CN116866281A

CN116866281A - Flow balancing method and device, computer equipment and storage medium

Info

Publication number: CN116866281A
Application number: CN202310785244.4A
Authority: CN
Inventors: 陈梦洲
Original assignee: Vipshop Guangzhou Software Co Ltd
Current assignee: Vipshop Guangzhou Software Co Ltd
Priority date: 2023-06-29
Filing date: 2023-06-29
Publication date: 2023-10-10

Abstract

The application relates to a flow balancing method, a flow balancing device, computer equipment and a storage medium. The flow balancing method comprises the following steps: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; selecting at least partial commodity subsets which do not complete the flow tasks from all the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to all the supporting commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; processing the flow supporting degree of each flow parameter to obtain each second supporting factor; constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; weighting the support subset with support parameters. The method can macroscopically regulate and control the flow, so as to be beneficial to reasonably distributing the flow and balancing the flow for multitasking.

Description

Flow balancing method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of information processing technologies, and in particular, to a flow balancing method, a flow balancing device, a computer device, and a computer readable storage medium.

Background

Traffic/exposure is a more precious resource in the e-commerce field. The e-commerce platform is often cooperated with a plurality of merchants, and each merchant has flow requirements adapting to the conditions of the merchant, and the e-commerce platform is used as a flow inlet and a distribution center and needs to coordinate flow distribution of each merchant by utilizing a certain strategy.

However, in the existing flow distribution strategy, the adjustment and control is generally performed for a commodity/merchant/class, that is, the adjustment and control object itself is microscopically adjusted and controlled, which is most likely to cause unbalanced flow distribution. In popular terms, a certain regulation object may distribute more flow to easily realize a flow task, and a certain regulation object may not obtain sufficient flow all the time.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a flow balancing method, a flow balancing device, a computer device, and a computer-readable storage medium that can macroscopically regulate and control flow to facilitate reasonable distribution of flow, balance flow, and multitasking.

In one aspect, a flow balancing method is provided, the flow balancing method including: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; the standard reaching task is to complete a flow task for supporting a commodity subset of a target number in a commodity set; selecting at least partial commodity subsets which do not complete the flow tasks from all the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to all the commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; processing the flow supporting degree of each flow parameter to obtain each second supporting factor; constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong; weighting the support subset with support parameters.

In one embodiment of the application, constraining the second support factor with the first support factor matched thereto includes: and carrying out weighted fusion on the second supporting factor and the first supporting factor matched with the second supporting factor to obtain supporting parameters, and controlling the supporting parameters to be not larger than the constraint parameters.

In one embodiment of the present application, the first supporting factor is taken as a constraint parameter; controlling the support parameter not to be greater than the constraint parameter includes: and in response to the support parameter being greater than the first support factor, replacing the support parameter with the first support factor to form a new support parameter.

In one embodiment of the present application, constraining the second supporting factor with the first supporting factor matched with the first supporting factor to obtain the supporting parameter includes: comparing the second supporting factor with the first supporting factor matched with the second supporting factor; in response to the second support factor being not greater than the first support factor, taking the second support factor as a support parameter; and responding to the second supporting factor not smaller than the first supporting factor, and taking the first supporting factor as supporting parameter.

In one embodiment of the present application, at least one of the progress-support process and the flow-support process comprises a control algorithm process; the control algorithm processing comprises the following steps: obtaining a control quantity; the control quantity is the progress control quantity obtained based on the progress parameter/the flow control quantity obtained based on the flow parameter; the control quantity and a plurality of control quantities of the same type which are acquired in advance are subjected to control algorithm operation; and superposing the operation result and the previous control factor to obtain the current control factor, and taking the current control factor as the first supporting factor or the second supporting factor.

In one embodiment of the application, the flow task is that the commodity subset reaches the target flow, and the flow parameter comprises the current flow of the commodity subset; the control amount is a flow control amount obtained based on the flow parameter, including: based on the current flow and the target flow, obtaining the ratio of the flow to be completed to the target flow, and taking the ratio as a flow control quantity; the progress parameters comprise the subset number of commodity subsets supporting the commodity set to complete the flow task; the control amount is a progress control amount obtained based on the progress parameter, and includes: based on the number of subsets and the target number, the number to be completed is obtained, and the number to be completed is used as a control quantity.

In one embodiment of the present application, weighting the subset of handoffs with the handoffs parameters is preceded by: obtaining the amplification factor of the supporting parameter by utilizing the balance parameter, and amplifying the supporting parameter; wherein, in each supporting subset, the balance parameter is obtained by the flow parameter; taking zero from the commodity subsets which are not taken as supporting subsets; selecting intermediate values of the amplified supporting parameters, the supporting upper limit parameters and the supporting lower limit parameters as new supporting parameters; wherein, the upper supporting limit parameter and the lower supporting limit parameter are both acquired in advance.

In one embodiment of the present application, the progress-support process and the flow-support process each include a control algorithm process, and the control algorithm is a PID algorithm.

In one embodiment of the present application, selecting at least a portion of the subset of articles from the set of supported articles that do not complete the flow task as the supported subset comprises: randomly distributing selection parameters for a commodity subset supporting incomplete flow tasks in a commodity set; respectively comparing the selection parameters of the commodity subsets with preset parameters; and taking the commodity subset as the supporting subset in response to the selection parameter meeting the preset condition.

In one embodiment of the application, the progress parameter includes a subset number of subsets of the commodity that support the commodity set to complete the flow task; selecting at least part of commodity subsets which do not complete the flow tasks from the supporting commodity set as supporting subsets further comprises: obtaining a difference value between the target quantity and the progress parameter as a demand quantity; comparing the number of the supporting subsets with the required number; responding to the fact that the number of the supporting subsets is smaller than the required number, and selecting the commodity subsets which are not used as the supporting subsets by combining balance parameters of the commodity subsets which are not used as the supporting subsets until the number of the supporting subsets is equal to the required number; wherein the balance parameter is obtained based on the flow parameter.

In one embodiment of the present application, selecting the subset of items not being a supporting subset as a supporting subset in combination with the balance parameters of the subset of items not being a supporting subset comprises: arranging the commodity subsets in descending order according to the balance parameters; and sequentially selecting the commodity subsets which are not used as the supporting subsets from the head end of the queue as the supporting subsets.

In an embodiment of the present application, the comparing the number of supporting subsets with the required number further includes: in response to the number of support subsets being greater than the required number, the support subsets are filtered such that the number of support subsets is equal to the required number.

In one embodiment of the present application, the balance parameter is obtained based on the flow parameter, including: weighting the flow parameters by using the balance factors to obtain balance parameters; the balance factor is obtained based on the flow parameter and the average completion rate in a preset period.

In an embodiment of the present application, the average completion rate is an average value of flow completion rates of each preset sub-period in the preset period; the flow completion rate is a ratio of an actual flow rate of the subset of the commodities to a target flow rate within a preset sub-period.

In one embodiment of the present application, in response to the traffic completion rate being greater than the preset completion rate, the preset completion rate is taken as a new traffic completion rate; and/or, reducing the dimension of the flow parameter.

In an embodiment of the present application, after obtaining the balance parameter, the method further includes: and carrying out standard deviation normalization processing on the balance parameters to update the balance parameters.

In an embodiment of the present application, obtaining progress parameters of executing up-to-standard tasks for a plurality of supporting commodity sets includes: acquiring progress parameters of executing up-to-standard tasks of each commodity set; and selecting at least part of the commodity sets which do not meet the standard and serve as supporting commodity sets.

In one embodiment of the application, weighting the subset of handoffs with the handoffs parameters includes: acquiring fine arrangement and regulation factors of all commodities in all commodity subsets; taking the product of the fine discharge and the regulation parameters as a final score, and arranging all commodities in a final score descending manner; the regulation factor is obtained by reducing the dimension of the ratio of the acquired fine discharge maximum value to the fine discharge minimum value; the regulation and control parameters are obtained by taking a preset value as a base number and taking the product of the supporting parameters and the regulation and control factors as a power to perform exponential operation.

In another aspect, a flow balancing device is provided, the flow balancing device comprising: a data layer and a control module; the data layer is used for counting flow parameters of all commodity subsets; the control module is used for realizing the following steps: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; the standard reaching task is to complete a flow task for supporting a commodity subset of a target number in a commodity set; selecting at least partial commodity subsets which do not complete the flow tasks from all the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to all the supporting commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; processing the flow supporting degree of each flow parameter to obtain each second supporting factor; constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong; weighting the support subset with support parameters.

In yet another aspect, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; the standard reaching task is to complete a flow task for supporting a commodity subset of a target number in a commodity set; selecting at least partial commodity subsets which do not complete the flow tasks from all the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to all the supporting commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; processing the flow supporting degree of each flow parameter to obtain each second supporting factor; constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong; weighting the support subset with support parameters.

In yet another aspect, a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; the standard reaching task is to complete a flow task for supporting a commodity subset of a target number in a commodity set; selecting at least partial commodity subsets which do not complete the flow tasks from all the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to all the supporting commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; processing the flow supporting degree of each flow parameter to obtain each second supporting factor; constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong; weighting the support subset with support parameters.

The flow balancing method, the flow balancing device, the computer equipment and the computer readable storage medium divide commodities into a plurality of commodity subsets, and divide the commodity subsets into a plurality of commodity sets. Each commodity set has a standard-reaching task and each commodity subset has a flow task to balance flow distribution by using multi-level tasks. And the first supporting factors of the supporting commodity sets are utilized to restrict the second supporting factors of the supporting subsets, and supporting parameters for supporting the supporting subsets are obtained, so that the flow is macroscopically regulated and controlled as a whole, the flow is reasonably distributed, and the completion of the flow tasks of all commodity subsets and the completion of the standard-reaching tasks of the commodity sets are promoted.

Drawings

FIG. 1 is a diagram of an application environment for one embodiment of a flow balancing method of the present application;

FIG. 2 is a flow chart of an embodiment of a flow balancing method according to the present application;

FIG. 3 is a flow chart of another embodiment of the flow balancing method of the present application;

FIG. 4 is a flowchart of an embodiment of a method for selecting a support subset according to the present application;

FIG. 5 is a schematic diagram of an embodiment of a flow balancing device according to the present application;

FIG. 6 is a schematic view of another embodiment of a flow balancing device of the present application;

FIG. 7 is a schematic diagram of a computer device according to an embodiment of the present application.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The flow balancing method provided by the application can be applied to an application environment shown in fig. 1, and fig. 1 is an application environment diagram of an embodiment of the flow balancing method of the application.

The terminal 102 communicates with the server 104 via a network.

The server 104 performs a flow balancing method to obtain supporting parameters, and weights the supporting subsets by using the supporting parameters. The server 104 feeds back the supporting action to the terminal 102, which is shown as disturbing the original page ordering in the terminal 102, and reorders the pages according to the calculation result of the server 104 flow balancing method, so as to promote each commodity subset to complete the flow task and the commodity set to complete the standard task.

The terminal 102 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers, and portable wearable devices, and the server 104 may be implemented by a stand-alone server or a server cluster composed of a plurality of servers.

In an embodiment, a flow balancing method is provided, as shown in fig. 2, and fig. 2 is a flow chart of an embodiment of the flow balancing method of the present application. Taking the application of the flow balancing method to the server in fig. 1 as an example, the method comprises the following steps:

s201: acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; and the standard reaching task is to complete the flow task for the subset of the commodities supporting the target quantity in the commodity set.

In this embodiment, each commodity set includes a plurality of commodity subsets, and each commodity subset includes a plurality of commodities.

For example, the commodity set may be a business tile and the commodity subset may be a pool. In particular, subsets of goods may be grouped into pools of goods by "brands x radicals". The parts may be general parts such as upper and lower parts, or refined parts such as shirts, jeans, sport pants, casual shoes, sport shoes, etc., and are not limited herein. Business tiles may be delimited modules that may be partitioned by brand or family or other rules such as importance. Alternatively, the commodity set may be a commodity pool, and the commodity subset may be each commodity in the commodity pool, which is not limited herein.

And each commodity set has a standard task, and each commodity subset has a flow task, so that a multi-target task is formed. In popular terms, a commodity/commodity subset has a traffic task, and it is also necessary to promote the belonging commodity set to complete the up-to-standard task.

As the name implies, the flow task is that the commodity subset needs to reach a certain flow exposure, and the progress parameter can represent the progress of the current commodity subset in carrying out the flow subset, which can be a numerical representation of the current flow, a proportion of the current flow to the required flow, and the like.

And the up-to-standard task is to complete the flow task for the commodity subset with the target number in the commodity set. The target number may be an integer number, a ratio, or the like. The progress parameter can represent the progress of the up-to-standard task performed by the current commodity set, and can be the number of commodity subsets currently completing the flow task, the proportion of the commodity subsets currently completing the flow task to the number of all commodity subsets in the commodity set, and the like.

The supporting commodity set is a commodity set which does not complete the standard-reaching task, namely, at least part of the commodity set which does not complete the standard-reaching task is selected as the supporting commodity set to carry out flow supporting. And acquiring flow parameters of a plurality of commodity subsets of each supporting commodity set for executing a flow task, so as to balance and distribute the flow of each commodity subset by combining the completion degree of the standard-reaching task and the completion degree of the flow task, thereby carrying out macroscopic regulation and control on the flow, and being beneficial to reasonably distributing the flow.

In general, it is understood that the total flow of daily access to the platform may be referred to as the large disk flow, and the flow in the present application may refer to the amount of access to the subset of goods.

S202: and selecting at least part of the commodity subsets which do not complete the flow task from the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to the supporting commodity sets.

In this embodiment, in response to obtaining the flow parameters of each subset of the commodities in the supporting commodity set, the subset of the commodities in the commodity set that do not complete the flow tasks can be learned, and at least a portion of the subset of the commodities that do not complete the flow tasks is selected as the supporting subset.

The subset of the commodities that do not complete the flow tasks may be selected as the supporting subset, or a certain number of subsets of the commodities may be selected as the supporting subset by combining the progress parameter and the target number, which is not limited herein.

Selecting the subset of the commodities which are at least partially complete the flow task as the supporting subset means that the subset of the commodities which are complete the flow task is allowed to be included in the supporting subsets, but the subset of the commodities which are complete in flow must exist in the supporting subsets of the supporting commodity set, so that the weighting supporting of the subset of the commodities which are complete in flow task is guaranteed effectively, and the realization of the flow task and the standard reaching task is promoted.

S203: performing progress support degree processing on each progress parameter to obtain each first support factor; and carrying out flow supporting degree processing on each flow parameter to obtain each second supporting factor.

In this embodiment, the supporting forces of the two self-attributes are obtained according to the properties of each supporting commodity set and each supporting subset, so that the actual supporting forces of the supporting subsets can be balanced by combining the attributes of the supporting commodity sets.

And performing progress supporting degree processing on the progress parameters of the supporting commodity sets to obtain first supporting factors belonging to the supporting commodity sets. And carrying out flow supporting degree processing on the flow parameters of each supporting subset to obtain second supporting factors belonging to each supporting subset.

S204: constraining the second supporting factors of each supporting subset by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong.

In this embodiment, for each supporting subset, the second supporting factor is constrained by the first supporting factor matched with the first supporting factor, so as to obtain the constrained supporting parameter.

The first supporting factor matched with the supporting subset refers to the first supporting factor of the supporting commodity set to which the supporting subset belongs.

And constraining the second supporting factor by using the first supporting factor, and controlling the value of the second supporting factor by using the first supporting factor to obtain supporting parameters, so as to macroscopically control the flow of the commodity set and the commodity subset.

That is, the first supporting factor is used for restraining the second supporting factor to macroscopically regulate and control the distribution of the flow, which is equivalent to the fact that the properties of the supporting commodity sets are used for restraining the flow distributed among the supporting commodity sets, so that the supporting subsets are promoted to realize the flow tasks, and meanwhile, the standard reaching tasks of the commodity sets are considered.

S205: and weighting the supporting subset by using the supporting parameters.

In this embodiment, the subset of handoffs is weighted with the handoffs parameters after constraining the second handoffs by the first handoffs. If so, the flow distribution can be macroscopically regulated and controlled, the reasonable flow distribution is facilitated, the completion of the flow tasks by all supporting subsets is promoted, the completion of the standard-reaching tasks by all supporting commodity sets is promoted, and the flow is balanced and simultaneously the multi-level tasks are carried out.

In the flow balancing method, each commodity set has a standard task, and each commodity subset has a flow task, so that the flow needs to balance multi-level tasks. And constraining the second supporting factors of the supporting subset by using the first supporting factors of the supporting commodity set to obtain supporting parameters for supporting the supporting subset. If so, the standard task of each supporting commodity set can be considered while supporting the subset to promote the subset to complete the flow task. The supporting subsets contained below the supporting commodity sets are restrained among each other, so that the flow is macroscopically regulated and controlled on the whole, the flow is reasonably distributed, and the commodity subsets are promoted to complete the flow task and the commodity sets complete the standard-reaching task.

In one embodiment, another flow balancing method is provided, as shown in fig. 3, and fig. 3 is a flow chart of another embodiment of the flow balancing method of the present application.

S301: and acquiring the progress parameters of each commodity set for executing the up-to-standard task.

In this embodiment, the commodity set is a set composed of a plurality of commodities, where the plurality of commodities are further divided into a plurality of commodity subsets according to a certain condition, and the number of commodity subsets included in each commodity set is allowed to be different.

The standard reaching task is to complete the flow task for the commodity subsets supporting the target quantity in the commodity set, and the flow tasks of the commodity subsets are allowed to be different.

The progress parameter includes a subset number of subsets of the commodity that support the commodity set to complete the flow task.

The step of obtaining the progress parameter of each commodity set for executing the standard-reaching task may be to obtain the progress parameter periodically with the preset period as an interval, and support the commodity set by using the current support parameter obtained by processing until the next support parameter is obtained.

Optionally, the step of obtaining the progress parameter of each commodity set may be to obtain the flow of each commodity first, sequentially count the flow parameters of the commodity subsets for executing the flow tasks according to the commodity subsets and the classifications of the commodity sets, and count the number of the commodity subsets for completing the flow tasks in each commodity set based on the flow parameters to obtain the progress parameter. Statistics may also be performed in other ways, and are not limited herein.

For example, 10000 products are shared, 500 product pools are classified according to the brand x department as product subsets, and 500 product subsets are classified according to importance degrees to form 5 service plates as product sets, namely a product set A, a product set B, a product set C, a product set D and a product set E.

The commodity set A, the commodity set B, the commodity set C, the commodity set D and the commodity set E comprise 100 commodity subsets, 200 commodity subsets, 80 commodity subsets, 70 commodity subsets and 50 commodity subsets respectively.

The target numbers of the commodity set A, the commodity set B, the commodity set C, the commodity set D and the commodity set E are respectively 90%, 80%, 70%, 60% and 50%.

That is, the up-to-standard task of the commodity set a requires 90 (100×90%) commodity subsets to complete the flow tasks, and so on, the up-to-standard task of the commodity set B requires 160 commodity subsets to complete the flow tasks, the up-to-standard task of the commodity set C requires 56 commodity subsets to complete the flow tasks, the up-to-standard task of the commodity set D requires 42 commodity subsets to complete the flow tasks, and the up-to-standard task of the commodity set E requires 25 commodity subsets to complete the flow tasks.

For ease of understanding, this example will be referred to directly below for illustration, but this implementation is merely by way of example and not by way of limitation to the specific implementation of this embodiment.

S302: and selecting at least part of the commodity sets which do not meet the standard and serve as supporting commodity sets.

In this embodiment, whether each commodity set completes the up-to-standard task may be determined according to the progress parameter. For each commodity set, the number of commodity subsets completing the flow tasks in the progress parameters can be compared with the target number of the standard-reaching tasks so as to judge whether the standard-reaching tasks are completed.

And selecting a part of the commodity sets which do not meet the standard as supporting commodity sets, or taking all the commodity sets which do not meet the standard as supporting commodity sets, wherein the method is not limited. If so, in step S301 and step S302, a plurality of progress parameters for supporting the commodity set to execute the up-to-standard task may be obtained.

For example, the current progress parameters of the commodity set a, the commodity set B, the commodity set C, the commodity set D, and the commodity set E are respectively 80, 170, 45, 20, and 24. That is, the commodity set a, the commodity set C, the commodity set D and the commodity set E do not meet the standard, and at least one commodity set may be selected from the commodity set a, the commodity set C, the commodity set D and the commodity set E as the supporting commodity set. For example, the commodity set A, the commodity set C, the commodity set D and the commodity set E can be all used as supporting subsets.

S303: and acquiring flow parameters of the commodity subsets for executing the flow tasks.

In this embodiment, the traffic task is that the subset of commodities reaches the target traffic.

The flow parameters include the current flow of the subset of items.

And acquiring flow parameters of all commodity subsets in all the supporting commodity sets, comparing the flow parameters with the target flow, and judging the commodity subset with the flow parameters smaller than the target flow as the commodity subset with incomplete flow tasks. If the flow task is not completed, the commodity subset which supports the incomplete flow task in the commodity set can be obtained, and the partial commodity subset is considered to be used as the commodity subset which needs to support the flow so as to promote the commodity subset which does not complete the flow task to complete the flow task, and further promote the supported commodity set to complete the standard-reaching task.

For example, a traffic task for a subset of commodities requires a UV (universal identifier) with a target traffic of 100w (ten thousand), and the current traffic in the obtained traffic parameter is 95wUV, so that it may be determined that the subset of commodities does not complete the traffic task. If the flow task of the commodity subset requires a target flow of 100w (ten thousand) of UV (Unique identifier), the current flow in the obtained flow parameters is more than 100wUV, and it can be determined that the commodity subset completes the flow task.

S304: and selecting at least partial commodity subsets which do not complete the flow tasks from the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to the supporting commodity sets.

In this embodiment, in response to obtaining the progress parameter of each supporting subset and the flow parameter of each commodity subset, at least a part of commodity subsets from the commodity subsets not completing the flow task are selected as supporting subsets of the belonging supporting commodity sets, and the supporting subsets are subjected to flow supporting. In other words, the supporting subset is a subset of commodities that need to be supported.

Specifically, a random number of commodity subsets can be selected as the supporting subsets; or, the commodity subsets with the quantity required for completing the standard-reaching task can be selected as supporting subsets; alternatively, a predetermined number of subsets of merchandise may be randomly selected as the support subset. The embodiments for selecting the supporting subset will be described below by way of example, and will not be described in detail herein.

For example, the commodity set a includes 100 commodity subsets, the target number is 90%, the current progress parameter is 80, and the commodity set a can be used as a supporting commodity set. Therefore, the commodity set A can complete the standard task by completing the flow task by 10 commodity subsets. Thus, at least one subset of the articles may be selected from the article set a as a supporting subset of the article set a, e.g., the number of selected subsets of articles may be 6, 10, 11, etc.

S305: performing progress support degree processing on each progress parameter to obtain each first support factor; and carrying out flow supporting degree processing on each flow parameter to obtain each second supporting factor.

In this embodiment, at least one of the progress-support process and the flow-support process includes a control algorithm process. That is, it may be that the progress-support degree processing includes control algorithm processing; alternatively, the flow support processing includes control algorithm processing; alternatively, the progress-support process and the flow-support process may each include a control algorithm process.

One specific process flow of the control algorithm process is illustrated below.

First, the control amount may be obtained based on a progress parameter or a flow parameter. That is, the control amount is a progress control amount based on the progress parameter/a flow control amount based on the flow parameter.

Specifically, when the first supporting factor for supporting the commodity set is acquired, the progress control amount is obtained based on the progress parameter, and the number to be completed can be acquired based on the subset number and the target number, and the number to be completed is taken as the control amount. The subset number is the number of the subset of the support commodity set including the commodity subset. The number to be completed is the ratio of the number of commodity subsets to be completed in the flow task/the number of supporting subsets to the number of all commodity subsets in the supporting commodity set.

When the first supporting factor for the supporting subset is acquired, the ratio of the flow to be completed to the target flow can be acquired, and the ratio is used as the control quantity.

And then, carrying out control algorithm operation on the control quantity and a plurality of control quantities of the same type which are acquired in advance, superposing an operation result and a previous control factor to obtain a current control factor, and taking the current control factor as a first supporting factor or a second supporting factor. In detail, if the control amount is a progress control amount, the control factor is taken as a first supporting factor; and if the control quantity is the flow control quantity, taking the control factor as a second supporting factor.

The plurality of previously acquired control amounts of the same category are the progress control amounts/flow control amounts acquired several times before. If the control quantity is the progress control quantity, the same-category control quantity is the progress control quantity when the flow balance is performed before; if the control amount is a flow control amount, the same type of control amount is a flow control amount at the time of flow balance before that.

Optionally, the progress support processing and the flow support processing both comprise control algorithm processing, and the control algorithm is a PID algorithm.

The first support factor may be obtained as follows:

Δuk＝K _p (err ₁ (k)-err ₁ (k-1))+K _i err ₁ (k)+K _d (err ₁ (k)-2*err ₁ (k-1)+err ₁ (k-2)) formula 1-1

PID1 = PID1 (k-1) +Δuk formula 1-2 wherein err ₁ Representing a control amount based on the set of supported commodities; k (K) _p Is proportional gain, K _i For integral gain, K _d Differential gain is adopted, and the differential gain, the differential gain and the PID algorithm are all the adaptive parameters of the PID algorithm; k represents the current time; k-1 represents the last time, which may be the time of the next previous acquired progress parameter and flow parameter; k-2 represents the last time; err (r) ₁ (k-1) represents a control amount of previous acquisition progress parameter calculation; PID1 (k-1) is the first supporting factor for the previous flow balance; PID1 is the calculated current first support factor.

Wherein err is ₁ The specific results of (2) can be shown as follows:

err ₁ ＝1-[count(completeRate≥1)]total type 1-3

Wherein count (complexrate is greater than or equal to 1) is the number of commodity subsets for completing the flow tasks in the commodity set; total is the number of subsets of items contained by the set of items.

For example, 10 commodity subsets are selected from the commodity set A as supporting subsets, and the control amount is 0.1 (10/100).

The second support factor may be obtained as follows:

Δu _k ＝K _p (err ₂ (k)-err ₂ (k-1))+K _i err ₂ (k)+K _d (err ₂ (k)-2*err ₂ (k-1)+err ₂ (k-2)) formulas 1 to 4

PID2＝PID2(k-1)+Δu _k 1-5

Wherein err is ₂ Representing a control amount based on the set of supported commodities; k (K) _p Is proportional gain, K _i For integral gain, K _d Differential gain is adopted, and the differential gain, the differential gain and the PID algorithm are all the adaptive parameters of the PID algorithm; k represents the current time, k-1 represents the last time, the last time may be the time of the next previous acquired progress parameter and flow parameter, and k-2 represents the last time; err (r) ₂ (k-1) represents the control amount calculated by the previously acquired flow rate parameter; PID2 (k-1) is the second supporting factor for the previous flow balance; PID2 is the calculated current second-order support factor.

Wherein err is ₂ The specific results of (2) can be shown as follows:

err ₂ ＝1-actualUV _hm /targetUV _hm 1-6

Wherein, actualUV _hm Current flow for the subset of commodities; targetUV _hm Is the target flow for the subset of goods.

For example, the traffic task for the support subset requires a target traffic of 100w (ten thousand) UV, the current traffic of 95wUV, and the support subset requires a traffic of 5wUV (100 wUV-95 wUV), then the control amount is 0.05 (5 w/100 w).

If so, the calculated first supporting factor PID1 and the calculated second supporting factor PID2 are fused with each other to form the regulating and controlling force at the same time, so that the supporting force of the commodity subset can be changed relatively stably.

S306: and constraining the second supporting factors of the supporting subsets by using the first supporting factors matched with the supporting subsets to obtain supporting parameters.

In this embodiment, the first supporting factor matched with the supporting subset is the first supporting factor of the supporting commodity set to which the supporting subset belongs.

Specifically, the constraining the second supporting factor by using the first supporting factor may be that the second supporting factor and the first supporting factor matched with the second supporting factor are weighted and fused to obtain a supporting parameter, and the supporting parameter is controlled to be not greater than the constraining parameter. That is, the constraint parameter is used to constrain the upper limit of the support parameter obtained by the weighted fusion.

For example, when the first supporting factor and the second supporting factor are weighted and fused, weights may be assigned to the first supporting factor and the second supporting factor, and the weights may be directly assigned empirical values or may be obtained through model calculation, which is not limited herein.

The weight sizes of both the first supporting factor and the second supporting factor are not particularly limited. That is, the first supporting factor and the second supporting factor are allowed to have the same weight, e.g., the first supporting factor is assigned a weight of 0.5, and the second supporting factor is assigned a weight of 0.5. The weight of the first supporting factor is allowed to be greater than the weight of the second supporting factor, e.g. the weight of the first supporting factor may be 0.7 and the weight of the second supporting factor may be 0.3. The weight of the first support factor is allowed to be less than the weight of the second support factor, e.g., the weight of the first support factor may be 0.25 and the weight of the second support factor may be 0.75.

Further, the constraint parameter may be a preset empirical value, or a value calculated using a model, or the constraint parameter may be obtained based on the first support factor.

For example, the first support factor may be used as the constraint parameter.

In particular, the first support factor may be substituted for the support parameter to form a new support parameter in response to the support parameter being greater than the first support factor. The first supporting factor is used as the upper limit of the supporting parameter, so that when the current supporting commodity set is promoted to complete the standard-reaching task, the risk of resource preemption is reduced, namely the flow preemption for executing the standard-reaching task on other supporting commodity sets is reduced, and therefore all the supporting commodity sets are balanced to complete the standard-reaching task.

Alternatively, the second support factor may be aligned with the first support factor that matches it. The first supporting factor matched with the first supporting factor refers to the first supporting factor of the supporting commodity set to which the supporting subset belongs.

In response to the second support factor being not greater than the first support factor, taking the second support factor as a support parameter; and responding to the second supporting factor not smaller than the first supporting factor, and taking the first supporting factor as supporting parameter.

S307: and obtaining the amplification factor of the supporting parameter by using the balance parameter, and amplifying the supporting parameter.

In this embodiment, the balance factor may be obtained by using the flow parameter and the average completion rate in the preset period. In other words, the balance factor is obtained based on the flow parameter and the average completion rate within the preset period. And weighting the flow parameters by using the balance factors to obtain balance parameters.

And the supporting parameters are adjusted according to the completion condition of the flow tasks of the commodity subsets in the preset period, so that the risk of sudden rising and falling of the flow distributed to the commodity subsets is reduced, the flow fluctuation amplitude is reduced, and accordingly the flow supporting of the commodity subsets is carried out smoothly and relatively stably.

The average completion rate is an average value of flow completion rates of all preset sub-periods in the preset period. For example, the preset period may be three days, one week, half month, one month, etc., which is not limited herein.

The flow completion rate is a ratio of an actual flow rate of the subset of the commodities to a target flow rate within a preset sub-period. The preset period includes a plurality of preset sub-periods. Taking the case that the preset period is one week, the preset sub-period may be one day.

Specifically, the balance factor is obtained and the balance parameter is obtained based on the balance factor as follows:

balanceFactor＝avg _7days (completeRate)/lg(targetUV _dt ) 1-7

balanceScore (x) = complexrate the balanceFactor formula 1-8 wherein balanceFactor is the balance factor; complexeRate is the traffic completion rate, avg _7days (complexrate) is the average completion rate of a subset of the items over the past week (seven days); targetUV _dt Is a flow parameter; balanceScore (x) is the equilibrium parameter.

In equations 1-5, the flow parameters may be reduced in dimension before weighting them with the balance factors to facilitate obtaining reasonable balance factors, improving the influence of the balance factors, i.e., lg (targetUV) _dt )。

Furthermore, the balance parameters can be updated by standard deviation normalization processing, which is further beneficial to supporting each commodity subset relatively stably. The standard deviation normalization processing for the balance parameters can be specifically shown as the following formula:

balance score = (balanceScore (x) - μ)/delta formula 1-9 wherein balance score is subjected to standard deviation normalization processing to update the balance parameters; mu is the average value of balanceScore (x) of all commodity subsets in the commodity set to which the commodity subsets belong; delta is the standard deviation of balanceScore (x) for each commodity subset in the commodity set to which the commodity subset belongs.

Optionally, the flow completion rate can be interfered according to the preset completion rate, and when the interference condition is met, the preset completion rate is used as a new flow completion rate, so that the flow distribution is interfered by reducing the limit of the flow completion rate in a preset sub-period, and the flow of the commodity subset is enabled to be dynamically changed relatively stably by an interference means. The preset completion rate can be set as an empirical value, or can be calculated according to a model.

Considering that the flow of goods during the campaign is likely to increase substantially, even far beyond the target flow, due to the preferential promotion campaigns of the merchant/e-commerce platform. Thus, the intervention condition may be that the flow completion rate is greater than the preset completion rate, thereby reducing the flow completion rate in the preset sub-period to interfere with flow distribution, which is beneficial to enabling the flow of the subset of goods to dynamically change relatively smoothly. Alternatively, the preset completion rate may be 110%, 120%, 125%, 130%, 140%.

And/or the intervention condition may be that the flow completion rate is less than a preset completion rate, which may be 15%, 20%, 25%, etc.

For example, the target flow rate of the supporting subset is 100wUV, a preset sub-period in the preset period is accumulated to obtain 200wUV, that is, the flow completion rate is 200%, and the preset completion rate is 120%, and the flow completion rate of the preset sub-period of the supporting subset is updated to 120% to participate in the calculation of the balance factor.

Still further, in consideration of the actual flow balancing process, the subset of commodities that complete the flow task and the subset of commodities that are not support subsets need to be regulated together. In each supporting subset, the balance parameters are obtained based on the flow parameters, and the regulation and control force is pulled up. It is considered that the balance parameter may be zero without additional weighting support for each subset of items not being a support subset.

Specifically, after the obtained supporting subset, only the supporting subset is calculated as above to obtain the balance parameter, and the balance parameter of the commodity subset which is not used as the supporting subset directly takes a zero value. Or, pre-calculating balance parameters of each commodity subset/the commodity subset of the incomplete flow task, and setting the balance parameters of the rest commodity subsets to zero values after determining the supporting subset.

The amplification factor of the supporting parameter is obtained by utilizing the balance parameter, and the amplified supporting parameter can be specifically shown by the following formula:

PID _n ＝PID _n1 * (1+balance score) formula 1-10

Wherein PID _n PID for amplified support parameters _n1 In order to constrain the support parameters obtained after the second support factor by the first support factor, the balance score is an equilibrium parameter.

S308: and selecting intermediate values of the amplified supporting parameters, the supporting upper limit parameters and the supporting lower limit parameters as new supporting parameters.

In the present embodiment, both the holding upper limit parameter and the holding lower limit parameter are acquired in advance.

And selecting an intermediate value from the amplified supporting parameters, the supporting upper limit parameters and the supporting lower limit parameters, which is beneficial to reducing the risk of excessively low/high flow distributed by the supporting subset, and restraining the supporting force by utilizing the supporting upper limit parameters and the supporting lower limit parameters. For example, the specific implementation may be as follows:

PID＝Math.min(PIDmax，Math.max(PIDmin，PID _n ) 1-11

The PID is a new supporting parameter and is used for finally carrying out weighted supporting; math.min () is to select the smaller value among them; math.max () is the value selected to be larger; PIDmax is a supporting upper limit parameter; PIdmin is the support lower limit parameter.

S309: weighting the support subset with support parameters.

In this embodiment, the fine arrangement and the regulatory factor of each commodity in each commodity subset may be obtained.

The fine ranking score may be obtained by using a fine ranking model, and is used for identifying a score of the commodity attribute. For example, the fine ranking score may be calculated based on parameters such as commodity flow, search times, prices, yield, and good score. Alternatively, the higher the fine ranking of the commodity may be considered to be better the commodity performance.

Taking the product of the fine discharge and the regulation and control parameters as a final score to obtain the flow rate which is specifically distributed to each commodity. The specific formula is as follows:

LogFactor＝log ₁₀ (ReqModelScaore _MAX /ReqModelScaore _MIN ) 1-12

FinalScore＝ModelScore*10 ^{PID*logFactor} 1-13

Wherein, logFactor is the regulatory factor; reqModelScaore _MAX Obtaining the fine discharge maximum value of all commodities for the current time; reqModelScaore _MIN Obtaining all quotients for the current timeMinimum value of fine discharge fraction of the product; finalScar is the final score; 10 ^{PID*logFactor} Is a regulating parameter; modelScare is the fine-ranking of the current commodity, and PID is the supporting parameter of the commodity subset to which the current commodity belongs.

Therefore, the regulating factor is obtained by reducing the dimension of the ratio of the acquired fine discharge maximum value to the fine discharge minimum value.

The regulation and control parameters are obtained by taking a preset value as a base number and taking the product of the supporting parameters and the regulation and control factors as a power to perform exponential operation.

The fine arrangement of different commodities may be different, so that the weight holding force allocated by each commodity may be different.

Specifically, the commodities can be arranged according to a final descending order, original commodity ordering is disturbed, flow among the commodities is balanced, and further flow tasks of the commodity subsets are promoted to be completed, and standard tasks of the commodity sets are promoted to be completed.

Therefore, in the embodiment, the flow control force/supporting force is controlled and regulated in a self-balancing way by using the double-layer supporting factors, so that the tedious work of manually adjusting parameters is reduced, the flow among various commodities can be balanced, and the macroscopic control on the commodity subsets and the commodity sets is realized. In popular terms, the double PD control is utilized, and the service plate PID is taken as the thought of the upper limit of the cargo pool PID, so that macroscopic regulation and control can be carried out on different cargo pools and different service plates belonging to various commodities, and each service plate is promoted to complete the standard-reaching task, so that the dual targets of the flow target and the standard-reaching rate target are achieved, the standard-reaching rate of the service plate is accurate and stable, and the flow preemption between the service blocks/among the cargo pools is reduced. Meanwhile, the application considers that the flow competition relationship exists between the similar units, so that macroscopic flow regulation and control are beneficial to promoting each unit to complete tasks.

Meanwhile, the target flow of the flow tasks and the target number of the up-to-standard tasks can be thinned in a time-sharing mode, and therefore the flow balance refinement is facilitated. For example, for a subset of products, one can compare 10: 00. 12:00, 14:00am, 18: 00. and setting target flow at specific target time points such as 23:00, and selecting the target flow at the current target time point or the next adjacent target time point as the target flow of the current flow balance when the flow is supported.

For example, when the target flow rate of the commodity subset is 100wuv at 10:00 and the flow rate parameter is acquired at 9:50, the target flow rate of 100wUV at 10:00 is taken as the target flow rate of the current flow balance.

Alternatively, the target flow rate for the subset of products at 10:00 is 100wUV, at 12: the target flow rate of 00 is 130wUV. When the commodity subset obtains the flow parameters at 10:00, the target flow 100wUV of 10:00 can be used as the target flow of the current flow balance, and the flow parameters can also be 12: the target flow rate 130 and wUV of 00 is not limited to this flow rate balance target flow rate.

In one embodiment, another method for selecting a support subset is provided, as shown in fig. 4, and fig. 4 is a flowchart illustrating an embodiment of a method for selecting a support subset according to the present application. The following specifically describes a method for selecting a supporting subset in the present application:

S3041: and obtaining the difference value between the target quantity and the progress parameter as the required quantity.

In this embodiment, the up-to-standard task of the commodity set is that the flow task is completed for the commodity subset requiring the target number. Therefore, the difference between the target quantity required in the up-to-standard task and the currently acquired progress parameter can be acquired as the required quantity, so that the quantity of the commodity subsets required to be weighted and supported can be reversely deduced, and the method specifically can be shown as the following formula:

completeRate＝actualUV _hm /targetUV _hm 2-1

reach rate= [ count (complexrate. Gtoreq.1) ]/total 2-2

brandnum=total-reach rate 2-3

The complexeRate is the completion degree of the flow task of the commodity subset, the complexeRate is more than or equal to 1, and the commodity subset is considered to complete the flow task, otherwise, the commodity subset is considered to complete the flow task; actual UV _hm Current flow for the subset of commodities; targetUV _hm Target flow for a subset of goods; the reach rate is the current standard rate of the commodity setThe method comprises the steps of carrying out a first treatment on the surface of the count (complexRate. Gtoreq.1) is the number of commodity subsets for completing the flow tasks in the commodity set; total is the number of commodity sets comprising all commodity subsets; brandNum is the required number; targetRate is the target number required for a qualifying task.

Wherein, the formula 2-1 may be used to determine the subset of commodities that do not complete the flow task in the foregoing embodiment, and after obtaining the complexeRate, further determine the relationship between the complexeRate and 1, where the complexeRate is greater than or equal to 1, and determine the subset of commodities that do not complete the flow task, otherwise determine the subset of commodities that do not complete the flow task.

The formula 2-2 may be applied to the step S302 of the foregoing embodiment, where the current standard-reaching rate of the commodity set is obtained, and the standard-reaching task is determined whether to complete the standard-reaching task by comparing the current standard-reaching rate with the target number of standard-reaching tasks.

That is, the flow task is completed by the subset of the required quantity of commodities, and the standard-reaching task can be completed by supporting the commodity set.

Therefore, the commodity subset requiring a plurality of incomplete flow tasks can be directly selected to participate in the current flow weighting support, so that the efficiency of promoting the commodity set to complete the up-to-standard tasks is improved.

For example, if the target flow rate of the commodity subset is 100wUV and the current flow rate of the commodity subset is 95wUV, it is determined that the current commodity subset does not complete the flow task. If the target flow of the commodity subset is 90wUV and the current flow of the commodity subset is 100wUV, the current commodity subset is judged to complete the flow task.

The commodity set comprises 100 commodity subsets, and the target number of up-to-standard tasks is 80%. Counting 70 commodity subsets of the commodity set for completing the flow tasks means that the commodity set can complete the up-to-standard tasks by completing the flow tasks by 10 commodity subsets, namely, the required number is 10.

S3042: and randomly distributing selection parameters for the subset of the commodities which support the incomplete flow tasks in the commodity set.

In this embodiment, selection parameters are allocated to the subset of the commodities supporting all the incomplete flow tasks in the commodity set, where the selection parameters are randomly generated and allocated to facilitate the randomness of selecting the subset of the commodities, and reduce the risk that the difference between the subsets of the commodities is gradually increased due to the constant selection of the subset of the commodities with good performance, thereby facilitating the balance of the performance of each pool. The selection parameter is matched with a preset parameter.

Alternatively, both the selection parameter and the preset parameter may be any value in the range of 0 to 1. For example, the selection parameters may be 0, 0.1, 0.13, 0.145, 0.5, 0.73, 0.8899, 1, etc., and the preset parameters may be 0.6.

S3043: and respectively comparing the selection parameters of the commodity subsets with preset parameters, and judging whether preset conditions are met.

In this embodiment, if the preset condition is satisfied, step S3044 is executed, and the next selection parameter is continuously compared with the preset parameter; if the preset condition is not met, the next selection parameter is continuously compared with the preset parameter.

The preset parameters may be preset empirical values, or may be calculated according to a model. And when the preset parameter is an empirical value, the adjustment can be performed according to multiple screening results.

The preset condition may be that the required selection parameter is not less than the preset parameter; or, the selection parameter is required to be not more than a preset parameter; or, the result of operation selection of the two is in accordance with the expectation; or, the selection parameter is required to be smaller than the preset parameter; or, the selection parameter is required to be larger than the preset parameter.

S3044: the subset of goods is taken as the supporting subset.

In this embodiment, the subset of items is taken as the support subset in response to the selection parameter satisfying the preset condition.

For example, the preset condition is that the required selection parameter is smaller than the preset parameter, and the preset parameter is 0.5. If the randomly assigned selection parameter for the subset of items is 0.49, it is selected as the supporting subset.

S3045: the number of support subsets is compared with the required number.

In this embodiment, if the number of supporting subsets is smaller than the required number, it is considered that a subset of the commodities that have not completed the flow task may be selected continuously as the supporting subset, and the flow ends or step S3046 is performed (the implementation of step S3046 is illustrated in fig. 4); if the number of support subsets is greater than the required number, the process ends or step S3047 is performed (the implementation of step S3047 is illustrated in fig. 4); if the number of the supporting subsets is equal to the required number, the process ends.

For example, if the required number of the commodity sets obtained in step S3041 is 10, and the number of the supporting subsets is 8 through steps S3042-S3044, then it is necessary to continue to select the commodity subsets with 2 incomplete flow tasks as the supporting subsets.

If the required number of the commodity set obtained in step S3041 is 10, and the number of the supporting subsets is 10 through steps S3042-S3044, the process is ended, and the 10 supporting subsets are weighted and supported.

If the required number of the commodity set obtained in step S3041 is 10, and the number of the supporting subsets is 15 through steps S3042-S3044, then 10 supporting subsets need to be selected as supporting subsets, or 5 supporting subsets need to be selected to cancel the identities of the supporting subsets without weighting supporting.

S3046: and selecting the commodity subset which is not used as the supporting subset by combining the balance parameters of the commodity subset which is not used as the supporting subset until the number of the supporting subsets is equal to the required number.

In this embodiment, in response to the number of support subsets being less than the required number, the subset of items not being support subsets is selected as support subsets in combination with the balance parameters of the subset of items not being support subsets until the number of support subsets is equal to the required number. Wherein the balance parameter is obtained based on the flow parameter.

Alternatively, the subsets of goods may be arranged in descending order of balance parameters; and sequentially selecting the commodity subsets which are not used as the supporting subsets from the head end of the queue as the supporting subsets.

The balance parameters may be calculated as exemplified in the previous embodiments. Specifically, the balance factor can be obtained by using the flow parameter and the average completion rate in the preset period, and the balance factor is used for weighting the flow parameter to obtain the balance parameter.

In an alternative implementation, the selection parameters may be further allocated to the subset of commodities that are not the supporting subset, and the selection parameters may be compared with the preset parameters to select the supporting subset.

Optionally, the subset of commodities that is not a support subset refers to a subset of commodities that does not complete the traffic task in the set of commodities and that is not a support subset.

S3047: the supporting subsets are filtered such that the number of supporting subsets is equal to the required number.

In this embodiment, in response to the number of support subsets being greater than the required number, the support subsets are filtered such that the number of support subsets is equal to the required number.

Alternatively, screening may be performed according to balance parameters; alternatively, it may be randomly selected; alternatively, the selection parameters may be reassigned and screened using preset parameters. That is, the number of supporting subsets may be the same as the required number, and the specific screening method is not limited herein.

In the present application, the CK (click house) technique can be used in combination with kafka for in-line polymerization. Compared with the traditional mode of realizing flow balance by utilizing a hive table, the method and the device can remarkably improve the calculation efficiency, shorten the supporting period of supporting the supporting parameters each time, and further facilitate the adjustment of the supporting parameters according to commodity states in time and promote the completion of flow tasks and standard-reaching tasks.

For example, when the hive table is used to store data, it may only be possible to obtain the flow data of the commodity every 10 minutes, and process the flow data to obtain the supporting parameters for supporting. The application can acquire the flow data of the commodity every 5 minutes, process the flow data to obtain the supporting parameters, and utilize the supporting parameters to support.

It should be understood that, although the steps in the flowcharts of fig. 2-4 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2-4 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily occur sequentially, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or steps.

In one embodiment, a flow balancing device is provided, as shown in fig. 5, and fig. 5 is a schematic structural diagram of an embodiment of the flow balancing device according to the present application.

The flow balancing means comprises a data layer 51 and a control module 52.

And the data layer 51 is used for counting the flow parameters of each commodity subset.

The control module 52 is configured to obtain progress parameters of the plurality of supporting commodity sets for executing the up-to-standard task, and obtain flow parameters of the plurality of commodity subsets for executing the flow task; the standard reaching task is that the flow task is completed for the commodity subset supporting the target quantity in the commodity set; selecting at least part of the commodity subsets which do not complete the flow task from the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to the supporting commodity sets; performing progress support degree processing on each progress parameter to obtain each first support factor; carrying out flow supporting degree processing on each flow parameter to obtain each second supporting factor; constraining the second supporting factors of each supporting subset by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; and weighting the supporting subset by using the supporting parameters.

For specific limitations of the flow balancing device, reference may be made to the limitations of the flow balancing method hereinabove, and no further description is given here. The various modules in the flow balancing device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

Further, the flow balancing system shown in fig. 5 is further refined, as shown in fig. 6, and fig. 6 is a schematic structural diagram of another embodiment of the flow balancing device of the present application.

In this embodiment, the control module includes a commodity collection layer and a commodity subset layer.

The commodity set layer is used for processing data related to the commodity set, generating a first supporting factor and feeding back the first supporting factor to the commodity subset layer.

The commodity subset layer is used for processing data related to the commodity subset, generating a second supporting factor, restraining the second supporting factor by utilizing the received first supporting factor to obtain supporting parameters, and feeding back the supporting parameters to the front end for adjusting page ordering of each commodity, so that supporting actions are realized.

In one embodiment, a computer device is provided, which may be a server, and an internal structure thereof may be shown in fig. 7, and fig. 7 is a schematic structural diagram of an embodiment of the computer device of the present application.

The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing progress parameters, flow parameters, balancing parameters, first support factors, second support factors, support parameters, etc., i.e. the data referred to in the foregoing embodiments regarding flow balancing methods. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a flow balancing method.

It will be appreciated by those skilled in the art that the structure shown in FIG. 7 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program:

S205: and weighting the supporting subset by using the supporting parameters.

In one embodiment, the processor when executing the computer program further performs the steps of:

In this embodiment, the up-to-standard task is to support a subset of the target number of commodities in the commodity set to complete the flow task.

In this embodiment, whether each commodity set completes the standard-reaching task may be determined, and a part of commodity sets that do not complete the standard-reaching task may be selected as the supporting commodity sets, or all commodity sets that do not complete the standard-reaching task may be used as the supporting commodity sets, which is not limited herein. If so, the progress parameters of a plurality of supporting commodity sets for executing the standard-reaching tasks can be obtained.

The flow parameters include that the current flow of the subset of items is based on the current flow and the target flow.

In this embodiment, the supporting subset is a subset of commodities that need to be supported.

The embodiments for selecting the supporting subset will be described below by way of example, and will not be described in detail herein.

In this embodiment, at least one of the progress-support process and the flow-support process includes a control algorithm process.

In this embodiment, the first supporting factor matched with the supporting subset is the first supporting factor of the supporting commodity set to which the supporting subset belongs. Specifically, the constraining the second supporting factor by using the first supporting factor may be that the second supporting factor and the first supporting factor matched with the second supporting factor are weighted and fused to obtain a supporting parameter, and the supporting parameter is controlled to be not greater than the constraining parameter.

Further, the first support factor may be substituted for the support parameter to form a new support parameter in response to the support parameter being greater than the first support factor. Alternatively, the second support factor may be aligned with the first support factor that matches it. In response to the second support factor being not greater than the first support factor, taking the second support factor as a support parameter; and responding to the second supporting factor not smaller than the first supporting factor, and taking the first supporting factor as supporting parameter.

In this embodiment, the balance factor is used to weight the flow parameters to obtain the balance parameters. The balance parameters can be further subjected to standard deviation normalization processing to update the balance parameters. Further, the flow parameters may be reduced in dimension before weighting the flow parameters with the balance factor. The balance factor is obtained based on the flow parameter and the average completion rate in a preset period. The average completion rate is an average value of flow completion rates of each preset sub-period in the preset period. The flow completion rate is a ratio of an actual flow rate of the subset of the commodities to a target flow rate within a preset sub-period. Optionally, in response to the traffic completion rate being greater than the preset completion rate, the preset completion rate is taken as the new traffic completion rate.

Still further, in each support subset, the balance parameter is derived based on the flow parameter; the balance parameter is zeroed for each subset of items that are not a support subset.

S309: weighting the support subset with support parameters.

In this embodiment, the fine arrangement and the regulatory factor of each commodity in each commodity subset may be obtained. Taking the product of the fine discharge and the regulation parameters as a final score, and arranging the commodities in a final score descending manner. The regulating factor is obtained by reducing the dimension of the ratio of the acquired fine discharge maximum value to the fine discharge minimum value. The regulation and control parameters are obtained by taking a preset value as a base number and taking the product of the supporting parameters and the regulation and control factors as a power to perform exponential operation.

In one embodiment, the processor when executing the computer program further performs the following steps to effect the selection of the support subset:

S3044: the subset of goods is taken as the supporting subset.

S3045: the number of support subsets is compared with the required number.

In the present embodiment, if the number of supporting subsets is smaller than the required number, step S3046 is performed; if the number of the supporting subsets is greater than the required number, ending the flow or executing step S3047; if the number of the supporting subsets is equal to the required number, the process ends.

Responding to the fact that the number of the supporting subsets is smaller than the required number, and selecting the commodity subsets which are not used as the supporting subsets by combining balance parameters of the commodity subsets which are not used as the supporting subsets until the number of the supporting subsets is equal to the required number; wherein the balance parameter is obtained based on the flow parameter.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

S205: and weighting the supporting subset by using the supporting parameters.

In one embodiment, the computer program when executed by the processor further performs the steps of:

Specifically, the constraining the second supporting factor by using the first supporting factor may be that the second supporting factor and the first supporting factor matched with the second supporting factor are weighted and fused to obtain a supporting parameter, and the supporting parameter is controlled to be not greater than the constraining parameter.

Further, the first support factor may be substituted for the support parameter to form a new support parameter in response to the support parameter being greater than the first support factor.

Alternatively, the second support factor may be aligned with the first support factor that matches it. In response to the second support factor being not greater than the first support factor, taking the second support factor as a support parameter; and responding to the second supporting factor not smaller than the first supporting factor, and taking the first supporting factor as supporting parameter.

In this embodiment, the balance factor is used to weight the flow parameters to obtain the balance parameters. The balance parameters can be further subjected to standard deviation normalization processing to update the balance parameters.

Further, the flow parameters may be reduced in dimension before weighting the flow parameters with the balance factor.

The balance factor is obtained based on the flow parameter and the average completion rate in a preset period.

The average completion rate is an average value of flow completion rates of each preset sub-period in the preset period.

The flow completion rate is a ratio of an actual flow rate of the subset of the commodities to a target flow rate within a preset sub-period. Optionally, in response to the traffic completion rate being greater than the preset completion rate, the preset completion rate is taken as the new traffic completion rate.

S309: weighting the support subset with support parameters.

In this embodiment, the fine arrangement and the regulatory factor of each commodity in each commodity subset may be obtained. Taking the product of the fine discharge and the regulation parameters as a final score, and arranging the commodities in a final score descending manner.

The regulating factor is obtained by reducing the dimension of the ratio of the acquired fine discharge maximum value to the fine discharge minimum value.

S3044: the subset of goods is taken as the supporting subset.

S3045: the number of support subsets is compared with the required number.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims

1. A flow balancing method, the flow balancing method comprising:

acquiring progress parameters of a plurality of supporting commodity sets for executing standard tasks, and acquiring flow parameters of a plurality of commodity subsets for executing flow tasks; the standard reaching task is that the flow task is completed for the commodity subset supporting the target quantity in the commodity set;

Selecting at least part of the commodity subsets which do not complete the flow task from the supporting commodity sets, and respectively taking the commodity subsets as supporting subsets belonging to the supporting commodity sets;

performing progress support degree processing on each progress parameter to obtain each first support factor; carrying out flow supporting degree processing on each flow parameter to obtain each second supporting factor;

constraining the second supporting factors of each supporting subset by using the first supporting factors matched with the supporting subsets to obtain supporting parameters; the first supporting factors matched with the supporting subsets are the first supporting factors of the supporting commodity sets to which the supporting subsets belong;

and weighting the supporting subset by using the supporting parameters.

2. The flow balancing method according to claim 1, wherein said constraining the second support factor thereof with the first support factor matched thereto comprises:

and carrying out weighted fusion on the second supporting factors and the first supporting factors matched with the second supporting factors to obtain supporting parameters, and controlling the supporting parameters not to be larger than constraint parameters.

3. The flow balancing method according to claim 2, characterized in that the first support factor is taken as the constraint parameter;

The controlling the support parameter not to be greater than the constraint parameter includes:

and in response to the support parameter being greater than the first support factor, replacing the support parameter with the first support factor to form a new support parameter.

4. The flow balancing method according to claim 1, wherein constraining the second support factor with the first support factor matched thereto to obtain a support parameter comprises:

comparing the second supporting factor with the first supporting factor matched with the second supporting factor;

responsive to the second support factor being not greater than the first support factor, taking the second support factor as the support parameter; and in response to the second support factor being not less than the first support factor, taking the first support factor as the support parameter.

5. The flow balancing method of claim 1, wherein at least one of the progress-support process and the flow-support process comprises a control algorithm process; the control algorithm process includes:

obtaining a control quantity; the control quantity is a progress control quantity obtained based on the progress parameter/a flow control quantity obtained based on the flow parameter;

The control quantity and a plurality of control quantities of the same type which are acquired in advance are subjected to control algorithm operation;

and superposing the operation result and the previous control factor to obtain a current control factor, and taking the current control factor as the first supporting factor or the second supporting factor.

6. The flow balancing method according to claim 5, wherein,

the flow task is that the commodity subset reaches a target flow, and the flow parameters comprise the current flow of the commodity subset; the control amount is a flow control amount obtained based on a flow parameter, and includes: based on the current flow and the target flow, obtaining the ratio of the flow to be completed to the target flow, and taking the ratio as the flow control quantity;

the progress parameters comprise the subset number of the commodity subsets for supporting the commodity set to complete the flow task; the control amount is a progress control amount obtained based on the progress parameter, and the control amount comprises: and acquiring the quantity to be completed based on the subset quantity and the target quantity, and taking the quantity to be completed as the control quantity.

7. The flow balancing method of claim 1, wherein prior to weighting the subset of struts with the struts parameters comprises:

Obtaining the amplification factor of the supporting parameter by utilizing the balance parameter, and amplifying the supporting parameter; wherein, in each supporting subset, the balance parameter is obtained by the flow parameter; taking zero from all commodity subsets which are not taken as supporting subsets;

selecting intermediate values of the amplified supporting parameters, the supporting upper limit parameters and the supporting lower limit parameters as new supporting parameters; wherein both the support upper limit parameter and the support lower limit parameter are acquired in advance.

8. The flow balancing method of any one of claims 1-7, wherein the progress and flow support processes each comprise a control algorithm process, the control algorithm being a PID algorithm.

9. The flow balancing method of claim 1, wherein selecting at least a portion of the subset of articles from the set of supported articles that did not complete the flow task as a supported subset comprises:

randomly distributing selection parameters for the commodity subset which does not complete the flow task in the supporting commodity set;

respectively comparing the selection parameters of the commodity subsets with preset parameters;

And responding to the selection parameters to meet preset conditions, and taking the commodity subset as the supporting subset.

10. The flow balancing method of claim 9, wherein the progress parameter comprises a subset number of the subset of commodities that the set of supported commodities complete a flow task;

the selecting at least a portion of the subset of articles from the set of supported articles that did not complete the traffic task as a supported subset further comprises:

obtaining the difference value between the target quantity and the progress parameter as a demand quantity;

comparing the number of the supporting subsets with the required number;

in response to the number of the supporting subsets being smaller than the required number, selecting the subset of commodities which are not used as the supporting subset by combining balance parameters of the subset of commodities which are not used as the supporting subset until the number of the supporting subset is equal to the required number; wherein the balance parameter is derived based on the flow parameter.

11. The flow balancing method of claim 10, wherein the selecting the subset of items not being a support subset as the support subset in combination with the balance parameters of the subset of items not being a support subset comprises:

Arranging the commodity subsets in descending order according to the balance parameters;

and sequentially selecting the commodity subset which is not used as the supporting subset from the head end of the queue as the supporting subset.

12. The flow balancing method of claim 10, wherein the comparing the number of the subset of support with the required number further comprises:

in response to the number of support subsets being greater than the required number, the support subsets are screened to make the number of support subsets equal to the required number.

13. The flow balancing method according to any one of claims 7, 10-12, wherein the balancing parameters are derived based on the flow parameters comprising:

weighting the flow parameters by using balance factors to obtain the balance parameters; the balance factor is obtained based on the flow parameter and an average completion rate in a preset period.

14. The flow balancing method according to claim 13, wherein the average completion rate is an average value of flow completion rates of respective preset sub-periods within a preset period; the flow completion rate is the ratio of the actual flow of the commodity subset to the target flow in a preset sub-period.

15. The flow balancing method according to claim 14, wherein the flow completion rate is a ratio of an actual flow rate of the subset of commodities to a target flow rate within a preset sub-period, further comprising:

and responding to the flow completion rate being greater than a preset completion rate, and taking the preset completion rate as a new flow completion rate.

16. The flow balancing method of claim 13, wherein prior to weighting the flow parameters with a balancing factor comprises: and reducing the dimension of the flow parameter.

17. The flow balancing method according to claim 13, wherein the obtaining the balancing parameters further comprises: and carrying out standard deviation normalization processing on the balance parameters to update the balance parameters.

18. The flow balancing method according to claim 1, wherein the obtaining a progress parameter of the plurality of support commodity sets for performing the up-to-standard task includes:

acquiring progress parameters of each commodity set for executing the standard task;

and selecting at least part of the commodity set which does not complete the standard reaching task as the supporting commodity set.

19. The flow balancing method of claim 1, wherein said weighting the subset of struts with the strut parameters comprises:

Acquiring fine arrangement and regulation factors of all commodities in all commodity subsets;

taking the product of the fine discharge and the regulation parameters as a final score, and arranging the commodities in a descending manner according to the final score;

the regulation factor is obtained by reducing the dimension of the ratio of the acquired fine discharge maximum value to the fine discharge minimum value; the regulation and control parameters are obtained by taking a preset value as a base number and taking the product of the supporting parameters and the regulation and control factors as power to perform exponential operation.

20. A flow balancing device, the flow balancing device comprising:

the data layer is used for counting flow parameters of all commodity subsets;

a control module for implementing the steps of the flow balancing method of any one of claims 1 to 19.

21. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the flow balancing method according to any one of claims 1 to 19 when the computer program is executed.

22. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the flow balancing method according to any one of claims 1 to 19.