464 A CN201910557464 A CN 201910557464A CN 112132498 A CN112132498 A CN 112132498A
- Authority
- CN
- China
- Prior art keywords
- resource
- target resource
- target
- inventory
- demand model
- 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.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 43
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 230000005012 migration Effects 0.000 claims abstract description 85
- 238000013508 migration Methods 0.000 claims abstract description 85
- 238000012549 training Methods 0.000 claims abstract description 72
- 238000012360 testing method Methods 0.000 claims description 52
- 238000000034 method Methods 0.000 claims description 44
- 230000006870 function Effects 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012804 iterative process Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 12
- 230000009466 transformation Effects 0.000 description 9
- 230000001932 seasonal effect Effects 0.000 description 8
- 238000013501 data transformation Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 4
- 238000013075 data extraction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012417 linear regression Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000032258 transport Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an inventory management method, device, equipment and storage medium, which receive and analyze an inventory adjustment request to obtain resource identification information of a target resource; acquiring resource attribute characteristics of the target resource corresponding to the resource identification information; determining the resource demand of the target resource according to the resource attribute characteristics and the resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources; and feeding back inventory adjustment information of the target resource determined based on the current inventory information of the target resource and the resource demand. According to the invention, the resource demand of the target resource is obtained through the resource identification information of the target resource and the resource demand model, and the time influence factor related to the resource is considered in the model, so that the inventory management can be more scientifically and accurately carried out, thus the inventory shortage or the inventory overstock can be effectively avoided, and the inventory cost and the operation cost are reduced.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, a device, and a storage medium for inventory management.
Background
In the process of managing the inventory, the enterprise is additionally provided with additional cost due to insufficient inventory or overstock of the inventory. In particular, the demand for resources such as seasonal products is influenced by many factors such as time, and seasonal products have high demand for inventory control due to these characteristics. Therefore, the inventory level is reasonably adjusted, the inventory management strategy is optimized, the insufficient inventory or the overstock of the inventory can be effectively avoided, and the overall supply chain cost and the operation cost are reduced.
In the prior art, a subjective decision method is generally adopted, which is greatly influenced by personal experience, some factors cannot be considered, and decisions of different people are generally different greatly, so that data quantification cannot be used. Other existing methods are generally less accurate in managing inventory of resources such as seasonal products. Therefore, the accuracy of the inventory management strategy in the prior art is not high, so that the insufficient inventory or the overstock of the inventory cannot be accurately and effectively avoided, and the inventory cost and the operation cost are reduced.
Disclosure of Invention
The invention provides an inventory management method, device, equipment and storage medium, which are used for more scientifically and accurately managing inventory, thereby effectively avoiding inventory shortage or inventory overstock and reducing inventory cost and operation cost.
A first aspect of an embodiment of the present invention provides an inventory management method, including:
receiving an inventory adjustment request, and analyzing the inventory adjustment request to obtain resource identification information of a target resource;
acquiring resource attribute characteristics of the target resource corresponding to the resource identification information;
determining the resource demand of the target resource according to the acquired resource attribute characteristics and a resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources;
feeding back inventory adjustment information for the target resource, the inventory adjustment information being determined based on current inventory information for the target resource and the resource demand.
On the basis of the above embodiment, the method further includes:
acquiring a training set and a test set according to the resource migration record information of the target resource;
constructing a preliminary resource demand model of the target resource;
and training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model, wherein the resource demand model comprises resource-related time influence factors.
On the basis of any one of the above embodiments, the time influence factor related to the resource includes at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the time influence factor related to the resource is selected according to a fluctuation cycle of the resource migration record information of the target resource.
On the basis of any of the above embodiments, the training the preliminary resource demand model according to the training set and the test set to obtain the resource demand model includes:
determining model coefficients of the preliminary resource demand model according to the training set;
acquiring the RSS and the square sum of the residual errors between the resource demand predicted by the preliminary resource demand model and the real historical migration quantity of the target resource according to the test set;
constructing a loss function according to the residual square sum RSS and the penalty item;
and repeatedly training and testing through an iterative process to minimize the loss function, thereby obtaining the resource demand model.
On the basis of any of the above embodiments, the resource demand model is:
wherein y is the resource demand of the target resource; x is the number of1-xnThe resource attribute characteristics of the target resource; x is the number ofmonth_j、xweek_k、xday_lRelated variables of month sub-factor, week sub-factor and date sub-factor respectively(ii) a Alpha, beta and gamma are model coefficients respectively.
On the basis of any of the above embodiments, the loss function is constructed according to the following formula:
wherein λ is L1 penalty term weight, and 0< λ < 1; p is the number of arguments.
On the basis of any of the above embodiments, before acquiring the training set and the test set according to the resource migration record information of the target resource, the method further includes:
determining the selection time range of the resource migration record information of the target resource according to the category of the target resource and the time period required to be predicted;
and acquiring the resource migration record information of the target resource according to the selection time range.
On the basis of any of the above embodiments, the acquiring a training set and a test set according to the resource migration record information of the target resource includes:
extracting the resource attribute characteristics of the target resource and the historical migration volume of the target resource according to the resource migration record information of the target resource; the resource attribute characteristics of the target resource comprise all resource attribute characteristics which can influence the historical migration quantity of the target resource;
and segmenting the resource attribute characteristics of the target resource and the historical migration volume of the target resource into the training set and the test set.
On the basis of any of the above embodiments, after the obtaining of the resource migration record information of the target resource, before the training of the preliminary resource demand model according to the training set and the test set, the method further includes:
performing data preprocessing on the resource attribute characteristics of the target resource and the historical migration volume of the target resource, wherein the data preprocessing comprises the following steps: and processing missing data in the historical migration amount of the target resource, and/or performing data transformation on predetermined data in the historical migration amount of the target resource.
On the basis of any of the above embodiments, the processing of the missing data in the historical migration volume of the target resource includes:
judging whether the missing rate of the missing data exceeds a preset threshold value or not; if yes, discarding the missing data; if not, completing the missing data by a difference method;
the data transformation of the predetermined data in the historical migration amount of the target resource comprises:
and performing direct transformation or Box-Cox transformation on the predetermined data so that the predetermined data meets a predetermined order of magnitude after being transformed.
A second aspect of an embodiment of the present invention is to provide an inventory management apparatus, including:
the receiving module is used for receiving the inventory adjustment request and analyzing the inventory adjustment request to obtain the resource identification information of the target resource;
a feature obtaining module, configured to obtain a resource attribute feature of the target resource corresponding to the resource identification information;
the processing module is used for determining the resource demand of the target resource according to the acquired resource attribute characteristics and the resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources;
and the sending module is used for feeding back inventory adjustment information of the target resource, and the inventory adjustment information is determined based on the current inventory information of the target resource and the resource demand.
On the basis of the foregoing embodiment, the processing module is further configured to:
acquiring a training set and a test set according to the resource migration record information of the target resource;
constructing a preliminary resource demand model of the target resource;
and training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model, wherein the resource demand model comprises resource-related time influence factors.
On the basis of any one of the above embodiments, the time influence factor related to the resource includes at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the time influence factor related to the resource is selected according to a fluctuation cycle of the resource migration record information of the target resource.
On the basis of any one of the above embodiments, the processing module is configured to:
determining model coefficients of the preliminary resource demand model according to the training set;
acquiring the RSS and the square sum of the residual errors between the resource demand predicted by the preliminary resource demand model and the real historical migration quantity of the target resource according to the test set;
constructing a loss function according to the residual square sum RSS and the penalty item;
and repeatedly training and testing through an iterative process to minimize the loss function, thereby obtaining the resource demand model.
On the basis of any of the above embodiments, the resource demand model is:
wherein y is the resource demand of the target resource; x is the number of1-xnThe resource attribute characteristics of the target resource; x is the number ofmonth_j、xweek_k、xday_lThe related variables are respectively a month sub-factor, a week sub-factor and a date sub-factor; alpha, beta and gamma are model coefficients respectively.
On the basis of any of the above embodiments, the processing module constructs the loss function according to the following formula:
wherein λ is L1 penalty term weight, and 0< λ < 1; p is the number of arguments.
On the basis of any of the above embodiments, the apparatus further includes a data acquisition module configured to:
before the preliminary resource demand model is trained according to the training set and the testing set, determining the selection time range of the resource migration record information of the target resource according to the category of the target resource and the time period required to be predicted;
and acquiring the resource migration record information of the target resource according to the selection time range.
On the basis of any of the above embodiments, the apparatus further includes a data extraction module configured to:
extracting the resource attribute characteristics of the target resource and the historical migration volume of the target resource according to the resource migration record information of the target resource;
the resource attribute characteristics of the target resource comprise all resource attribute characteristics which can influence the historical migration quantity of the target resource;
and segmenting the resource attribute characteristics of the target resource and the historical migration volume of the target resource into the training set and the test set.
On the basis of any of the above embodiments, the apparatus further includes a data preprocessing module configured to:
performing data preprocessing on the resource attribute characteristics of the target resource and the historical migration volume of the target resource, wherein the data preprocessing comprises the following steps: and processing missing data in the historical migration amount of the target resource, and/or performing data transformation on predetermined data in the historical migration amount of the target resource.
On the basis of any one of the above embodiments, the data preprocessing module is configured to:
judging whether the missing rate of the missing data exceeds a preset threshold value or not; if yes, discarding the missing data; if not, completing the missing data by a difference method; and/or
And performing direct transformation or Box-Cox transformation on the predetermined data so that the predetermined data meets a predetermined order of magnitude after being transformed.
A third aspect of an embodiment of the present invention provides an inventory management device, including:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of the first aspect.
A fourth aspect of embodiments of the present invention is to provide a computer-readable storage medium having stored thereon a computer program;
which when executed by a processor implements the method according to the first aspect.
According to the inventory management method, the inventory management device, the inventory management equipment and the inventory management storage medium, the inventory adjustment request is received, and the inventory adjustment request is analyzed to obtain the resource identification information of the target resource; acquiring resource attribute characteristics of the target resource corresponding to the resource identification information; determining the resource demand of the target resource according to the acquired resource attribute characteristics and a resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources; feeding back inventory adjustment information for the target resource, the inventory adjustment information being determined based on current inventory information for the target resource and the resource demand. According to the embodiment of the invention, the resource demand of the target resource is obtained through the resource identification information of the target resource and the resource demand model, and the time influence factor related to the resource is considered in the model, so that the inventory management can be more scientifically and accurately carried out, the insufficient inventory or the overstock of the inventory can be effectively avoided, and the inventory cost and the operation cost are reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a method for inventory management according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for inventory management according to another embodiment of the present invention;
FIG. 3 is a flow chart of a method for inventory management according to another embodiment of the present invention;
FIG. 4 is a block diagram of an inventory management device according to an embodiment of the present invention;
fig. 5 is a structural diagram of an inventory management device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of an inventory management method according to an embodiment of the present invention. The embodiment provides an inventory management method, which comprises the following specific steps:
s101, receiving an inventory adjustment request, and analyzing the inventory adjustment request to obtain resource identification information of a target resource.
In this embodiment, the target resource may be a seasonal product, such as a fresh product, a clothing product, and the like. When a user needs to perform inventory adjustment on a target resource, an inventory adjustment request is input, resource identification information of the target resource, such as the name, the category, the ID and the like of the target resource, is carried in the inventory adjustment request, and the resource identification information of the target resource can be obtained by analyzing the inventory adjustment request.
S102, acquiring the resource attribute characteristics of the target resource corresponding to the resource identification information.
In this embodiment, after the resource identification information is obtained, the target resource may be determined, and then the resource attribute characteristics of the target resource, such as average price, category, discount, advertisement resource slot number in the predetermined time period of the target resource, and the week, date, month, holiday, air temperature, etc. of the predetermined time period may be obtained, and the resource attribute characteristics may be any resource attribute characteristics that may affect the resource demand. It should be noted that, for different target resources, there may be common resource attribute features, such as product price, discount strength, inventory level, quantity of active resource bits, activity level, shelf life, and the like, and certainly there may also be different resource attribute features, such as for fresh products, the resource attribute features may include production date, shelf life, storage temperature, and the like, and for clothing products, these resource attribute features are not included.
S103, determining the resource demand of the target resource according to the acquired resource attribute characteristics and the resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources.
In this embodiment, the acquired resource attribute characteristics of the target resource are input into a resource demand model trained in advance, so that the resource demand of the target resource can be predicted. The resource demand model is obtained by training resource migration record information (i.e., inventory adjustment history information), and the resource demand model includes time influence factors related to resources, which is described in detail in the following embodiments.
And S104, feeding back inventory adjustment information of the target resource, wherein the inventory adjustment information is determined based on the current inventory information of the target resource and the resource demand.
In this embodiment, the inventory adjustment information can be obtained according to the current inventory information of the target resource and the resource demand, further feeding back the inventory adjustment information of the target resource to adjust the inventory, specifically, generating an inventory adjustment instruction according to the inventory adjustment information, and sending the inventory adjustment instruction to the warehousing robot and/or the logistics vehicle, and automatically adjusting the inventory of the target resource by the warehousing robot and/or the logistics vehicle according to the inventory adjustment instruction, for example, a first warehousing robot can load the target resource in a first warehouse into the logistics vehicle according to the inventory adjustment instruction, the logistics vehicle transports the target resource to a second warehouse according to the inventory adjustment instruction, the target resource is then unloaded by the second warehousing robot to a predetermined location of the second warehouse according to the inventory adjustment instructions, and each warehousing robot and/or logistics vehicle can also be allocated according to the inventory adjustment instruction. Of course, the inventory adjustment information of the target resource can also be fed back to the terminal of the warehouse manager, so that the warehouse manager can adjust the inventory according to the inventory adjustment information.
In the inventory management method provided by this embodiment, an inventory adjustment request is received, and the inventory adjustment request is analyzed to obtain resource identification information of a target resource; acquiring resource attribute characteristics of the target resource corresponding to the resource identification information; determining the resource demand of the target resource according to the acquired resource attribute characteristics and a resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources; feeding back inventory adjustment information for the target resource, the inventory adjustment information being determined based on current inventory information for the target resource and the resource demand. According to the embodiment of the invention, the resource demand of the target resource is obtained through the resource identification information of the target resource and the resource demand model, and the time influence factor related to the resource is considered in the model, so that the resource demand can be predicted more scientifically and accurately, the prediction result is prevented from being influenced by personal subjectivity, the accuracy and the interpretability of the prediction result are improved, and the inventory adjustment can be effectively guided to reduce the inventory cost and the operation cost.
On the basis of the foregoing embodiment, as shown in fig. 2, the inventory management method provided in this embodiment further includes a training process of a resource demand model, and the specific steps are as follows:
s201, acquiring a training set and a test set according to the resource migration record information of the target resource.
In this embodiment, the resource migration record information of the target resource may be obtained first, and then the training set and the test set may be obtained according to the resource migration record information of the target resource.
Specifically, the selection time range of the resource migration record information of the target resource can be determined according to the category of the target resource and the time period required to be predicted; and acquiring the resource migration record information of the target resource according to the selection time range. For example, if the target resource is an extremely time-sensitive product, that is, the sales volume fluctuates greatly within a short time window, the selection time window (that is, the time range) of the migration record information needs to be shortened to ensure the accuracy of the data.
Further, after the resource migration record information of the target resource is obtained, a training set and a test set are obtained according to the following steps:
s2011, according to the resource migration record information of the target resource, extracting the resource attribute characteristics of the target resource and the historical migration volume of the target resource; the resource attribute characteristics of the target resource comprise all resource attribute characteristics which can influence the historical migration quantity of the target resource;
s2012, the resource attribute characteristics of the target resource and the historical migration volume of the target resource are segmented into the training set and the test set.
In this embodiment, after the resource migration record information of the target resource is acquired, the historical migration amount of the target resource in any time period (that is, the resource amount for performing inventory adjustment on the target resource in the time period, in other words, the resource demand amount of the target resource in the time period) and the resource attribute characteristics of the corresponding target resource in the time period, including all the resource attribute characteristics that may affect the historical migration amount of the target resource, may be acquired according to the resource migration record information of the target resource. Furthermore, the obtained resource attribute characteristics of the multiple pairs of target resources and the historical migration volume of the target resources are segmented into a training set and a test set, wherein the training set is used for training a resource demand model, and a user of the test set tests the resource demand model.
S202, constructing a preliminary resource demand model of the target resource.
In this embodiment, a preliminary resource demand model between the resource demand of the target resource and the resource attribute feature is first constructed, where the resource demand model includes a plurality of model coefficients to be determined, and an optimal model coefficient may be determined through the following training process.
S203, training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model, wherein the resource demand model comprises time influence factors related to resources.
In this embodiment, the resource demand model is trained through a training set and a testing set, and an optimal model coefficient is determined, so that an optimal resource demand model is obtained. The training process in this embodiment may specifically be to determine a model coefficient of the resource demand model according to a training set, then obtain a loss function value according to a test set, and train and test repeatedly through iteration to minimize the loss function, thereby obtaining an optimal resource demand model.
On the basis of the above embodiment, the resource demand model may include a linear regression portion and a time influence factor portion, where the time influence factor portion includes at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the season factor portion is selected according to a fluctuation period of the resource migration record information of the target resource.
In this embodiment, a linear regression part of the resource demand model may be constructed according to a linear regression model, and since the resource demand may fluctuate greatly with time, a time influence factor part needs to be added on the basis of the linear regression model; in addition, due to the fact thatDifferent fluctuation periods of the resource demand of different target resources are different, which finally results in different fluctuation periods of the resource migration record information, that is, the resource demand of the target resources is different in sensitivity degree to months, weeks and/or dates, seasonal factors of different dimensions can be added, the seasonal factors can include at least one of month sub-factors, week sub-factors and date sub-factors, for example, for the target resources sensitive to months, 11 virtual variables (dummy) x can be addedmonthX of a month when the time series of predicted time periods is in that monthmonthValue 1, x for the remaining monthsmonthThe value is 0, and the time when the independent variable and the dependent variable occur in the data is used in the training process and the testing process, for example, the historical migration amount and the resource attribute characteristic of the target resource with the test data of 2018, 10, month and 1, xmonth101, other xmonth0. Of which 11 xmonthIt is sufficient to represent the 12 month-year-round situation, of course, the virtual variable xmonth12 can also be added; similarly, 6 week-related and 30 date-related virtual variables may be added.
More specifically, the resource requirement model in this embodiment may be as follows:
wherein y is the resource demand of the target resource; x is the number of1-xnThe resource attribute characteristics of the target resource; x is the number ofmonth_j、xweek_k、xday_lThe related variables are respectively a month sub-factor, a week sub-factor and a date sub-factor; alpha, beta and gamma are model coefficients respectively.
It should be noted that the seasonal factor may include at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the above formula lists the case where the month sub-factor, the week sub-factor, and the date sub-factor are included at the same time.
Based on the resource demand model, the step S203 of training the preliminary resource demand model according to the training set and the test set to obtain the resource demand model, as shown in fig. 3, may specifically include:
s2031, determining a model coefficient of the preliminary resource demand model according to the training set;
s2032, acquiring a residual square sum RSS between the resource demand predicted by the preliminary resource demand model and the real historical migration quantity of the target resource according to the test set;
s2033, constructing a loss function according to the residual square sum RSS and the penalty item;
s2034, repeatedly training and testing through an iterative process to minimize the loss function, thereby obtaining the resource demand model.
In this embodiment, the existing regression method may be adopted to determine the model coefficient of the resource demand model according to the training set, and details are not repeated here. In the embodiment, during testing, an RSS (Residual Sum of Squares) and a penalty term are used to construct a loss function to constrain the number of independent variables and the parameter range and reduce overfitting of the model, so that in the iterative training process according to the loss function, low-correlation independent variables can be screened and eliminated, overfitting is prevented, high-correlation independent variable combinations are retained, and accuracy and model interpretability are improved.
Specifically, in this embodiment, the resource demand model may be simplified as follows:
the residual sum of squares RSS can then be obtained by the following formula:
wherein, yiThe actual historical migration volume of the target resource of the ith test data in the test set, f (x)i) Representing input of test data into resource requirementsThe resource demand predicted by the model.
To construct a loss function according to the following formula:
wherein λ is the L1 penalty term weight and 0< λ < 1; p is the number of arguments.
In this embodiment, two penalty terms, L1 and L2, are added to the loss function:
the L1 is used for setting a model coefficient before an independent variable with low correlation with the resource demand of the target resource as 0 in the iteration process; and the L2 is used for reducing the model coefficient with a larger value in the iterative process, so as to avoid that the smaller fluctuation of the independent variable caused by the overlarge model coefficient has larger influence on the prediction result, and reduce the fluctuation of the resource demand model. In the embodiment, by adding the penalty terms L1 and L2, the method has the advantages of LASSO Regression (last Absolute Shrinkage and Selection Operator) and Ridge Regression (Ridge Regression), and the model interpretability is improved.
According to the resource demand model provided by the embodiment, the seasonal factors are added into the resource demand model, so that the resource demand can be effectively predicted, meanwhile, the penalty item is added into the loss function, the number and the parameter range of independent variables can be restricted, low-correlation independent variables are screened and eliminated, overfitting is prevented, high-correlation independent variable combinations are kept, and the accuracy and the model interpretability are improved.
On the basis of any of the above embodiments, after obtaining the resource migration record information of the target resource, before training the preliminary resource demand model according to the training set and the test set, the method may further include:
performing data preprocessing on the resource attribute characteristics of the target resource and the historical migration volume of the target resource, wherein the data preprocessing comprises the following steps: and processing missing data in the historical migration amount of the target resource, and/or performing data transformation on predetermined data in the historical migration amount of the target resource.
In this embodiment, the processing of the missing data in the historical migration amount of the target resource may specifically include:
judging whether the missing rate of the missing data exceeds a preset threshold value or not; if yes, discarding the missing data; if not, completing the missing data by a difference method;
performing data transformation on the predetermined data in the historical migration amount of the target resource, which may specifically include:
and performing direct transformation or Box-Cox transformation on the predetermined data so that the predetermined data meets a predetermined order of magnitude after being transformed. Wherein, the direct transformation includes taking logarithm, square root, reciprocal, etc. The data transformation can avoid that the magnitude of certain data is too large or too small, which causes great difference between the predicted result and the actual value.
It should be noted that, in this embodiment, the data extraction process and the data preprocessing process may be performed simultaneously, or of course, the data extraction may be performed first, and then the extracted data is preprocessed.
On the basis of the foregoing embodiment, the present embodiment provides the following specific implementation manner of the inventory management method:
after the category of the target resource is determined, the relevant data of the category is extracted through a big data platform (through hive extraction). And cleaning the original data, including missing data repairing, abnormal value processing and data conversion. And simultaneously, all characteristics are selected by combining the experience of the service side and the visual analysis of related data, and the characteristic data is cleaned. Here the python machine learning package sickit-leann will be used for the next data segmentation, model training and model testing. The cleaned data is first sliced using the train _ test _ split () function into a training set and a test set. And then introducing ElasticNet from a sklean. linear _ model, and training the resource demand data Y and the resource attribute feature data X in the training set by using a fit () function to obtain a regression model. The resource requirement value is compared with the actual value by using the resource attribute characteristic data in the test set and the predict () function resource requirement value, so that the model accuracy can be tested.
Fig. 4 is a structural diagram of an inventory management device according to an embodiment of the present invention. The inventory management device provided in this embodiment may execute the processing flow provided in the embodiment of the inventory management method, as shown in fig. 3, where the inventory management device 30 includes: a receiving module 31, a feature obtaining module 32, a processing module 33 and a sending module 34.
The receiving module 31 is configured to receive an inventory adjustment request, and analyze the inventory adjustment request to obtain resource identification information of a target resource;
a feature obtaining module 32, configured to obtain a resource attribute feature of the target resource corresponding to the resource identification information;
the processing module 33 is configured to determine a resource demand amount of the target resource according to the acquired resource attribute characteristics and the resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources;
a sending module 34, configured to feed back inventory adjustment information for the target resource, where the inventory adjustment information is determined based on the current inventory information of the target resource and the resource demand.
On the basis of the above embodiment, the processing module 33 is further configured to:
acquiring a training set and a test set according to the resource migration record information of the target resource;
constructing a preliminary resource demand model of the target resource;
and training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model, wherein the resource demand model comprises resource-related time influence factors.
On the basis of any one of the above embodiments, the time influence factor related to the resource includes at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the time influence factor related to the resource is selected according to a fluctuation cycle of the resource migration record information of the target resource.
On the basis of any of the above embodiments, the processing module 33 is configured to:
determining model coefficients of the preliminary resource demand model according to the training set;
acquiring the RSS and the square sum of the residual errors between the resource demand predicted by the preliminary resource demand model and the real historical migration quantity of the target resource according to the test set;
constructing a loss function according to the residual square sum RSS and the penalty item;
and repeatedly training and testing through an iterative process to minimize the loss function, thereby obtaining the resource demand model.
On the basis of any of the above embodiments, the resource demand model is:
wherein y is the resource demand of the target resource; x is the number of1-xnThe resource attribute characteristics of the target resource; x is the number ofmonth_j、xweek_k、xday_lThe related variables are respectively a month sub-factor, a week sub-factor and a date sub-factor; alpha, beta and gamma are model coefficients respectively.
On the basis of any of the above embodiments, the processing module 33 constructs the loss function according to the following formula:
wherein λ is L1 penalty term weight, and 0< λ < 1; p is the number of arguments.
On the basis of any of the above embodiments, the apparatus 30 further includes a data acquisition module 35, configured to:
before the preliminary resource demand model is trained according to the training set and the testing set, determining the selection time range of the resource migration record information of the target resource according to the category of the target resource and the time period required to be predicted;
and acquiring the resource migration record information of the target resource according to the selection time range.
On the basis of any of the above embodiments, the apparatus 30 further includes a data extraction module 36, configured to:
extracting the resource attribute characteristics of the target resource and the historical migration volume of the target resource according to the resource migration record information of the target resource;
the resource attribute characteristics of the target resource comprise all resource attribute characteristics which can influence the historical migration quantity of the target resource;
and segmenting the resource attribute characteristics of the target resource and the historical migration volume of the target resource into the training set and the test set.
On the basis of any of the above embodiments, the apparatus 30 further includes a data preprocessing module 37, configured to:
performing data preprocessing on the resource attribute characteristics of the target resource and the historical migration volume of the target resource, wherein the data preprocessing comprises the following steps: and processing missing data in the historical migration amount of the target resource, and/or performing data transformation on predetermined data in the historical migration amount of the target resource.
On the basis of any of the above embodiments, the data preprocessing module 37 is configured to:
judging whether the missing rate of the missing data exceeds a preset threshold value or not; if yes, discarding the missing data; if not, completing the missing data by a difference method; and/or
And performing direct transformation or Box-Cox transformation on the predetermined data so that the predetermined data meets a predetermined order of magnitude after being transformed.
The inventory management device provided in the embodiment of the present invention may be specifically configured to execute the method embodiments provided in fig. 1 to 3, and specific functions are not described herein again.
The inventory management device provided by the embodiment of the invention receives the inventory adjustment request, and analyzes the inventory adjustment request to obtain the resource identification information of the target resource; acquiring resource attribute characteristics of the target resource corresponding to the resource identification information; determining the resource demand of the target resource according to the acquired resource attribute characteristics and a resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources; feeding back inventory adjustment information for the target resource, the inventory adjustment information being determined based on current inventory information for the target resource and the resource demand. According to the embodiment of the invention, the resource demand of the target resource is obtained through the resource identification information of the target resource and the resource demand model, and the time influence factor related to the resource is considered in the model, so that the inventory management can be more scientifically and accurately carried out, the insufficient inventory or the overstock of the inventory can be effectively avoided, and the inventory cost and the operation cost are reduced.
Fig. 5 is a schematic structural diagram of an inventory management device according to an embodiment of the present invention. The inventory management device provided by the embodiment of the present invention may execute the processing flow provided by the inventory management method embodiment, as shown in fig. 5, the inventory management device 40 includes a memory 41, a processor 42, a computer program, and a communication interface 43; wherein a computer program is stored in the memory 41 and is configured to be executed by the processor 42 for performing the inventory management method as described in the above embodiments.
The inventory management device in the embodiment shown in fig. 5 may be used to implement the technical solution of the above method embodiment, and the implementation principle and technical effect are similar, which are not described herein again.
In addition, the present embodiment also provides a computer-readable storage medium on which a computer program is stored, the computer program being executed by a processor to implement the inventory management method described in the above embodiments.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
It is obvious to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to perform all or part of the above described functions. For the specific working process of the device described above, reference may be made to the corresponding process in the foregoing method embodiment, which is not described herein again.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An inventory management method, comprising:
receiving an inventory adjustment request, and analyzing the inventory adjustment request to obtain resource identification information of a target resource;
acquiring resource attribute characteristics of the target resource corresponding to the resource identification information;
determining the resource demand of the target resource according to the acquired resource attribute characteristics and a resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources;
feeding back inventory adjustment information for the target resource, the inventory adjustment information being determined based on current inventory information for the target resource and the resource demand.
2. The method of claim 1, further comprising:
acquiring a training set and a test set according to the resource migration record information of the target resource;
constructing a preliminary resource demand model of the target resource;
and training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model, wherein the resource demand model comprises resource-related time influence factors.
3. The method according to claim 2, wherein the resource-related time influence factor includes at least one of a month sub-factor, a week sub-factor, and a date sub-factor, and the resource-related time influence factor is selected according to a fluctuation period of the resource migration record information of the target resource.
4. The method of claim 3, wherein training the preliminary resource demand model according to the training set and the testing set to obtain the resource demand model comprises:
determining model coefficients of the preliminary resource demand model according to the training set;
acquiring the RSS and the square sum of the residual errors between the resource demand predicted by the preliminary resource demand model and the real historical migration quantity of the target resource according to the test set;
constructing a loss function according to the residual square sum RSS and the penalty item;
and repeatedly training and testing through an iterative process to minimize the loss function, thereby obtaining the resource demand model.
5. The method of claim 4, wherein the resource demand model is:
wherein y is the resource demand of the target resource; x is the number of1-xnThe resource attribute characteristics of the target resource; x is the number ofmonth_j、xweek_k、xday_lThe related variables are respectively a month sub-factor, a week sub-factor and a date sub-factor; alpha, beta and gamma are model coefficients respectively.
6. The method of claim 5, wherein the loss function is constructed according to the following equation:
wherein λ is L1 penalty term weight, and 0< λ < 1; p is the number of arguments.
7. The method of claim 2, wherein before obtaining the training set and the test set according to the resource migration record information of the target resource, the method further comprises:
determining the selection time range of the resource migration record information of the target resource according to the category of the target resource and the time period required to be predicted;
and acquiring the resource migration record information of the target resource according to the selection time range.
8. An inventory management device, comprising:
the receiving module is used for receiving the inventory adjustment request and analyzing the inventory adjustment request to obtain the resource identification information of the target resource;
a feature obtaining module, configured to obtain a resource attribute feature of the target resource corresponding to the resource identification information;
the processing module is used for determining the resource demand of the target resource according to the acquired resource attribute characteristics and the resource demand model; the resource demand model is obtained through resource migration record information training, and the resource demand model comprises time influence factors related to resources;
and the sending module is used for feeding back inventory adjustment information of the target resource, and the inventory adjustment information is determined based on the current inventory information of the target resource and the resource demand.
9. An inventory management device, comprising:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of any one of claims 1-7.
10. A computer-readable storage medium, having stored thereon a computer program;
the computer program, when executed by a processor, implementing the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910557464.5A CN112132498A (en) | 2019-06-25 | 2019-06-25 | Inventory management method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910557464.5A CN112132498A (en) | 2019-06-25 | 2019-06-25 | Inventory management method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112132498A true CN112132498A (en) | 2020-12-25 |
Family
ID=73849469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910557464.5A Pending CN112132498A (en) | 2019-06-25 | 2019-06-25 | Inventory management method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112132498A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045325A1 (en) * | 2021-09-26 | 2023-03-30 | 烟台杰瑞石油服务集团股份有限公司 | Resource storage and acquisition method and apparatus |
| CN116091175A (en) * | 2023-04-10 | 2023-05-09 | 南京航空航天大学 | Transaction information data management system and method based on big data |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137944A1 (en) * | 2003-12-21 | 2005-06-23 | Li-Chin Lu | Automatic inventory management system |
| CN103235822A (en) * | 2013-05-03 | 2013-08-07 | 富景天策(北京)气象科技有限公司 | Database generating and querying method |
| CN106897795A (en) * | 2017-02-17 | 2017-06-27 | 联想(北京)有限公司 | A kind of inventory forecast method and device |
| CN107730173A (en) * | 2017-10-13 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of automatic procurement practice in Mini Supermarkets based on data analysis and system |
| CN108364092A (en) * | 2018-01-29 | 2018-08-03 | 西安理工大学 | A kind of catering trade vegetable Method for Sales Forecast method based on deep learning |
| CN108764974A (en) * | 2018-05-11 | 2018-11-06 | 国网电子商务有限公司 | A kind of procurement of commodities amount prediction technique and device based on deep learning |
| CN109714395A (en) * | 2018-12-10 | 2019-05-03 | 平安科技（深圳）有限公司 | Cloud platform resource uses prediction technique and terminal device |
| CN109840734A (en) * | 2017-11-29 | 2019-06-04 | 北京京东尚科信息技术有限公司 | Information output method and device |
| CN109902850A (en) * | 2018-08-24 | 2019-06-18 | 华为技术有限公司 | Determine the method, apparatus and storage medium of Strategy of Inventory Control |
| CN109919710A (en) * | 2019-01-25 | 2019-06-21 | 广州富港万嘉智能科技有限公司 | A kind of method automatically generating procurement of commodities inventory, electronic equipment and storage medium |
-
2019
- 2019-06-25 CN CN201910557464.5A patent/CN112132498A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050137944A1 (en) * | 2003-12-21 | 2005-06-23 | Li-Chin Lu | Automatic inventory management system |
| CN103235822A (en) * | 2013-05-03 | 2013-08-07 | 富景天策(北京)气象科技有限公司 | Database generating and querying method |
| CN106897795A (en) * | 2017-02-17 | 2017-06-27 | 联想(北京)有限公司 | A kind of inventory forecast method and device |
| CN107730173A (en) * | 2017-10-13 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of automatic procurement practice in Mini Supermarkets based on data analysis and system |
| CN109840734A (en) * | 2017-11-29 | 2019-06-04 | 北京京东尚科信息技术有限公司 | Information output method and device |
| CN108364092A (en) * | 2018-01-29 | 2018-08-03 | 西安理工大学 | A kind of catering trade vegetable Method for Sales Forecast method based on deep learning |
| CN108764974A (en) * | 2018-05-11 | 2018-11-06 | 国网电子商务有限公司 | A kind of procurement of commodities amount prediction technique and device based on deep learning |
| CN109902850A (en) * | 2018-08-24 | 2019-06-18 | 华为技术有限公司 | Determine the method, apparatus and storage medium of Strategy of Inventory Control |
| CN109714395A (en) * | 2018-12-10 | 2019-05-03 | 平安科技（深圳）有限公司 | Cloud platform resource uses prediction technique and terminal device |
| CN109919710A (en) * | 2019-01-25 | 2019-06-21 | 广州富港万嘉智能科技有限公司 | A kind of method automatically generating procurement of commodities inventory, electronic equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045325A1 (en) * | 2021-09-26 | 2023-03-30 | 烟台杰瑞石油服务集团股份有限公司 | Resource storage and acquisition method and apparatus |
| CN116091175A (en) * | 2023-04-10 | 2023-05-09 | 南京航空航天大学 | Transaction information data management system and method based on big data |
| CN116091175B (en) * | 2023-04-10 | 2023-08-22 | 南京航空航天大学 | Transaction information data management system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108256898B (en) | Product sales prediction method, system and storage medium | |
| CN110991875A (en) | Platform user quality evaluation system | |
| CN112633962B (en) | Service recommendation method and device, computer equipment and storage medium | |
| US20100125487A1 (en) | System and method for estimating settings for managing a supply chain | |
| CN111727451A (en) | Information processing apparatus and the like for calculating prediction data | |
| CN110689355A (en) | Client classification method, device, computer equipment and storage medium | |
| CN112036631B (en) | Purchasing quantity determining method, purchasing quantity determining device, purchasing quantity determining equipment and storage medium | |
| CN107958297A (en) | A kind of product demand forecasting method and product demand prediction meanss | |
| CN112132498A (en) | Inventory management method, device, equipment and storage medium | |
| EP3706369B1 (en) | Sequence, frequency, and time interval based journey recommendation | |
| CN109492863A (en) | The automatic generation method and device of financial document | |
| CN115545828A (en) | E-commerce data push analysis system based on artificial intelligence | |
| CN111274531A (en) | Commodity sales amount prediction method, commodity sales amount prediction device, computer equipment and storage medium | |
| CN114782123A (en) | Credit assessment method and system | |
| Chen-Ritzo et al. | Component rationing for available-to-promise scheduling in configure-to-order systems | |
| CN112990703A (en) | International engineering market matching degree evaluation method, electronic device and storage medium | |
| CN112150179B (en) | Information pushing method and device | |
| CN116596547A (en) | Client relationship management method and system based on multidimensional information data | |
| CN110689177A (en) | Method and device for predicting order preparation time, electronic equipment and storage medium | |
| WO2023017244A1 (en) | A system and method for selecting a service supplier | |
| CN107545347B (en) | Attribute determination method and device for risk prevention and control and server | |
| CN113743994A (en) | Provider's season-busy prediction method, system, equipment and storage medium | |
| CN114693243A (en) | Enterprise platform management method, system and storage medium | |
| US20020128858A1 (en) | Method and system for population classification | |
| CN114548620A (en) | Logistics punctual insurance service recommendation method and device, computer equipment and storage medium |
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 |