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. According to one specific embodiment, the method comprises the steps that feature extraction processing is conducted on historical article circulation time sequence information corresponding to a target article, historical time sequence feature information is obtained, and the historical article circulation time sequence information comprises a historical article circulation volume sequence; generating a model according to the historical time sequence feature information and a pre-trained smooth model coefficient information set, and generating a smooth model coefficient information set; according to the smoothing model coefficient information set and a preset smoothing model set, smoothing the historical article flow volume sequence to obtain a smooth article flow volume sequence; and generating article flow volume information corresponding to the target article according to the smooth article flow volume sequence and a pre-trained article flow volume information generation model. The implementation mode is related to an intelligent supply chain, the loss of articles can be reduced, and transportation resources are saved.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of intelligent supply chain, and specifically to methods, devices, equipment, media and program products for generating item turnover information. Background Art

[0002] Predicting the turnover of items (such as sales) helps warehouse managers to make reasonable inventory plans. At present, when predicting the turnover of items, the method usually adopted is: using statistical methods such as mean-standard deviation method or manually defined rule method as smoothing method to smooth the historical item turnover sequence of a period of history, and then predicting the item turnover of a period of time in the future based on the smoothed historical item turnover sequence to obtain item turnover information.

[0003] However, the inventors have discovered that when the above-mentioned method is used to predict the item turnover volume, the following technical problems often occur: when a statistical method or an artificially defined rule method is used as a smoothing algorithm, for items with regular characteristics of the item turnover volume or items subject to unclear rules, the accuracy of the historical item turnover volume sequence obtained by smoothing is low, resulting in a low accuracy in the predicted item turnover volume. As a result, when the predicted item turnover volume is high, more items are lost, and when the predicted item turnover volume is low, it leads to a waste of transportation resources.

[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 program products for generating item turnover information to solve 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: performing feature extraction processing on historical item turnover time series information corresponding to a target item to obtain historical time series feature information, wherein the historical item turnover time series information comprises a historical item turnover volume sequence; generating a model based on the historical time series feature information and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set, wherein the smoothing model coefficient information set generation model corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; performing smoothing processing on the historical item turnover volume sequence based on the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item turnover volume sequence; generating item turnover volume information corresponding to the target item based on the smoothed item turnover volume sequence and the pre-trained item turnover volume information generation model.

[0008] Optionally, the above-mentioned historical item circulation time series information also includes historical item circulation association information; and the above-mentioned historical item circulation time series information corresponding to the target item is subjected to feature extraction processing to obtain historical time series feature information, including: feature extraction processing is performed on the above-mentioned historical item circulation quantity sequence and the above-mentioned historical item circulation association information to obtain historical time series feature information.

[0009] Optionally, the above-mentioned historical item turnover volume sequence is smoothed according to the above-mentioned smoothing model coefficient information set and the above-mentioned preset smoothing model set to obtain a smoothed item turnover volume sequence, including: determining the smoothing model coefficient information that meets the preset coefficient conditions in the above-mentioned smoothing model coefficient information set as the target smoothing model coefficient information; determining the preset smoothing model corresponding to the above-mentioned target smoothing model coefficient information in the above-mentioned preset smoothing model set as the target smoothing model; inputting the above-mentioned historical item turnover volume sequence into the above-mentioned target smoothing model to obtain a smoothed item turnover volume sequence.

[0010] Optionally, the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set includes a smoothing model weight value, and the above-mentioned smoothing model coefficient information set generation model is obtained by training through the following steps: obtaining a sample set, wherein the samples in the above-mentioned sample set include sample historical item circulation time series information and sample actual item circulation quantity sequence, and the above-mentioned sample historical item circulation time series information includes sample historical item circulation quantity sequence; for each sample in the above-mentioned sample set, executing the following training steps: for each preset smoothing model included in the above-mentioned preset smoothing model set, executing the following steps: converting the sample historical item circulation quantity sequence included in the above-mentioned sample into the sample historical item circulation quantity sequence; The preset smoothing model is input into the above-mentioned preset smoothing model to obtain a sample smoothed item turnover volume sequence; the sample smoothed item turnover volume sequence is input into the above-mentioned item turnover volume information generation model to obtain the item turnover volume information as the sample estimated item turnover volume information; according to the sample actual item turnover volume sequence included in the above-mentioned sample and the above-mentioned sample estimated item turnover volume information, the estimated loss value is determined; the sample historical item turnover time series information included in the above-mentioned sample is subjected to feature extraction processing to obtain the sample historical time series feature information; according to the determined respective estimated loss values ​​and the above-mentioned respective sample historical time series feature information, a smoothing model coefficient information set generation model is generated.

[0011] Optionally, the method further includes: generating item replenishment information corresponding to the target item based on the item turnover information; and controlling the associated replenishment equipment to perform a replenishment operation corresponding to the target item based on the item replenishment information.

[0012] In a second aspect, some embodiments of the present disclosure provide an apparatus for generating item turnover volume information, the apparatus comprising: a feature extraction unit, configured to perform feature extraction processing on historical item turnover time series information corresponding to a target item, to obtain historical time series feature information, wherein the historical item turnover time series information comprises a historical item turnover volume sequence; a first generation unit, configured to generate a model based on the historical time series feature information and a pre-trained smoothing model coefficient information set, to generate a smoothing model coefficient information set, wherein the model generated by the smoothing model coefficient information set corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; a smoothing processing unit, configured to perform smoothing processing on the historical item turnover volume sequence based on the smoothing model coefficient information set and the preset smoothing model set, to obtain a smoothed item turnover volume sequence; a second generation unit, configured to generate item turnover volume information corresponding to the target item based on the smoothed item turnover volume sequence and a pre-trained item turnover volume information generation model.

[0013] Optionally, the above-mentioned historical item circulation time series information also includes historical item circulation association information; the feature extraction unit is further configured to perform feature extraction processing on the above-mentioned historical item circulation quantity sequence and the above-mentioned historical item circulation association information to obtain historical time series feature information.

[0014] Optionally, the smoothing processing unit is further configured to determine the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set that meets the preset coefficient conditions as the target smoothing model coefficient information; determine the preset smoothing model in the above-mentioned preset smoothing model set that corresponds to the above-mentioned target smoothing model coefficient information as the target smoothing model; and input the above-mentioned historical item turnover volume sequence into the above-mentioned target smoothing model to obtain a smoothed item turnover volume sequence.

[0015] Optionally, the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set includes a smoothing model weight value, and the above-mentioned smoothing model coefficient information set generation model is obtained by training through the following steps: obtaining a sample set, wherein the samples in the above-mentioned sample set include sample historical item circulation time series information and sample actual item circulation quantity sequence, and the above-mentioned sample historical item circulation time series information includes sample historical item circulation quantity sequence; for each sample in the above-mentioned sample set, executing the following training steps: for each preset smoothing model included in the above-mentioned preset smoothing model set, executing the following steps: converting the sample historical item circulation quantity sequence included in the above-mentioned sample into the sample historical item circulation quantity sequence; The preset smoothing model is input into the above-mentioned preset smoothing model to obtain a sample smoothed item turnover volume sequence; the sample smoothed item turnover volume sequence is input into the above-mentioned item turnover volume information generation model to obtain the item turnover volume information as the sample estimated item turnover volume information; according to the sample actual item turnover volume sequence included in the above-mentioned sample and the above-mentioned sample estimated item turnover volume information, the estimated loss value is determined; the sample historical item turnover time series information included in the above-mentioned sample is subjected to feature extraction processing to obtain the sample historical time series feature information; according to the determined respective estimated loss values ​​and the above-mentioned respective sample historical time series feature information, a smoothing model coefficient information set generation model is generated.

[0016] Optionally, the article circulation volume information generating device further comprises: a third generating unit and a control unit. The third generating unit is configured to generate article replenishment information corresponding to the target article according to the article circulation volume information; the control unit is configured to control the associated replenishment device to perform a replenishment operation corresponding to the target article according to the article replenishment information.

[0017] 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.

[0018] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0019] 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.

[0020] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the loss of items can be reduced and transportation resources can be saved through the method for generating the item turnover information of some embodiments of the present disclosure. Specifically, the reason for the high loss of items or the waste of transportation resources is that: the statistical method or the artificially defined rule method is used as the smoothing algorithm. For the regular characteristics of the item turnover or the items subject to unclear rules, the accuracy of the historical item turnover sequence obtained by smoothing is low, resulting in a low accuracy of the predicted item turnover. Therefore, when the predicted item turnover is high, the loss of items is high, and when the predicted item turnover is low, it leads to a waste of transportation resources. Based on this, the method for generating the item turnover information of some embodiments of the present disclosure, first, performs feature extraction processing on the historical item turnover time series information corresponding to the target item to obtain historical time series feature information. Among them, the above-mentioned historical item turnover time series information includes the historical item turnover sequence. Thus, information characterizing the characteristics of the historical item turnover sequence can be obtained, which can be used to predict the item turnover sequence in the future time period. Secondly, a model is generated based on the above historical time series feature information and the pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set. Among them, the above smoothing model coefficient information set generation model corresponds to the preset smoothing model set, and the above smoothing model coefficient information set corresponds to the above preset smoothing model set. Therefore, by establishing the association between the characteristics of the historical item turnover sequence and each preset smoothing model, the coefficient information corresponding to each preset smoothing model can be obtained, which can be used to smooth the historical item turnover sequence. Then, according to the above smoothing model coefficient information set and the above preset smoothing model set, the above historical item turnover sequence is smoothed to obtain a smoothed item turnover sequence. Thus, a smoothed item turnover sequence generated according to each preset smoothing model can be obtained, thereby improving the accuracy of the smoothed item turnover sequence. Finally, according to the above smoothed item turnover sequence and the pre-trained item turnover information generation model, the item turnover information corresponding to the above target item is generated. Thus, more accurate item turnover information for a period of time in the future can be obtained, thereby improving the accuracy of the item turnover information. Also, when smoothing the historical item turnover sequence, the compatibility between each smoothing model and the characteristics of the historical item turnover sequence is considered, thereby improving the accuracy of the smoothed item turnover sequence, and further improving the accuracy of the item turnover information. As a result, the loss of items can be reduced and transportation resources can be saved. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] 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.

[0022] Figure 1 is a schematic diagram of an application scenario of a method for generating item turnover information according to some embodiments of the present disclosure;

[0023] Figure 2 is a flow chart of some embodiments of the method for generating item turnover information according to the present disclosure;

[0024] Figure 3 is a schematic diagram of an application scenario of a method for generating item turnover information according to other embodiments of the present disclosure;

[0025] Figure 4 is a flow chart of other embodiments of the method for generating item turnover information according to the present disclosure;

[0026] 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;

[0027] 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

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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".

[0032] 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.

[0033] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0034] Figure 1 It is a schematic diagram of an application scenario of a method for generating item turnover information according to some embodiments of the present disclosure.

[0035] exist Figure 1 In the application scenario, first, the computing device 101 can perform feature extraction processing on the historical item circulation time series information 103 corresponding to the target item 102 to obtain the historical time series feature information 104. The above-mentioned historical item circulation time series information 103 includes a historical item circulation volume sequence 1031. For example, the above-mentioned target item 102 can be "banana". Then, the computing device 101 can generate a model 105 according to the above-mentioned historical time series feature information 104 and a pre-trained smoothing model coefficient information set, and generate a smoothing model coefficient information set 106. The above-mentioned smoothing model coefficient information set generates a model 105 corresponding to a preset smoothing model set 107. The above-mentioned smoothing model coefficient information set 106 corresponds to the above-mentioned preset smoothing model set 107. Afterwards, the above-mentioned historical item circulation volume sequence 1031 can be smoothed according to the above-mentioned smoothing model coefficient information set 106 and the above-mentioned preset smoothing model set 107 to obtain a smoothed item circulation volume sequence 108. Finally, the computing device 101 can generate the item turnover information 110 corresponding to the target item 102 according to the smoothed item turnover sequence 108 and the pre-trained item turnover information generation model 109. As an example, the target item 102 can be "banana", the historical item turnover sequence 1031 can be {125, 145, 141, 120, 120, 150, 150}, the preset smoothing model set 107 can be {moving average model, decision tree model, generative adversarial network model}, the smoothing model coefficient information set 106 can be {0.5, 0.2, 0.3}, the smoothed item turnover sequence 109 can be {125, 135, 137, 133, 130, 133, 136}, and the item turnover information 110 can be {125, 140, 130, 130, 130, 135, 140}.

[0036] It should be noted that the computing device 101 can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, or as a single software or software module. No specific limitation is made here.

[0037] It should be understood that Figure 1 The number of computing devices in the embodiment is only illustrative. Any number of computing devices may be provided according to implementation requirements.

[0038] Continue to refer Figure 2 , shows a process 200 of some embodiments of the method for generating item turnover information according to the present disclosure. The method for generating item turnover information comprises the following steps:

[0039] Step 201 , performing feature extraction processing on the historical item circulation time series information corresponding to the target item to obtain historical time series feature information.

[0040] In some embodiments, the execution subject of the method for generating the item turnover information (for example Figure 1The computing device shown in the figure can perform feature extraction processing on the historical item circulation time series information corresponding to the target item to obtain historical time series feature information. Among them, the above-mentioned target item can be an item whose circulation volume (for example, sales volume) needs to be predicted. The above-mentioned target item can be a fresh item. For example, the above-mentioned target item can be "banana". The above-mentioned historical item circulation time series information can be information on the actual circulation of the target item corresponding to the historical time period. The above-mentioned historical item circulation time series information can include but is not limited to a historical item circulation volume sequence. The above-mentioned historical item circulation volume sequence can be a sequence in which the circulation volumes of each historical item are arranged in ascending order of time. The historical item circulation volume in the above-mentioned historical item circulation volume sequence can be the circulation quantity (for example, sales volume) of the above-mentioned target item in a certain unit time period within the historical time period. For example, when the current time is July 1, 2023, the above-mentioned historical time period can be from June 23, 2023 to June 30, 2023, and a unit time period can be a day between June 23, 2023 and June 30, 2023. The above-mentioned historical time series feature information can characterize the features of the historical item turnover sequence. In practice, the above-mentioned execution subject can perform feature extraction processing on the historical item turnover sequence corresponding to the target item through a preset feature extraction algorithm to obtain the historical time series feature information. Among them, the above-mentioned preset feature extraction algorithm can be an algorithm for extracting features from time series data. The above-mentioned preset feature extraction algorithm can include but is not limited to a feature extraction algorithm based on statistical methods and a feature extraction algorithm based on transformation. For example, the above-mentioned feature extraction algorithm based on statistical methods can be a mean algorithm or a variance algorithm. The above-mentioned feature extraction algorithm based on transformation can be a feature extraction algorithm based on Fourier transform or a principal component analysis method.

[0041] Optionally, the historical item circulation time sequence information may further include historical item circulation association information. The historical item circulation association information may be information associated with the historical item circulation amount sequence. The historical item circulation association information may include but is not limited to weather information, item value reduction (e.g., promotion) information, and item circulation planning (e.g., marketing) information. The weather information may characterize the weather of the historical time period corresponding to the historical item circulation amount sequence. The weather information may include but is not limited to a weather temperature set, a weather humidity set, and a rainfall set. The weather temperature in the weather temperature set may be the temperature of the corresponding unit time period in the historical time period. The weather humidity in the weather humidity set may be the humidity of the corresponding unit time period in the historical time period. The rainfall in the rainfall set may be the rainfall of the corresponding unit time period in the historical time period. The item value reduction information may be information on item value reduction. The item value reduction information may include but is not limited to the value reduction type, the value reduction duration, and the value reduction strength. The value reduction type may characterize the form of value reduction. For example, the value reduction type may be a full amount reduction or a buy one get one free. The above-mentioned value reduction duration may be the duration corresponding to the start time to the end time of the preset value reduction. The above-mentioned item circulation planning information may be the information of the circulation planning. The above-mentioned item circulation planning information may include but is not limited to the circulation planning type, circulation planning duration, and circulation planning intensity. The above-mentioned circulation planning type may characterize the form of the circulation planning. For example, the above-mentioned circulation planning type may be advertising, social media promotion, or fan interaction. The above-mentioned circulation planning duration may be the duration corresponding to the start time to the end time of the preset circulation planning.

[0042] In some optional implementations of some embodiments, the execution subject may perform feature extraction processing on the historical item circulation sequence and the historical item circulation association information to obtain historical time series feature information. In practice, the execution subject may perform feature extraction processing on the historical item circulation sequence and the historical item circulation association information by using the preset feature extraction algorithm to obtain historical time series feature information.

[0043] Step 202, generating a model based on historical time series feature information and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set.

[0044] In some embodiments, the execution subject may generate a model based on the historical time series feature information and the pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set. The smoothing model coefficient information set generation model may be a classification model that takes the historical time series feature information as input and the smoothing model coefficient information set as output. The classification model may be a decision tree or a neural network. The smoothing model coefficient information set generation model corresponds to a preset smoothing model set. The preset smoothing model in the preset smoothing model set may be a pre-set model for smoothing time series data to eliminate the influence of random fluctuations, noise and outliers and retain trend and seasonal information. The preset smoothing model may be, but is not limited to, one of the following: a moving average model, a weighted moving average model, an exponential smoothing model, a piecewise weighted average model, a polynomial fitting model, a local smoothing model, a decision tree model, a support vector machine model, a local anomaly factor model, an integral emission model, a GAN (Generative Adversarial Nets, Generative Adversarial Network) model, a VAE (Variational Auto-Encoders, Variational Auto-Encoder) model. The smoothing model coefficient information in the above-mentioned smoothing model coefficient information set may correspond one to one with the preset smoothing models in the above-mentioned preset smoothing model set. The smoothing model coefficient information in the above-mentioned smoothing model coefficient information set may represent the coefficients of the corresponding preset smoothing model. The smoothing model coefficient information in the above-mentioned smoothing model coefficient information set may include a smoothing model ratio. The above-mentioned smoothing model ratio may be the proportion of the corresponding preset smoothing model in each preset smoothing model, and may also represent the important proportion of the output result of the corresponding preset smoothing model in the output result set corresponding to each preset smoothing model. The execution subject of the training smoothing model coefficient information set generation model may be other computing devices. The above-mentioned other computing devices may be non Figure 1 The computing device shown. For example, the other computing device may be a server. In practice, first, the execution subject may send the historical time series feature information to the other device to generate a smoothing model coefficient information set. Then, the smoothing model coefficient information set sent by the other device is received.

[0045] Optionally, the execution entity of the training smoothing model coefficient information set generation model can be Figure 1 The computing device shown.

[0046] Optionally, the execution entity may directly input the historical time series feature information into a pre-trained smoothing model coefficient information set generation model to obtain a smoothing model coefficient information set.

[0047] Step 203, smoothing the historical item turnover sequence according to the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item turnover sequence.

[0048] In some embodiments, the execution subject may perform smoothing on the historical item turnover sequence according to the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item turnover sequence. In practice, the execution subject may perform the following sub-steps for each preset smoothing model included in the preset smoothing model set:

[0049] In the first sub-step, the smoothing model coefficient information corresponding to the preset smoothing model in the smoothing model coefficient information set is determined as the target smoothing model coefficient information.

[0050] The second sub-step is to input the above historical item turnover sequence into the above preset smoothing model to obtain a first smoothed item turnover sequence.

[0051] The third sub-step is to determine the product of the coefficient corresponding to the above-mentioned target smoothing model coefficient information and the first smoothing item circulation sequence as the second smoothing item circulation sequence.

[0052] Then, for each unit time period corresponding to the historical item turnover volume sequence, the sum of each second smoothed item turnover volume corresponding to the above unit time period in each determined second smoothed item turnover volume sequence is determined as the smoothed item turnover volume.

[0053] Finally, the determined smoothed item turnover volumes are arranged in ascending order according to the corresponding unit time periods to obtain a smoothed item turnover volume sequence.

[0054] Step 204, generating the item turnover information corresponding to the target item according to the smoothed item turnover sequence and the pre-trained item turnover information generation model.

[0055] In some embodiments, the execution subject may generate the item flow volume information corresponding to the target item based on the smoothed item flow volume sequence and the pre-trained item flow volume information generation model. Figure 1The computing device shown may also be other computing devices. The item turnover information generation model may be a time series prediction model that takes a smoothed item turnover sequence as input and item turnover information as output. The above-mentioned item turnover information may be information on predicted item turnover. The above-mentioned item turnover information may include but is not limited to an estimated item turnover sequence. The above-mentioned estimated item turnover sequence may be a sequence in which each estimated item turnover is arranged in ascending order of time. The estimated item turnover in the above-mentioned estimated item turnover sequence may be the estimated turnover quantity of the target item in a unit time period in the corresponding future time period. The above-mentioned time series prediction model may be an LSTM (Long Short Term Memory Network) or a moving average model.

[0056] Optionally, the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set may include a smoothing model weight value. The above-mentioned smoothing model weight value may be the weight of the corresponding preset smoothing model. The above-mentioned smoothing model coefficient information set generation model is trained by the following steps:

[0057] The first step is to obtain a sample set. The samples in the sample set may include sample historical item circulation time series information and sample actual item circulation volume sequence. The sample historical item circulation time series information includes the sample historical item circulation volume sequence. The sample actual item circulation volume sequence may be an actual item circulation volume sequence in a future time period corresponding to the historical item circulation time series information. The future time period of the historical item circulation time series information may be: a time period corresponding to the item circulation volume sequence that can be predicted based on the historical item circulation time series information.

[0058] In the second step, for each sample in the above sample set, perform the following training steps:

[0059] Training step 1: for each preset smoothing model included in the preset smoothing model set, perform the following sub-steps:

[0060] In the first sub-step, the sample historical item turnover volume sequence included in the above sample is input into the above preset smoothing model to obtain the sample smoothed item turnover volume sequence.

[0061] The second sub-step is to input the sample smoothed item turnover sequence into the item turnover information generation model to obtain the item turnover information as the sample estimated item turnover information. The sample estimated item turnover information includes the sample estimated item turnover sequence. The sample estimated item turnover sequence and the sample actual item turnover sequence have the same time period.

[0062] The third sub-step is to determine the estimated loss value according to the sample actual item turnover volume sequence and the sample estimated item turnover volume information included in the sample. In practice, first, the execution subject can perform the following determination steps for each sample actual item turnover volume in the sample actual item turnover volume sequence:

[0063] In the first determination step, the sample estimated item turnover amount corresponding to the sample actual item turnover amount in the sample estimated item turnover amount sequence is determined as the target sample actual item turnover amount. The corresponding sample actual item turnover amount may be: the unit time period corresponding to the sample estimated item turnover amount is the same as the unit time period corresponding to the sample actual item turnover amount.

[0064] The second determination step is to determine the absolute value of the difference between the actual item turnover volume of the sample and the actual item turnover volume of the target sample as the turnover loss value.

[0065] Secondly, the average value of each determined flow loss value is determined as the estimated loss value.

[0066] Training step 2: extracting features from the sample history item flow time series information included in the sample to obtain sample history time series feature information. In practice, the execution subject may extract features from the sample history item flow time series information included in the sample using the preset feature extraction algorithm to obtain sample history time series feature information.

[0067] The second step is to generate a smoothing model coefficient information set generation model based on the determined estimated loss values ​​and the above-mentioned historical time series feature information of each sample. In practice, first, the above-mentioned execution subject can input the determined estimated loss values ​​into the preset loss function to obtain the loss function to be solved. Among them, the above-mentioned preset loss function can be expressed by the following formula:

[0068]

[0069] Wherein, L can represent the estimated loss value. k can represent the kth preset smoothing model. q can represent the qth sample historical item turnover sequence. L kq It can represent the estimated loss value obtained by using the kth preset smoothing model for the qth sample historical item turnover sequence. ω can represent the weight. ω(f q ) k It can represent the weight of the kth preset smoothing model used by the qth sample historical item turnover sequence. k can represent the total number of preset smoothing models in the preset smoothing model set. Q can represent the total number of samples in the sample set.

[0070] Then, the loss function to be solved is solved to obtain the coefficient information of each smoothing model. In practice, the execution subject can solve the loss function to be solved by a preset objective function solving algorithm to obtain the weights corresponding to each sample and the preset smoothing model set as the coefficient information of each smoothing model. Among them, the preset objective function solving algorithm can be an algorithm for solving an unconstrained objective function. For example, the preset objective function solving algorithm can be a gradient descent method. Afterwards, for each sample historical time series feature information included in the above-mentioned sample historical time series feature information, the following sub-steps are performed:

[0071] Sub-step 1: Determine each smoothing model coefficient information corresponding to the above-mentioned sample historical time series feature information in the above-mentioned each smoothing model coefficient information as a smoothing model coefficient information set.

[0072] Sub-step two: determining the above sample historical time series feature information and the above smoothing model coefficient information set as a corresponding information pair.

[0073] Next, the determined corresponding information pairs are input into the initial smoothing model coefficient information set generation model to obtain the smoothing model coefficient information set generation model. The initial smoothing model coefficient information set generation model can be a correspondence table representing the sample historical time series feature information and the smoothing model coefficient information set.

[0074] Optionally, the above execution entity may further perform the following steps:

[0075] The first step is to generate item replenishment information corresponding to the above-mentioned target items based on the above-mentioned item turnover information. Among them, the above-mentioned item replenishment information can be information that characterizes the quantity of the above-mentioned target items that need to be replenished in the warehouse. The above-mentioned item replenishment information may include but is not limited to the item replenishment quantity. The above-mentioned item replenishment quantity can be the quantity of the above-mentioned target items that need to be replenished in the warehouse. In practice, firstly, the above-mentioned execution entity can determine the standard deviation of the estimated item turnover sequence included in the above-mentioned item turnover information. Secondly, the above-mentioned estimated item turnover sequence and the standard deviation of the estimated item turnover sequence are input into the preset replenishment model to obtain the safety stock. Among them, the unit of each unit time period corresponding to the estimated item turnover sequence can be day. The above-mentioned preset replenishment model can be expressed by the following formula:

[0076]

[0077] Wherein, TI may represent the safety stock. TI_day may represent the number of each unit time period corresponding to the estimated item turnover sequence. μ may represent the estimated item turnover in the estimated item turnover sequence. i may represent the i-th unit time period. μ iIt can represent the estimated item turnover in the i-th unit period. CR can represent the critical ratio. Z CR It can represent the quantile of a preset standard normal distribution. For example, the value range of the quantile can be 0.8 to 0.95. σ can represent the standard deviation. σ i It can represent the standard deviation of the estimated item turnover sequence. Then, the product of the above safety inventory and the preset replenishment point percentage is determined as the replenishment point. Among them, the above preset replenishment point percentage can be the percentage of the inventory corresponding to the critical point that needs to be replenished and the safety inventory. For example, the preset replenishment point percentage can be 0.7.

[0078] Afterwards, in response to determining that the existing inventory is less than the replenishment point, the difference between the safety inventory and the existing inventory is determined as the item replenishment quantity. The existing inventory may be the number of target items already in the warehouse. The existing inventory may be the available inventory or the sum of the available inventory and the in-transit inventory. The available inventory may be the number of target items actually stored in the warehouse. The in-transit inventory may be the number of target items in transit.

[0079] Finally, the replenishment quantity of the above items is determined as the item replenishment information.

[0080] The second step is to control the associated replenishment equipment to perform the replenishment operation corresponding to the target item according to the item replenishment information. The replenishment equipment may be a device for transporting items to a target location. The target location may be a warehouse or a shelf. For example, the replenishment equipment may be a transport robot. The replenishment operation may be an operation of transporting the target item to the target location. In practice, the execution subject may control the associated replenishment equipment to transport the target items of the item replenishment quantity included in the item replenishment information to the target location.

[0081] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the loss of items can be reduced and transportation resources can be saved through the method for generating the item turnover information of some embodiments of the present disclosure. Specifically, the reason for the high loss of items or the waste of transportation resources is that: the statistical method or the artificially defined rule method is used as the smoothing algorithm. For the regular characteristics of the item turnover or the items subject to unclear rules, the accuracy of the historical item turnover sequence obtained by smoothing is low, resulting in a low accuracy of the predicted item turnover. Therefore, when the predicted item turnover is high, the loss of items is high, and when the predicted item turnover is low, it leads to a waste of transportation resources. Based on this, the method for generating the item turnover information of some embodiments of the present disclosure, first, performs feature extraction processing on the historical item turnover time series information corresponding to the target item to obtain historical time series feature information. Among them, the above-mentioned historical item turnover time series information includes the historical item turnover sequence. Thus, information characterizing the characteristics of the historical item turnover sequence can be obtained, which can be used to predict the item turnover sequence in the future time period. Secondly, a model is generated based on the above historical time series feature information and the pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set. Among them, the above smoothing model coefficient information set generation model corresponds to the preset smoothing model set, and the above smoothing model coefficient information set corresponds to the above preset smoothing model set. Therefore, by establishing the association between the characteristics of the historical item turnover sequence and each preset smoothing model, the coefficient information corresponding to each preset smoothing model can be obtained, which can be used to smooth the historical item turnover sequence. Then, according to the above smoothing model coefficient information set and the above preset smoothing model set, the above historical item turnover sequence is smoothed to obtain a smoothed item turnover sequence. Thus, a smoothed item turnover sequence generated according to each preset smoothing model can be obtained, thereby improving the accuracy of the smoothed item turnover sequence. Finally, according to the above smoothed item turnover sequence and the pre-trained item turnover information generation model, the item turnover information corresponding to the above target item is generated. Thus, more accurate item turnover information for a period of time in the future can be obtained, thereby improving the accuracy of the item turnover information. Also, when smoothing the historical item turnover sequence, the compatibility between each smoothing model and the characteristics of the historical item turnover sequence is considered, thereby improving the accuracy of the smoothed item turnover sequence, and further improving the accuracy of the item turnover information. As a result, the loss of items can be reduced and transportation resources can be saved.

[0082] Figure 3 It is a schematic diagram of an application scenario of a method for generating item turnover information according to other embodiments of the present disclosure.

[0083] exist Figure 3In the application scenario, first, the computing device 301 can perform feature extraction processing on the historical item circulation time series information 303 corresponding to the target item 302 to obtain the historical time series feature information 304. The historical item circulation time series information 303 includes the historical item circulation volume sequence 3031. For example, the target item 302 can be "banana". Secondly, the computing device 301 can generate a model 305 based on the historical time series feature information 304 and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set 306. The smoothing model coefficient information set generation model 305 corresponds to a preset smoothing model set 307. The smoothing model coefficient information set 306 corresponds to the preset smoothing model set 307. Then, the computing device 301 can determine the smoothing model coefficient information that meets the preset coefficient condition 308 in the smoothing model coefficient information set 306 as the target smoothing model coefficient information 309. Afterwards, the preset smoothing model 3071 corresponding to the target smoothing model coefficient information in the preset smoothing model set 307 is determined as the target smoothing model 310. Next, the computing device 301 may input the above historical item turnover sequence 3031 into the above target smoothing model 310 to obtain a smoothed item turnover sequence 311. Finally, the computing device 301 may generate the item turnover information 313 corresponding to the above target item 302 based on the above smoothed item turnover sequence 311 and the pre-trained item turnover information generation model 312. As an example, the target item 302 mentioned above may be “banana”, the historical item turnover sequence 305 may be {125, 145, 141, 120, 120, 150, 150}, the preset smoothing model set 307 may be {moving average model, decision tree model, generative adversarial network model}, the smoothing model coefficient information set may be {0.5, 0.2, 0.3}, the target smoothing model coefficient information 309 may be “0.5”, the target smoothing model 310 may be “moving average model”, the smoothed item turnover sequence 311 may be {125, 135, 137, 133, 130, 133, 136}, and the item turnover information 313 may be {125, 140, 130, 130, 130, 135, 140}.

[0084] It should be noted that the computing device 301 can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, or it can be implemented as a single software or software module. No specific limitation is made here.

[0085] It should be understood that Figure 3The number of computing devices in the embodiment is only illustrative. Any number of computing devices may be provided according to implementation requirements.

[0086] 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:

[0087] Step 401 , performing feature extraction processing on the historical item circulation time series information corresponding to the target item to obtain historical time series feature information.

[0088] Step 402, generating a model based on historical time series feature information and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set.

[0089] In some embodiments, the specific implementation of steps 401-402 and the technical effects thereof can be referred to in Figure 2 The steps 201-202 in the corresponding embodiment are not described in detail here.

[0090] Step 403: determine the smoothing model coefficient information that meets the preset coefficient condition in the smoothing model coefficient information set as the target smoothing model coefficient information.

[0091] In some embodiments, the execution subject may determine the smoothing model coefficient information in the smoothing model coefficient information set that satisfies a preset coefficient condition as the target smoothing model coefficient information. The preset coefficient condition may be that the coefficient corresponding to the smoothing model coefficient information is the maximum value among the coefficients corresponding to the smoothing model coefficient information set.

[0092] Step 404: determine the preset smoothing model corresponding to the target smoothing model coefficient information in the preset smoothing model set as the target smoothing model.

[0093] In some embodiments, the execution entity may determine a preset smoothing model in the preset smoothing model set corresponding to the target smoothing model coefficient information as the target smoothing model.

[0094] Step 405: input the historical item turnover sequence into the target smoothing model to obtain a smoothed item turnover sequence.

[0095] In some embodiments, the execution entity may input the historical item turnover sequence into the target smoothing model to obtain a smoothed item turnover sequence.

[0096] Step 406, generating item turnover information corresponding to the target item according to the smoothed item turnover sequence and the pre-trained item turnover information generation model.

[0097] In some embodiments, the specific implementation of step 406 and the technical effects thereof can be referred to in Figure 2 The corresponding step 204 in the embodiment will not be described in detail here.

[0098] from Figure 4 It can be seen that Figure 2 Compared with the description of some corresponding embodiments, Figure 4 The process 400 of the method for generating the item turnover volume information in some corresponding embodiments embodies the steps of expanding how to generate the smoothed item turnover volume sequence. Therefore, the schemes described in these embodiments can improve the accuracy of the smoothed item turnover volume sequence, thereby improving the accuracy of the item turnover volume information, thereby reducing the loss of items and saving transportation resources.

[0099] 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 2 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0100] like Figure 5 As shown, in some embodiments, the article circulation volume information generating device 500 includes: a feature extraction unit 501, a first generation unit 502, a smoothing processing unit 503 and a second generation unit 504. The feature extraction unit 501 is configured to perform feature extraction processing on the historical article circulation time series information corresponding to the target article to obtain the historical time series feature information, wherein the historical article circulation time series information includes the historical article circulation volume sequence; the first generation unit 502 is configured to generate a model according to the historical time series feature information and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set, wherein the smoothing model coefficient information set generation model corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; the smoothing processing unit 503 is configured to perform smoothing processing on the historical article circulation volume sequence according to the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed article circulation volume sequence; the second generation unit 504 is configured to generate the article circulation volume information corresponding to the target article according to the smoothed article circulation volume sequence and the pre-trained article circulation volume information generation model.

[0101] Optionally, the above-mentioned historical item circulation time series information also includes historical item circulation association information; the feature extraction unit 501 can be further configured to: perform feature extraction processing on the above-mentioned historical item circulation quantity sequence and the above-mentioned historical item circulation association information to obtain historical time series feature information.

[0102] Optionally, the smoothing processing unit 503 can be further configured to: determine the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set that meets the preset coefficient conditions as the target smoothing model coefficient information; determine the preset smoothing model in the above-mentioned preset smoothing model set that corresponds to the above-mentioned target smoothing model coefficient information as the target smoothing model; input the above-mentioned historical item turnover volume sequence into the above-mentioned target smoothing model to obtain a smoothed item turnover volume sequence.

[0103] Optionally, the smoothing model coefficient information in the above-mentioned smoothing model coefficient information set includes a smoothing model weight value, and the above-mentioned smoothing model coefficient information set generation model is obtained by training through the following steps: obtaining a sample set, wherein the samples in the above-mentioned sample set include sample historical item circulation time series information and sample actual item circulation quantity sequence, and the above-mentioned sample historical item circulation time series information includes sample historical item circulation quantity sequence; for each sample in the above-mentioned sample set, executing the following training steps: for each preset smoothing model included in the above-mentioned preset smoothing model set, executing the following steps: converting the sample historical item circulation quantity sequence included in the above-mentioned sample into the sample historical item circulation quantity sequence; The preset smoothing model is input into the above-mentioned preset smoothing model to obtain a sample smoothed item turnover volume sequence; the sample smoothed item turnover volume sequence is input into the above-mentioned item turnover volume information generation model to obtain the item turnover volume information as the sample estimated item turnover volume information; according to the sample actual item turnover volume sequence included in the above-mentioned sample and the above-mentioned sample estimated item turnover volume information, the estimated loss value is determined; the sample historical item turnover time series information included in the above-mentioned sample is subjected to feature extraction processing to obtain the sample historical time series feature information; according to the determined respective estimated loss values ​​and the above-mentioned respective sample historical time series feature information, a smoothing model coefficient information set generation model is generated.

[0104] Optionally, the item circulation volume information generating device 500 further includes: a third generating unit and a control unit (not shown in the figure). The third generating unit is configured to generate item replenishment information corresponding to the target item according to the item circulation volume information; the control unit is configured to control the associated replenishment device to perform a replenishment operation corresponding to the target item according to the item replenishment information.

[0105] It is understandable that the units recorded in the article circulation information generating device 500 are similar to the reference Figure 2 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.

[0106] Reference below Figure 6 , which shows an electronic device 600 (eg, Figure 1The electronic devices in some embodiments of the present disclosure may include but are not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The 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.

[0107] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, 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.

[0108] 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.

[0109] 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.

[0110] 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.

[0111] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text 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.

[0112] 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: performs feature extraction processing on the historical item flow time series information corresponding to the target item to obtain historical time series feature information, wherein the historical item flow time series information includes a historical item flow volume sequence; generates a model based on the historical time series feature information and a pre-trained smoothing model coefficient information set to generate a smoothing model coefficient information set, wherein the smoothing model coefficient information set generation model corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; performs smoothing processing on the historical item flow volume sequence based on the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item flow volume sequence; generates item flow volume information corresponding to the target item based on the smoothed item flow volume sequence and the pre-trained item flow volume information generation model.

[0113] 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).

[0114] 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.

[0115] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor including a feature extraction unit, a first generation unit, a smoothing processing unit, and a second generation unit. The names of these units do not constitute limitations on the units themselves in certain circumstances, for example, the feature extraction unit may also be described as "a unit that performs feature extraction processing on the historical item flow time series information corresponding to the target item to obtain historical time series feature information".

[0116] 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.

[0117] 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.

[0118] 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 information on item turnover, include: Performing feature extraction processing on the historical item circulation time series information corresponding to the target item to obtain historical time series feature information, wherein the historical item circulation time series information includes a historical item circulation quantity sequence; Generate a smoothing model coefficient information set based on the historical time series feature information and the pre-trained smoothing model coefficient information set generation model, wherein the smoothing model coefficient information set generation model corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; According to the smoothing model coefficient information set and the preset smoothing model set, the historical item turnover sequence is smoothed to obtain a smoothed item turnover sequence; The item turnover volume information corresponding to the target item is generated according to the smoothed item turnover volume sequence and a pre-trained item turnover volume information generation model.

2. The method according to claim 1, in, The historical item circulation time sequence information also includes historical item circulation association information; and The feature extraction process is performed on the historical item circulation time series information corresponding to the target item to obtain the historical time series feature information, including: Feature extraction processing is performed on the historical item circulation quantity sequence and the historical item circulation association information to obtain historical time series feature information.

3. The method according to claim 1, in, The step of smoothing the historical item turnover sequence according to the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item turnover sequence includes: Determine the smoothing model coefficient information that meets the preset coefficient condition in the smoothing model coefficient information set as the target smoothing model coefficient information; Determine a preset smoothing model in the preset smoothing model set corresponding to the target smoothing model coefficient information as a target smoothing model; The historical item turnover sequence is input into the target smoothing model to obtain a smoothed item turnover sequence.

4. The method according to claim 1, in, The smoothing model coefficient information in the smoothing model coefficient information set includes a smoothing model weight value, and the smoothing model coefficient information set generation model is trained by the following steps: Acquire a sample set, wherein the samples in the sample set include sample historical item circulation time sequence information and sample actual item circulation quantity sequence, and the sample historical item circulation time sequence information includes the sample historical item circulation quantity sequence; For each sample in the sample set, the following training steps are performed: For each preset smoothing model included in the preset smoothing model set, the following steps are performed: Inputting the sample historical item turnover volume sequence included in the sample into the preset smoothing model to obtain the sample smoothed item turnover volume sequence; Inputting the sample smoothed item turnover volume sequence into the item turnover volume information generation model to obtain item turnover volume information as sample estimated item turnover volume information; Determine an estimated loss value according to the sample actual item turnover volume sequence included in the sample and the sample estimated item turnover volume information; Performing feature extraction processing on the sample history item circulation time series information included in the sample to obtain the sample history time series feature information; A smoothing model coefficient information set generation model is generated based on the determined estimated loss values ​​and the historical time series characteristic information of each sample.

5. The method according to any one of claims 1 to 4, in, The method further comprises: Generate item replenishment information corresponding to the target item according to the item circulation information; According to the item replenishment information, the associated replenishment equipment is controlled to perform a replenishment operation corresponding to the target item.

6. A device for generating information on the amount of goods transferred, include: A feature extraction unit is configured to perform feature extraction processing on the historical item circulation time series information corresponding to the target item to obtain historical time series feature information, wherein the historical item circulation time series information includes a historical item circulation amount sequence; A first generating unit is configured to generate a model according to the historical time series feature information and a pre-trained smoothing model coefficient information set, and generate a smoothing model coefficient information set, wherein the smoothing model coefficient information set generating model corresponds to a preset smoothing model set, and the smoothing model coefficient information set corresponds to the preset smoothing model set; a smoothing processing unit configured to perform smoothing processing on the historical item turnover sequence according to the smoothing model coefficient information set and the preset smoothing model set to obtain a smoothed item turnover sequence; The second generating unit is configured to generate the item turnover volume information corresponding to the target item according to the smoothed item turnover volume sequence and a pre-trained item turnover volume information generation model.

7. An electronic device, include: 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 5.

8. A computer readable medium having a computer program stored thereon, in, When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. 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 5.

Images