Article flow volume information generation method and device, equipment, medium and program product

Abstract

The embodiment of the invention discloses an article flow volume information generation method and device, equipment, a medium and a program product. A specific embodiment of the method comprises the following steps: extracting last-layer time series data from hierarchical time series data of an article flow volume of a corresponding target article; determining the time sequence data of each layer different from the last layer as a previous layer time sequence data set; obtaining an article flow prediction result set of the front-layer time sequence data set corresponding to a preset future time period as article flow prediction constraint information; generating an article flow prediction result according to the last-layer time sequence data and an article flow generation model; and according to the article flow prediction constraint information, carrying out correction processing on the article flow prediction result. The implementation mode is related to an intelligent supply chain, and the time sequence data of the article flow volumes of different levels can meet consistency, so that the method can be suitable for collaborative decision-making processes of the levels, and the accuracy of the predicted article flow volumes is improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to methods, devices, equipment, media, and program products for generating item turnover information. Background Art

[0002] Hierarchical time series prediction is a type of technology in the field of time series prediction. Hierarchical time series refers to multiple time series that can be associated through a certain hierarchical relationship. Time series at different levels can be decomposed or aggregated according to certain attributes (such as region, item type, etc.) to achieve mutual transformation. At present, when conducting hierarchical time series prediction of item turnover, the method usually adopted is: regardless of the hierarchical relationship, a single time series prediction is performed on the time series at each level.

[0003] However, the inventors have found that when the above method is used for hierarchical time series forecasting, the following technical problems often occur: the forecast results at different levels cannot automatically meet the consistency requirements and cannot be applied to the collaborative decision-making processes at each level, resulting in poor accuracy of the predicted hierarchical time series of item turnover.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0005] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] Some embodiments of the present disclosure propose methods, devices, electronic devices, computer-readable media and computer program products for generating item turnover information to solve one or more of the technical problems mentioned in the above background technology section.

[0007] In a first aspect, some embodiments of the present disclosure provide a method for generating item turnover volume information, the method comprising: extracting the time series data of the corresponding last layer from the stratified time series data of the item turnover volume corresponding to the target item as the last layer time series data; determining the time series data of each layer different from the last layer in the above stratified time series data as a previous layer time series data set; obtaining the item turnover volume prediction result set corresponding to a preset future time period from the above previous layer time series data set as item turnover volume prediction constraint information, wherein each item turnover volume prediction result in the above item turnover volume prediction result set satisfies a consistency condition; generating an item turnover volume prediction result corresponding to the above preset future time period according to the above last layer time series data and an item turnover volume generation model; and correcting the above item turnover volume prediction result according to the above item turnover volume prediction constraint information to obtain the corrected item turnover volume prediction result as the output result of the above item turnover volume generation model.

[0008] Optionally, the above-mentioned generation of the item turnover volume prediction result corresponding to the above-mentioned preset future time period based on the above-mentioned last-layer time series data and the item turnover volume generation model includes: performing missing value filling processing on the last-layer time series data to update the last-layer time series data; generating the item turnover volume prediction result corresponding to the above-mentioned preset future time period based on the updated last-layer time series data and the above-mentioned item turnover volume generation model.

[0009] Optionally, before performing missing value filling processing on the last layer of time series data to update the last layer of time series data, the method further includes: in response to determining that there are abnormal values ​​in the last layer of time series data, removing the abnormal values ​​in the last layer of time series data.

[0010] Optionally, the above-mentioned item turnover volume prediction result is corrected according to the above-mentioned item turnover volume prediction constraint information, including: generating residual information according to the above-mentioned item turnover volume generation model corresponding to the historical time period and each historical fitting value and each actual value of the above-mentioned last layer; generating a covariance matrix according to the above-mentioned residual information; generating a corrected item turnover volume prediction result according to the above-mentioned covariance matrix, the above-mentioned item turnover volume prediction constraint information and the above-mentioned item turnover volume prediction result.

[0011] Optionally, the above-mentioned generating a corrected item turnover volume prediction result based on the above-mentioned covariance matrix, the above-mentioned item turnover volume prediction constraint information and the above-mentioned item turnover volume prediction result includes: determining the constraint matrix corresponding to the above-mentioned layered time series data; generating item turnover volume constraint information based on the above-mentioned constraint matrix and the above-mentioned item turnover volume prediction constraint information; constructing an item turnover volume prediction optimization target based on the above-mentioned item turnover volume prediction result and the above-mentioned covariance matrix; generating a corrected item turnover volume prediction result based on the above-mentioned item turnover volume constraint information and the above-mentioned item turnover volume prediction optimization target.

[0012] Optionally, the method also includes: selecting each item circulation volume sequence corresponding to the pre-equipment cargo time period from the corrected item circulation volume prediction results; for each pre-equipment cargo area in the pre-equipment cargo area set, performing the following steps: selecting each item circulation volume sequence corresponding to the above-mentioned pre-equipment cargo area from the above-mentioned each item circulation volume sequence as the item circulation volume sequence of each area; determining the sum of each item circulation volume included in the above-mentioned each area item circulation volume sequence as the predicted total item circulation volume corresponding to the above-mentioned pre-equipment cargo area; generating a replenishment quantity according to the above-mentioned predicted total item circulation volume and the remaining inventory of the above-mentioned target item; and controlling the item transportation equipment corresponding to the above-mentioned pre-equipment cargo area to transport the target item of the above-mentioned replenishment quantity to the item storage location corresponding to the above-mentioned pre-equipment cargo area.

[0013] In a second aspect, some embodiments of the present disclosure provide an item turnover volume information generating device, the device comprising: an extraction unit, configured to extract the time series data of the corresponding last layer from the stratified time series data of the item turnover volume corresponding to the target item as the last layer time series data; a determination unit, configured to determine the time series data of each layer different from the last layer in the stratified time series data as a previous layer time series data set; an acquisition unit, configured to acquire the item turnover volume prediction result set corresponding to the previous layer time series data set for a preset future time period as item turnover volume prediction constraint information, wherein each item turnover volume prediction result in the above item turnover volume prediction result set satisfies a consistency condition; a generation unit, configured to generate an item turnover volume prediction result corresponding to the above preset future time period based on the above last layer time series data and an item turnover volume generation model; a correction unit, configured to correct the above item turnover volume prediction result based on the above item turnover volume prediction constraint information, and obtain the corrected item turnover volume prediction result as the output result of the above item turnover volume generation model.

[0014] Optionally, the generation unit is further configured to: perform missing value filling processing on the last layer time series data to update the last layer time series data; and generate an item turnover prediction result corresponding to the above-mentioned preset future time period based on the updated last layer time series data and the above-mentioned item turnover generation model.

[0015] Optionally, the generating unit includes: a removing unit configured to remove the abnormal values ​​in the last layer of time series data in response to determining that there are abnormal values ​​in the last layer of time series data.

[0016] Optionally, the correction unit is further configured to: generate residual information based on the historical time period corresponding to the above-mentioned item turnover generation model and the historical fitting values ​​and actual values ​​of the above-mentioned last layer; generate a covariance matrix based on the above-mentioned residual information; generate a corrected item turnover prediction result based on the above-mentioned covariance matrix, the above-mentioned item turnover prediction constraint information and the above-mentioned item turnover prediction result.

[0017] Optionally, the correction unit is further configured to: determine a constraint matrix corresponding to the above-mentioned layered time series data; generate item turnover constraint information based on the above-mentioned constraint matrix and the above-mentioned item turnover prediction constraint information; construct an item turnover prediction optimization target based on the above-mentioned item turnover prediction result and the above-mentioned covariance matrix; generate a corrected item turnover prediction result based on the above-mentioned item turnover constraint information and the above-mentioned item turnover prediction optimization target.

[0018] Optionally, the device for generating item turnover information further includes: a selection unit and an execution unit. The selection unit is configured to select each item turnover sequence corresponding to the pre-equipment cargo time period from the item turnover prediction result after correction processing. The execution unit is configured to perform the following steps for each pre-equipment cargo area in the pre-equipment cargo area set: select each item turnover sequence corresponding to the above-mentioned pre-equipment cargo area from the above-mentioned each item turnover sequence as the item turnover sequence of each area; determine the sum of each item turnover included in the item turnover sequence of each area as the predicted total item turnover corresponding to the above-mentioned pre-equipment cargo area; generate the replenishment quantity according to the above-mentioned predicted total item turnover and the remaining inventory of the above-mentioned target item; control the item transportation equipment corresponding to the above-mentioned pre-equipment cargo area to transport the target item of the above-mentioned replenishment quantity to the item storage position corresponding to the above-mentioned pre-equipment cargo area.

[0019] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0020] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0021] In a fifth aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which implements the method described in any implementation manner of the above-mentioned first aspect when executed by a processor.

[0022] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the method for generating the information of the article turnover volume of some embodiments of the present disclosure, the time series data of the article turnover volume at different levels can be made to meet the consistency, so that it can be applied to the collaborative decision-making process at each level, and the accuracy of the layered time series data of the predicted article turnover volume is improved. Specifically, the reason for the poor accuracy of the layered time series data of the predicted article turnover volume is that the prediction results of different levels cannot automatically meet the consistency and cannot be applied to the collaborative decision-making process at each level, resulting in the poor accuracy of the layered time series of the predicted article turnover volume. Based on this, the method for generating the information of the article turnover volume of some embodiments of the present disclosure, first, from the layered time series data of the article turnover volume corresponding to the target article, the time series data of the corresponding last layer is extracted as the last layer time series data. Thus, the extracted last layer time series data can represent the time series data of the last layer in the original layered time series data. Then, the time series data of each layer different from the above-mentioned last layer in the above-mentioned layered time series data is determined as the previous layer time series data set. Thus, the time series data of each layer before the last layer can be extracted. Afterwards, the item turnover volume prediction result set corresponding to the preset future time period of the above-mentioned front-layer time series data set is obtained as the item turnover volume prediction constraint information. Among them, each item turnover volume prediction result in the above-mentioned item turnover volume prediction result set meets the consistency condition. Therefore, the obtained item turnover volume prediction result set can represent the prediction results of the time series data of each layer before the last layer input externally, so that the prediction results of the high-level time series input externally can be used as the consistency constraint of the low-level time series. Secondly, according to the above-mentioned last-layer time series data and the item turnover volume generation model, the item turnover volume prediction result corresponding to the above-mentioned preset future time period is generated. Therefore, the original prediction result of the last-layer time series data can be generated by the item turnover volume generation model. Finally, according to the above-mentioned item turnover volume prediction constraint information, the above-mentioned item turnover volume prediction result is corrected and processed, and the corrected item turnover volume prediction result is obtained as the output result of the above-mentioned item turnover volume generation model. Therefore, the original prediction result of the last-layer time series data can be corrected with the externally input item turnover volume prediction result set as the constraint, so that the corrected result can be used as the output result of the item turnover volume generation model. Also, because when the original prediction results of the last layer of time series data are corrected, the set of prediction results of the externally input item turnover volume is used as a constraint, so that the prediction results of the externally input high-level time series can be used as the consistency constraint of the low-level time series. In turn, the time series data of item turnover at different levels can meet the consistency, so that it can be applied to the collaborative decision-making process at each level, and the accuracy of the layered time series data of the predicted item turnover volume is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0024] Figure 1 is a flow chart of some embodiments of the method for generating item turnover information according to the present disclosure;

[0025] Figure 2 It is a schematic diagram of a structure of hierarchical time series data including two levels according to some embodiments of the method for generating article turnover information disclosed in the present invention;

[0026] Figure 3 It is a schematic diagram of a structure of hierarchical time series data including three levels according to some embodiments of the method for generating article turnover information disclosed in the present invention;

[0027] Figure 4 is a flow chart of other embodiments of the method for generating item turnover information according to the present disclosure;

[0028] Figure 5 It is a schematic diagram of the structure of some embodiments of the device for generating the article turnover information according to the present disclosure;

[0029] Figure 6 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0030] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0031] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0032] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0033] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0034] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0035] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0036] Figure 1 The process 100 of some embodiments of the method for generating item turnover information according to the present disclosure is shown. The method for generating item turnover information comprises the following steps:

[0037] Step 101 : extracting the time series data corresponding to the last layer from the layered time series data of the article turnover volume corresponding to the target article as the last layer time series data.

[0038] In some embodiments, the execution subject (e.g., a computing device) of the method for generating item turnover information may extract the time series data of the corresponding last layer from the hierarchical time series data of the item turnover of the corresponding target item as the last layer time series data. The above-mentioned target item may be any item. The item turnover may be the circulation quantity of the above-mentioned target item. The hierarchical time series data corresponding to the item turnover of the target item may be the data of the item turnover of the above-mentioned target item organized in the form of a hierarchical time series. Here, the item turnover in the hierarchical time series data may be the item turnover of the above-mentioned target item in a historical time period. The above-mentioned last layer may be the last layer of the hierarchical time series data. The last layer time series data may include the item turnover time series of each last layer. The item turnover time series may include the item turnover of each item in a historical time period. Each item turnover time series may correspond to a node of the last layer.

[0039] As an example, the structure of a 2-layer hierarchical time series data can be referenced Figure 2 . Figure 2 In the example, the time series data of the first layer may include y1, and the time series data of the second layer may include y2 and y3.

[0040] The structure of 3-layer hierarchical time series data can be referred to Figure 3 . Figure 3 In the example, the first layer of time series data may include y1. The second layer of time series data may include y2 and y3. The third layer of time series data may include y4, y5, y6, and y7. Here, the i-th item turnover time series from time 1 to time M can be expressed as i=1,…,n. n is the number of time series of item turnover. Figure 3 In the example, the number of time series of item turnover can be 7. The constraints satisfied by the hierarchical structure can be y1=y2+y3, y2=y4+y5, y3=y6+y7.

[0041] Step 102, determining the time series data of each layer different from the last layer in the layered time series data as a previous layer time series data set.

[0042] In some embodiments, the execution subject may determine the time series data of each layer different from the last layer in the layered time series data as the previous layer time series data set. In practice, the execution subject may determine the time series data of each layer before the last layer in the layered time series data as the previous layer time series data set.

[0043] Step 103, obtaining a set of prediction results of the item turnover volume corresponding to a preset future time period of the previous layer time series data set as item turnover volume prediction constraint information.

[0044] In some embodiments, the execution subject may obtain the set of predicted results of the item turnover volume corresponding to the preset future time period of the previous layer time series data set as the item turnover volume prediction constraint information. Among them, the set of predicted results of the item turnover volume may be the prediction results of the time series data of each layer before the last layer input externally. Here, the predicted results of the item turnover volume of each layer may be the predicted item turnover volume of each node of the layer within the preset future time period according to the expert experience knowledge. The predicted results of the item turnover volume in the set of predicted results of the above-mentioned item turnover volume meet the consistency condition. The consistency condition may be that the sum of the item turnover volumes corresponding to the same time point and the same branch in the item turnover volume prediction result of the current layer is equal to the sum of the item turnover volumes of the nodes corresponding to the branch and the time point in the previous layer.

[0045] Step 104, generating a prediction result of the item turnover volume corresponding to a preset future time period based on the last layer time series data and the item turnover volume generation model.

[0046] In some embodiments, the above-mentioned execution subject can generate a prediction result of the item turnover volume corresponding to the above-mentioned preset future time period based on the above-mentioned last-layer time series data and the item turnover volume generation model. Among them, the above-mentioned item turnover volume generation model can be a model for predicting the turnover volume of each item in the future time period. Here, the above-mentioned item turnover volume generation model can be an algorithm model or a neural network model. For example, the above-mentioned item turnover volume generation model can be but not limited to: ARIMA, ETS, xgboost, lightgbm, deepar. In practice, the above-mentioned execution subject can input the above-mentioned last-layer time series data into the above-mentioned item turnover volume generation model to obtain the prediction result of the item turnover volume corresponding to the above-mentioned preset future time period. Here, the item turnover volume prediction result can be the predicted item turnover volume of each node in the last layer within the preset future time period.

[0047] In some optional implementations of some embodiments, first, the execution subject may perform missing value filling processing on the last layer of time series data to update the last layer of time series data. In practice, the execution subject may extract the item turnover time series corresponding to the missing value from the layered time series data in response to determining that there are missing values ​​in the last layer of time series data. Then, the item turnover time series may be used to generate the filling value corresponding to the missing value. As an example, an interpolation method may be used to generate the filling value corresponding to the missing value according to the item turnover time series. Then, the item turnover prediction result corresponding to the preset future time period may be generated according to the updated last layer of time series data and the item turnover generation model. Here, the specific implementation method of generating the item turnover prediction result corresponding to the preset future time period may refer to step 104, which will not be repeated here. Thus, the missing values ​​in the last layer of time series data may be filled in advance.

[0048] Optionally, before performing the missing value filling process on the last layer of time series data to update the last layer of time series data, the execution subject may also remove the outliers in the last layer of time series data in response to determining that there are outliers in the last layer of time series data. Thus, the outliers in the last layer of time series data can be removed.

[0049] Step 105, correcting the item circulation volume prediction result according to the item circulation volume prediction constraint information, and obtaining the corrected item circulation volume prediction result as the output result of the item circulation volume generation model.

[0050] In some embodiments, the execution subject may modify the item turnover prediction result according to the item turnover prediction constraint information, and obtain the modified item turnover prediction result as the output result of the item turnover generation model. In practice, the execution subject may first determine the constraint matrix corresponding to the last layer. The constraint matrix may be a sum matrix that aggregates the lowest level time series to the higher level time series. For example, Figure 2 In the hierarchical time series data containing 2 levels, the constraint matrix can be C = [1 1]. Figure 3 In the case of stratified time series data with three levels, the constraint matrix can be:

[0051]

[0052] C1=[1 1 1 1]

[0053]

[0054] Then, the item circulation constraint information can be generated based on the constraint matrix and the item circulation prediction constraint information. In practice, the execution subject can determine the constraint formula as the item circulation constraint information. The constraint formula can be expressed as Among them, b represents the forecast result of item turnover after correction. represents the above-mentioned item turnover prediction constraint information. Here, b and All are in the form of matrices. Afterwards, the preset optimization goal can be solved based on the item circulation constraint information to obtain the corrected item circulation prediction result. Here, the preset optimization goal can be to minimize the absolute value of the difference between the item circulation prediction result and the corrected item circulation prediction result.

[0055] Optionally, the execution subject may also select each item flow volume sequence corresponding to the pre-equipment and stocking time period from the corrected item flow volume prediction result. The pre-equipment and stocking time period may be a preset future time period for stocking. The time range of the pre-equipment and stocking time period is within the preset future time period. In practice, for each node of the last layer, the execution subject may select each item flow volume within the pre-equipment and stocking time period from each item flow volume of the node included in the corrected item flow volume prediction result as the item flow volume sequence. Thus, each item flow volume sequence may be obtained.

[0056] Then, for each pre-equipped cargo area in the pre-equipped cargo area set, perform the following steps:

[0057] The first step is to select the item circulation sequences corresponding to the above-mentioned pre-equipment cargo area from the above-mentioned item circulation sequences as the item circulation sequences for each area. Among them, the above-mentioned pre-equipment cargo area set can be each area that needs to be stocked. It should be noted that the layered time series data can be layered according to regional attributes. In practice, the above-mentioned execution entity can select the item circulation sequences under the branches corresponding to the above-mentioned pre-equipment cargo area from the above-mentioned item circulation sequences as the item circulation sequences for each area.

[0058] In the second step, the sum of the item turnover volumes included in the item turnover volume sequences of the above-mentioned respective regions is determined as the predicted total item turnover volume corresponding to the above-mentioned pre-equipped cargo region.

[0059] The third step is to generate a replenishment quantity according to the predicted total item turnover and the remaining inventory of the target item. The remaining inventory may be the number of target items remaining in the item storage location of the pre-equipped cargo area. The item storage location may be a warehouse. In practice, the execution subject may determine the difference between the predicted total item turnover and the remaining inventory of the target item as the replenishment quantity in response to determining that the predicted total item turnover is greater than the remaining inventory of the target item.

[0060] The fourth step is to control the item transport equipment corresponding to the above-mentioned pre-equipped cargo area to transport the target items of the above-mentioned replenishment quantity to the item storage location corresponding to the above-mentioned pre-equipped cargo area. The above-mentioned item transport equipment can be an unmanned transport vehicle. In this way, automatic replenishment of the stocking area can be achieved.

[0061] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the method for generating the information of the article turnover volume of some embodiments of the present disclosure, the time series data of the article turnover volume at different levels can be made to meet the consistency, so that it can be applied to the collaborative decision-making process at each level, and the accuracy of the layered time series data of the predicted article turnover volume is improved. Specifically, the reason for the poor accuracy of the layered time series data of the predicted article turnover volume is that the prediction results of different levels cannot automatically meet the consistency and cannot be applied to the collaborative decision-making process at each level, resulting in the poor accuracy of the layered time series of the predicted article turnover volume. Based on this, the method for generating the information of the article turnover volume of some embodiments of the present disclosure, first, from the layered time series data of the article turnover volume corresponding to the target article, the time series data of the corresponding last layer is extracted as the last layer time series data. Thus, the extracted last layer time series data can represent the time series data of the last layer in the original layered time series data. Then, the time series data of each layer different from the above-mentioned last layer in the above-mentioned layered time series data is determined as the previous layer time series data set. Thus, the time series data of each layer before the last layer can be extracted. Afterwards, the item turnover volume prediction result set corresponding to the preset future time period of the above-mentioned front-layer time series data set is obtained as the item turnover volume prediction constraint information. Among them, each item turnover volume prediction result in the above-mentioned item turnover volume prediction result set meets the consistency condition. Therefore, the obtained item turnover volume prediction result set can represent the prediction results of the time series data of each layer before the last layer input externally, so that the prediction results of the high-level time series input externally can be used as the consistency constraint of the low-level time series. Secondly, according to the above-mentioned last-layer time series data and the item turnover volume generation model, the item turnover volume prediction result corresponding to the above-mentioned preset future time period is generated. Therefore, the original prediction result of the last-layer time series data can be generated by the item turnover volume generation model. Finally, according to the above-mentioned item turnover volume prediction constraint information, the above-mentioned item turnover volume prediction result is corrected and processed, and the corrected item turnover volume prediction result is obtained as the output result of the above-mentioned item turnover volume generation model. Therefore, the original prediction result of the last-layer time series data can be corrected with the externally input item turnover volume prediction result set as the constraint, so that the corrected result can be used as the output result of the item turnover volume generation model. Also, because when the original prediction results of the last layer of time series data are corrected, the set of prediction results of the externally input item turnover volume is used as a constraint, so that the prediction results of the externally input high-level time series can be used as the consistency constraint of the low-level time series. In turn, the time series data of item turnover at different levels can meet the consistency, so that it can be applied to the collaborative decision-making process at each level, and the accuracy of the layered time series data of the predicted item turnover volume is improved.

[0062] Further references Figure 4 , which shows a process 400 of another embodiment of a method for generating item turnover information. The process 400 of the method for generating item turnover information includes the following steps:

[0063] Step 401 : extracting the time series data corresponding to the last layer from the layered time series data of the article turnover volume corresponding to the target article as the last layer time series data.

[0064] Step 402, determining the time series data of each layer different from the last layer in the layered time series data as a previous layer time series data set.

[0065] Step 403, obtaining a set of item turnover prediction results corresponding to a preset future time period of the previous layer time series data set as item turnover prediction constraint information.

[0066] Step 404, generating a prediction result of the item turnover volume corresponding to a preset future time period based on the last layer time series data and the item turnover volume generation model.

[0067] In some embodiments, the specific implementation of steps 401-404 and the technical effects brought about can be referred to Figure 1 The corresponding steps 101-104 in the embodiments are not described in detail here.

[0068] Step 405, generating residual information according to each historical fitting value and each actual value of the item turnover generation model corresponding to the historical time period and the last layer.

[0069] In some embodiments, the execution subject (e.g., computing device) of the method for generating the item turnover volume information may generate residual information according to the item turnover volume generation model corresponding to the historical time period and the historical fitting values ​​and actual values ​​of the last layer. In practice, the execution subject may determine the residual of each actual value among the actual values ​​and the historical fitting value corresponding to the actual value, and obtain a residual set as the residual information. The historical time period may be a time period before the preset future time period.

[0070] Step 406: Generate a covariance matrix based on the residual information.

[0071] In some embodiments, the execution subject may generate a covariance matrix according to the residual information. In practice, the execution subject may determine the variance-covariance matrix of the residual information as the covariance matrix.

[0072] Step 407, generating a corrected item turnover volume prediction result according to the covariance matrix, the item turnover volume prediction constraint information and the item turnover volume prediction result.

[0073] In some embodiments, the execution entity may generate a corrected item turnover volume prediction result based on the covariance matrix, the item turnover volume prediction constraint information, and the item turnover volume prediction result.

[0074] In some optional implementations of some embodiments, the execution subject may generate a corrected item circulation volume prediction result according to the covariance matrix, the item circulation volume prediction constraint information and the item circulation volume prediction result through the following steps:

[0075] The first step is to determine the constraint matrix corresponding to the layered time series data. In practice, the execution subject may determine a preset sum matrix that aggregates the lowest level time series of the layered time series data to a higher level time series as the constraint matrix.

[0076] The second step is to generate the item circulation constraint information according to the constraint matrix and the item circulation prediction constraint information. Here, the specific implementation method of generating the item circulation constraint information can refer to the method of generating the item circulation constraint information in step 105, which will not be repeated here.

[0077] The third step is to construct the optimization target of item turnover prediction based on the above-mentioned item turnover prediction results and the above-mentioned covariance matrix. In practice, the above-mentioned execution entity can determine the following formula as the optimization target of item turnover prediction:

[0078]

[0079] in, Indicates the forecast result of item turnover after correction. W represents the forecast result of the above-mentioned item turnover volume. b represents the above covariance matrix. It can represent the adjustment item of the bottom-level prediction result after correction. The above formula can represent the minimum sum of squares of the adjustment items weighted by the residual covariance matrix. It can be understood that for time series with larger residual standard deviation (less accurate prediction), larger adjustment items can be tolerated under the same loss function value.

[0080] The fourth step is to generate a corrected item circulation volume prediction result based on the above-mentioned item circulation volume constraint information and the above-mentioned item circulation volume prediction optimization target. In practice, the above-mentioned execution subject can use the above-mentioned item circulation volume constraint information as a constraint condition to solve the above-mentioned item circulation volume prediction optimization target and obtain the corrected item circulation volume prediction result. As an example, it can be solved by the Lagrangian method. Specifically, it can be solved by the following formula:

[0081]

[0082] from Figure 4 It can be seen that Figure 1 Compared with the description of some corresponding embodiments, Figure 4 The process 400 of the method for generating the item turnover information in some corresponding embodiments reflects the extended steps of correcting the item turnover prediction results. Therefore, the schemes described in these embodiments introduce the covariance matrix of the item turnover generation model, so that the fitting effect of the item turnover generation model on the low-level time series can be considered when correcting the item turnover prediction results, thereby improving the accuracy of the item turnover prediction results after the correction processing.

[0083] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a device for generating article turnover information. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0084] like Figure 5 As shown, the device 500 for generating item turnover information in some embodiments includes: an extraction unit 501, a determination unit 502, an acquisition unit 503, a generation unit 504 and a correction unit 505. Among them, the extraction unit 501 is configured to extract the time series data of the corresponding last layer from the stratified time series data of the item turnover volume corresponding to the target item as the last layer time series data; the determination unit 502 is configured to determine the time series data of each layer different from the above-mentioned last layer in the above-mentioned stratified time series data as the previous layer time series data set; the acquisition unit 503 is configured to obtain the above-mentioned previous layer time series data set corresponding to the preset future time period item turnover volume prediction result set as the item turnover volume prediction constraint information, wherein each item turnover volume prediction result in the above-mentioned item turnover volume prediction result set satisfies the consistency condition; the generation unit 504 is configured to generate the item turnover volume prediction result corresponding to the above-mentioned preset future time period according to the above-mentioned last layer time series data and the item turnover volume generation model; the correction unit 505 is configured to correct the above-mentioned item turnover volume prediction result according to the above-mentioned item turnover volume prediction constraint information, and obtain the corrected item turnover volume prediction result as the output result of the above-mentioned item turnover volume generation model.

[0085] Optionally, the generation unit 504 can be further configured to: perform missing value filling processing on the last layer of time series data to update the last layer of time series data; generate an item turnover volume prediction result corresponding to the above-mentioned preset future time period based on the updated last layer of time series data and the above-mentioned item turnover volume generation model.

[0086] Optionally, the generating unit 504 may include: a removing unit (not shown in the figure), configured to remove the abnormal values ​​in the above-mentioned last-layer time series data in response to determining that there are abnormal values ​​in the above-mentioned last-layer time series data.

[0087] Optionally, the correction unit 505 can be further configured to: generate residual information based on the historical time period corresponding to the above-mentioned item turnover generation model and the various historical fitting values ​​and actual values ​​of the above-mentioned last layer; generate a covariance matrix based on the above-mentioned residual information; generate a corrected item turnover prediction result based on the above-mentioned covariance matrix, the above-mentioned item turnover prediction constraint information and the above-mentioned item turnover prediction result.

[0088] Optionally, the correction unit 505 can be further configured to: determine a constraint matrix corresponding to the above-mentioned layered time series data; generate item turnover constraint information based on the above-mentioned constraint matrix and the above-mentioned item turnover prediction constraint information; construct an item turnover prediction optimization target based on the above-mentioned item turnover prediction result and the above-mentioned covariance matrix; generate a corrected item turnover prediction result based on the above-mentioned item turnover constraint information and the above-mentioned item turnover prediction optimization target.

[0089] Optionally, the article circulation volume information generating device 500 may further include: a selection unit and an execution unit (not shown in the figure). The selection unit is configured to select each article circulation volume sequence corresponding to the pre-equipment cargo time period from the corrected article circulation volume prediction result. The execution unit is configured to perform the following steps for each pre-equipment cargo area in the pre-equipment cargo area set: select each article circulation volume sequence corresponding to the above-mentioned pre-equipment cargo area from the above-mentioned each article circulation volume sequence as the article circulation volume sequence of each area; determine the sum of each article circulation volume included in the above-mentioned article circulation volume sequence of each area as the predicted total article circulation volume corresponding to the above-mentioned pre-equipment cargo area; generate the replenishment volume according to the above-mentioned predicted total article circulation volume and the remaining inventory of the above-mentioned target article; control the article transportation equipment corresponding to the above-mentioned pre-equipment cargo area to transport the target article of the above-mentioned replenishment volume to the article storage position corresponding to the above-mentioned pre-equipment cargo area.

[0090] It is understood that the units described in the device 500 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 500 and the units included therein, and will not be described in detail here.

[0091] Reference below Figure 6 , which shows a structural schematic diagram of an electronic device 600 suitable for implementing some embodiments of the present disclosure. Figure 6The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0092] like Figure 6 As shown, the electronic device 600 may include a processing device 601 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0093] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0094] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0095] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0096] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0097] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: extracts the time series data corresponding to the last layer from the hierarchical time series data of the item turnover corresponding to the target item as the last layer time series data; determines the time series data of each layer different from the last layer in the hierarchical time series data as the previous layer time series data set; obtains the item turnover prediction result set corresponding to the preset future time period of the previous layer time series data set as the item turnover prediction constraint information, wherein each item turnover prediction result in the item turnover prediction result set satisfies the consistency condition; generates the item turnover prediction result corresponding to the preset future time period according to the last layer time series data and the item turnover generation model; and performs correction processing on the item turnover prediction result according to the item turnover prediction constraint information, and obtains the corrected item turnover prediction result as the output result of the item turnover generation model.

[0098] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0099] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0100] The units described in some embodiments of the present disclosure may be implemented by software or hardware. The units described may also be set in a processor, for example, may be described as: a processor includes an extraction unit, a determination unit, an acquisition unit, a generation unit, and a correction unit. The names of these units do not constitute limitations on the units themselves in certain cases. For example, the extraction unit may also be described as "a unit that extracts the time series data of the corresponding last layer from the layered time series data of the item turnover volume corresponding to the target item as the last layer time series data."

[0101] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0102] Some embodiments of the present disclosure also provide a computer program product, including a computer program, which implements any of the above-mentioned methods for generating item turnover volume information when executed by a processor.

[0103] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A method for generating item circulation information, comprising: Extracting the time series data corresponding to the last layer from the layered time series data of the item turnover volume corresponding to the target item as the last layer time series data; Determine the time series data of each layer different from the last layer in the layered time series data as a front layer time series data set; Obtaining a set of item circulation volume prediction results corresponding to a preset future time period of the front-layer time series data set as item circulation volume prediction constraint information, wherein each item circulation volume prediction result in the set of item circulation volume prediction results satisfies a consistency condition; Generate a forecast result of the item turnover volume corresponding to the preset future time period according to the last layer time series data and the item turnover volume generation model; According to the item circulation volume prediction constraint information, the item circulation volume prediction result is corrected to obtain the corrected item circulation volume prediction result as the output result of the item circulation volume generation model.

2. The method according to claim 1, wherein: The generating of the item turnover prediction result corresponding to the preset future time period according to the last layer time series data and the item turnover generation model includes: Perform missing value filling processing on the last layer of time series data to update the last layer of time series data; According to the updated last-layer time series data and the article turnover generation model, an article turnover prediction result corresponding to the preset future time period is generated.

3. The method according to claim 2, wherein: Before performing missing value filling processing on the last layer of time series data to update the last layer of time series data, the method further includes: In response to determining that there are abnormal values ​​in the last layer of time series data, the abnormal values ​​in the last layer of time series data are eliminated.

4. The method according to any one of claims 1 to 3, wherein: The step of correcting the item circulation volume prediction result according to the item circulation volume prediction constraint information includes: Generate residual information according to the historical time period corresponding to the item turnover generation model and each historical fitting value and each actual value of the last layer; Generate a covariance matrix according to the residual information; A corrected item circulation volume prediction result is generated according to the covariance matrix, the item circulation volume prediction constraint information and the item circulation volume prediction result.

5. The method according to claim 4, wherein: The generating a corrected item circulation volume prediction result according to the covariance matrix, the item circulation volume prediction constraint information and the item circulation volume prediction result comprises: determining a constraint matrix corresponding to the stratified time series data; Generate item circulation quantity constraint information according to the constraint matrix and the item circulation quantity prediction constraint information; Constructing an optimization target for item turnover prediction based on the item turnover prediction result and the covariance matrix; According to the item circulation volume constraint information and the item circulation volume prediction optimization target, a corrected item circulation volume prediction result is generated.

6. The method according to claim 1, wherein: The method further comprises: Select each item flow volume sequence corresponding to the estimated equipment loading time period from the corrected item flow volume forecast results; For each pre-equipped cargo area in the pre-equipped cargo area collection, perform the following steps: Selecting, from the various article circulation volume sequences, the various article circulation volume sequences corresponding to the pre-equipped cargo area as the article circulation volume sequences for each area; Determine the sum of the respective item turnover amounts included in the item turnover amount sequence of each area as the predicted total item turnover amount corresponding to the pre-equipped cargo area; Generate a replenishment quantity according to the predicted total item turnover volume and the remaining inventory of the target item; The article transport equipment corresponding to the pre-equipped cargo area is controlled to transport the target article of the replenishment quantity to the article storage location corresponding to the pre-equipped cargo area.

7. A device for generating information on item turnover, comprising: An extraction unit is configured to extract the time series data corresponding to the last layer from the layered time series data of the article circulation amount corresponding to the target article as the last layer time series data; a determination unit configured to determine the time series data of each layer different from the last layer in the layered time series data as a front-layer time series data set; an acquisition unit, configured to acquire a set of item circulation volume prediction results corresponding to a preset future time period of the front-layer time series data set as item circulation volume prediction constraint information, wherein each item circulation volume prediction result in the set of item circulation volume prediction results satisfies a consistency condition; A generating unit, configured to generate a prediction result of the item circulation volume corresponding to the preset future time period according to the last layer time series data and the item circulation volume generating model; The correction unit is configured to correct the item circulation volume prediction result according to the item circulation volume prediction constraint information, and obtain the corrected item circulation volume prediction result as the output result of the item circulation volume generation model.

8. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images