The content of the invention
Inventor has found that prior art carries out cutting single scheme to the order being cancelled, and wastes human and material resources,
And if cutting list unsuccessfully causes commodity ex-warehouse, bigger loss will be further resulted in.
A technical problem to be solved by this invention is:How damage that cancellation of order to order production process bring is reduced
Lose.
According to one embodiment of present invention there is provided a kind of order processing method, including:According to various business in order
The information attribute value prediction order of product is cancelled probability;According to each order to be cancelled the ascending order of probability right
Each order is ranked up;The production of each order is arranged according to the sequence of each order.
In one embodiment, predict that order is cancelled probability packet according to the information attribute value of extensive stock in order
Include:The information attribute value of extensive stock is respectively inputted into disaggregated model obtain extensive stock and be cancelled probability;According to order
In include extensive stock be cancelled the determine the probability order be cancelled probability.
In one embodiment, disaggregated model is determined using following methods:Obtain the business of the extensive stock in History Order
The production status of product attribute information and History Order is as training data, and production status includes being cancelled and not cancelling;Utilize
Training data is trained determination disaggregated model to disaggregated model.
In one embodiment, using training data disaggregated model is trained including:Have according to preset ratio scope
That puts back to randomly selects the part information attribute value of extensive stock, and is combined with corresponding production status as one certainly
The training data of plan tree, is trained to the decision tree and determines the decision tree;Said process is repeated, until determining in disaggregated model
Whole decision trees.
In one embodiment, the information attribute value of extensive stock is inputted into disaggregated model respectively and obtains extensive stock
Being cancelled probability includes:Choosing information attribute value input corresponding with the decision tree in disaggregated model for every kind of commodity should be certainly
Plan tree-model, obtains classification of the commodity in the decision tree;According to commodity in the classification of each decision tree and the sum of decision tree
Determine commodity is cancelled probability.
In one embodiment, the probability that is cancelled of order is subtracted respectively for the probability sum that is cancelled of extensive stock in order
Plant commodity and be cancelled probability simultaneously.
According to another embodiment of the invention there is provided a kind of order processing unit, including:Order probabilistic forecasting mould
Block, for predicting that order is cancelled probability according to the information attribute value of extensive stock in order;Order Sorting module, is used for
Each order is ranked up according to the ascending order of probability that is cancelled of each order;Order scheduling of production module, is used
The production of each order is arranged in the sequence according to each order.
In one embodiment, order probabilistic forecasting module, for the information attribute value of extensive stock to be inputted respectively
Disaggregated model obtains extensive stock and is cancelled probability, being cancelled determine the probability this is ordered according to the extensive stock that is included in order
Single is cancelled probability.
In one embodiment, the processing unit also includes:Disaggregated model determining module, for obtaining in History Order
The information attribute value of extensive stock and the production status of History Order as training data, production status include being cancelled and
Do not cancel, determination disaggregated model is trained to disaggregated model using training data.
In one embodiment, disaggregated model determining module, for thering is that puts back to randomly select according to preset ratio scope
The part information attribute value of extensive stock, and it is combined the training number as a decision tree with corresponding production status
According to, the decision tree is trained and determines the decision tree, said process is repeated, whole decision-makings in determination disaggregated model
Tree.
In one embodiment, order probabilistic forecasting module, for being chosen and determining in disaggregated model for every kind of commodity
The corresponding information attribute value of plan tree inputs the decision-tree model, classification of the commodity in the decision tree is obtained, according to commodity each
What the classification of decision tree and the sum of decision tree determined commodity is cancelled probability.
In one embodiment, the probability that is cancelled of order is subtracted respectively for the probability sum that is cancelled of extensive stock in order
Plant commodity and be cancelled probability simultaneously.
According to still another embodiment of the invention there is provided a kind of order processing unit, including:Memory;And coupling
The processor of memory is connected to, processor is configured as, based on the instruction being stored in memory devices, performing as foregoing any
The processing method of order in individual embodiment.
According to still a further embodiment there is provided a kind of computer-readable recording medium, be stored thereon with calculating
Machine program, the program realizes the processing method of the order in any one foregoing embodiment when being executed by processor the step of.
The present invention is cancelled probability based on the attribute forecast order of extensive stock in order, is taken according to each order
Disappear the production of probability scheduling order.It is cancelled after the big order of probability is arranged at and produces, the production for reducing such order is excellent
First level so that such order is likely to not enter into production process when being cancelled, and reduces the waste of human and material resources, together
When ensure that the continuity of production operation flow process in whole warehouse, improve production efficiency.
By referring to the drawings to the detailed description of the exemplary embodiment of the present invention, further feature of the invention and its
Advantage will be made apparent from.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Below
Description only actually at least one exemplary embodiment is illustrative, is never used as to the present invention and its application or makes
Any limitation.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, belongs to the scope of protection of the invention.
For that can only carry out cutting a single scheme, wave after order produces completion for the order being cancelled in the prior art
The problem of having taken human and material resources, propose this programme.
The processing unit of order in embodiments of the invention can realize respectively by various computing devices or computer system,
It is described with reference to Fig. 1 and Fig. 2.
Fig. 1 is the structure chart of one embodiment of the processing unit of order of the present invention.As shown in figure 1, the dress of the embodiment
Putting 10 includes:Memory 110 and the processor 120 for being coupled to the memory 110, processor 120 are configured as based on storage
Instruction in the memory 110, performs the processing method of the order in the present invention in any one embodiment.
Wherein, memory 110 is such as can include system storage, fixed non-volatile memory medium.System is stored
Device is such as the operating system that is stored with, application program, Boot loader (Boot Loader), database and other programs.
Fig. 2 is the structure chart of another embodiment of the processing unit of order of the present invention.As shown in Fig. 2 the embodiment
Device 10 includes:Memory 110 and processor 120, can also include input/output interface 230, network interface 240, storage
Interface 250 etc..It can for example pass through bus 260 between these interfaces 230,240,250 and memory 110 and processor 120
Connection.Wherein, input/output interface 230 be display, the input-output equipment such as mouse, keyboard, touch-screen connecting interface is provided.
Network interface 240 provides connecting interface for various networked devices, for example, may be coupled to database server or high in the clouds storage
Server etc..The external storages such as memory interface 250 is SD card, USB flash disk provide connecting interface.
The present invention provides a kind of processing method of order, is described with reference to Fig. 3.
Fig. 3 is the flow chart of processing method one embodiment of order of the present invention.As shown in figure 3, the method for the embodiment
Including:
Step S302, predicts that order is cancelled probability according to the information attribute value of extensive stock in order.
Comprising one or more commodity in one order, every kind of commodity have its corresponding information attribute value, item property
Information is such as including commodity sign, cargo price, quantity purchase, merchandise classification, the time that places an order, the place that places an order, further business
Product attribute information can also include the information of corresponding purchase user, such as user's mark, user credit degree.
User credit degree can be determined according to the purchaser record of user.The user credit that can be determined for a certain commodity
Degree, for example, obtaining the History Order and corresponding production status of user, production status includes being cancelled and not cancelling;Then
The extensive stock and its corresponding production status in History Order are extracted, the number of times being cancelled according to commodity and total order numbers
Ratio determines the user credit degree for the commodity.Can also overall assessment user credit degree, gone through for example, obtaining user first
History order and corresponding production status, production status include being cancelled and not cancelling;Then it is cancelled and is ordered according to user's history
Odd number and the ratio of user's total orders determine user credit degree.
Step S304, is ranked up according to the ascending order of probability that is cancelled of each order to each order.
Step S306, the production of each order is arranged according to the sequence of each order.
The method of above-described embodiment is cancelled probability based on the attribute forecast order of extensive stock in order, according to each
The production for being cancelled probability scheduling order of order.It is cancelled after the big order of probability is arranged at and produces, reduces such and order
Single production priority so that such order is likely to not enter into production process when being cancelled, reduces man power and material's money
The waste in source, while ensure that the continuity of the production operation flow process in whole warehouse, improves production efficiency.
For how according to the information attribute value of extensive stock in order to predict that order is taken in above-mentioned steps S302
Disappear probability the problem of, the present invention provide following exemplary embodiment:
It is preferred that, extract the information attribute value of extensive stock in order;The information attribute value of extensive stock is distinguished
Input disaggregated model acquisition extensive stock is cancelled probability;Determine the probability is cancelled according to the extensive stock included in order
The order is cancelled probability.
In one embodiment, the information attribute value of extensive stock is divided into feature attribute of commodity information according to feature
With commodity customer attribute information;The feature attribute of commodity information of extensive stock is inputted into disaggregated model respectively and obtains extensive stock
First is cancelled probability, and determine extensive stock according to commodity customer attribute information second is cancelled probability, for every kind of commodity
It is cancelled probability and second to first to be cancelled probability and be weighted summation, obtain the commodity is cancelled probability;According to order
In include extensive stock be cancelled the determine the probability order be cancelled probability.Wherein, second it is cancelled probability and is, for example,
For the user credit degree of the commodity.
Further, the probability sum that is cancelled that probability is, for example, extensive stock in order that is cancelled of order subtracts various business
Product are cancelled probability simultaneously.Extensive stock is cancelled the product for being cancelled probability that probability is, for example, extensive stock simultaneously.
Disaggregated model is, for example, decision-tree model or model-naive Bayesian, and decision-tree model is, for example, Random Forest model
Or gradient lifting decision-tree model etc..Above-mentioned model is prior art, be will not be repeated here.
Using disaggregated model obtain extensive stock be cancelled probability before, it is possible to use History Order data to classification
Model, which is trained, obtains disaggregated model.It is described with reference to Fig. 4.
Fig. 4 is the flow chart of another embodiment of the processing method of order of the present invention.As shown in figure 4, step S302 it
Before can also include:
The information attribute value and the production status of History Order of extensive stock in step S402, acquisition History Order
It is used as training data.
Production status includes being cancelled and not cancelling.Production status can carry out binary conversion treatment, for example, 0 represents not take
Disappear, 1 represents to be cancelled.
Step S404, determination disaggregated model is trained to disaggregated model using training data.
Specifically, when disaggregated model is made up of many decision trees, thering is that puts back to randomly select according to preset ratio scope
The part information attribute value of extensive stock, and it is combined the training number as a decision tree with corresponding production status
According to, to the decision tree be trained determination the decision tree;Said process is repeated, until determining whole decision-makings in disaggregated model
Tree.
In extraction History Order after the information attribute value and production status of extensive stock, it can enter according to the form of table 1
Row storage.
Table 1
The information attribute value and corresponding production status of a kind of commodity are represented in table 1 per a line.Wherein, wherein E1~Ei
It is independent variable, (i is positive integer, and information attribute value is planted in representing i-th), D is dependent variable.When disaggregated model is by many decision trees
During composition, partial data can be extracted according to table 1 and generate multiple training subsets, each training subset as 1 decision tree instruction
Practice data.Specifically, having the column data of the part of the representative information attribute value in the extraction table 1 put back to, with corresponding production
Combinations of states as a training subset, in each training subset the species of information attribute value be, for example, total species 10%~
90%.Data in each training subset are used to build a decision tree, until building whole decision-makings in disaggregated model
Tree, and then obtain disaggregated model.
For example, for a decision tree, cargo price, quantity purchase and the production status for choosing extensive stock are combined
Obtain the training data of the decision tree;For another decision tree, the time that places an order, quantity purchase and the production of extensive stock are chosen
State is combined the training data for obtaining the decision tree.
Understood based on above-mentioned training process, every decision tree corresponds to different information attribute values, therefore, it is determined that business
When being cancelled probability of product, can be in the following ways:
Information attribute value corresponding with the decision tree in disaggregated model, which is chosen, for every kind of commodity inputs the decision tree mould
Type, obtains classification of the commodity in the decision tree;According to commodity business is determined in the classification of each decision tree and the sum of decision tree
Product are cancelled probability.
In extraction order after the information attribute value and production status of extensive stock, it can be deposited according to the form of table 2
Storage.
Table 2
For the data input decision tree of the corresponding row of every kind of commodity in different decision tree selection tables 2, various business are obtained
Classification of the product in each decision tree, a kind of probability that is cancelled of commodity is, for example, to be divided into be cancelled class in disaggregated model
Number of times divided by decision tree sum.
For example, it is desirable to which obtain the commodity that commodity ID is 312341 is cancelled probability, first, for first decision tree,
The decision tree is built according to two information attribute values of cargo price and quantity purchase, then by 312341 cargo price
2.15 and quantity purchase 5 input first decision tree, obtain 312341 be categorized as be cancelled class, with reference to said process, obtain
312341 each decision tree classification, for example, 312341 being divided into 40 decision trees and being cancelled class, in 10 are set
312341 are divided into and do not cancel class, then 312341 probability that is cancelled is 80%.
Disaggregated model can also be updated using newly-increased History Order data every predetermined period in the present invention.
The present invention also provides a kind of processing unit of order, is described with reference to Fig. 5.
Fig. 5 is the structure chart of processing unit one embodiment of order of the present invention.As shown in figure 5, the device 50 includes:
Order probabilistic forecasting module 502, the quilt for predicting order according to the information attribute value of extensive stock in order
Cancel probability.
It is preferred that, order probabilistic forecasting module 502, for the information attribute value of extensive stock to be inputted into classification mould respectively
Type obtains extensive stock and is cancelled probability, according to the quilt for being cancelled the determine the probability order of the extensive stock included in order
Cancel probability.
Further, order probabilistic forecasting module 502, for being chosen and the decision tree pair in disaggregated model for every kind of commodity
The information attribute value answered inputs the decision-tree model, classification of the commodity in the decision tree is obtained, according to commodity in each decision-making
What the sum of the classification of tree and decision tree determined commodity is cancelled probability.
It is preferred that, to subtract extensive stock same for the probability sum that is cancelled of extensive stock in order for the probability that is cancelled of order
When be cancelled probability.
Order Sorting module 504, for being cancelled the ascending order of probability to each order according to each order
It is ranked up.
Order scheduling of production module 506, the production of each order is arranged for the sequence according to each order.
In one embodiment, the processing unit 50 of order can also include:
Disaggregated model determining module 508, for obtaining the information attribute value of the extensive stock in History Order and going through
The production status of history order is as training data, and production status includes being cancelled and not cancelling, using training data to classification mould
Type is trained determination disaggregated model.
It is preferred that, disaggregated model determining module 508, for thering is that puts back to randomly select various business according to preset ratio scope
The part information attribute value of product, and the training data as a decision tree is combined with corresponding production status, to this
Decision tree, which is trained, determines the decision tree, repeats said process, until determining whole decision trees in disaggregated model.
The present invention also provides a kind of computer-readable recording medium, is stored thereon with computer program, and the program is processed
Device realizes the processing method of the order in any one foregoing embodiment when performing the step of.
Those skilled in the art should be understood that embodiments of the invention can be provided as method, system or computer journey
Sequence product.Therefore, in terms of the present invention can be using complete hardware embodiment, complete software embodiment or combination software and hardware
The form of embodiment.Moreover, the present invention can be used in one or more calculating for wherein including computer usable program code
Machine can use the meter implemented on non-transient storage medium (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
The present invention is the flow with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Figure and/or block diagram are described.Being interpreted as can be by each in computer program instructions implementation process figure and/or block diagram
Flow and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer journeys can be provided
Sequence instruction to all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices processor with
Produce a machine so that being produced by the instruction of computer or the computing device of other programmable data processing devices is used for
Realize the dress for the function of being specified in one flow of flow chart or multiple flows and/or one square frame of block diagram or multiple square frames
Put.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.