Article flow control method and device, terminal equipment and computer medium

Abstract

Embodiments of the present disclosure disclose an article flow control method, device, terminal equipment and computer medium. A specific implementation of the method includes: obtaining a historical article outbound information set; generating a feature set based on the historical article outbound information set; generating an article flow result data pair set; generating an article flow prediction result data pair set; and sending the article flow prediction result data pair set to a target terminal equipment, wherein the target terminal equipment controls a prepositioned equipment connected by communication to complete an article flow related operation according to the article flow prediction result data pair set. This implementation generates the article flow prediction result data pair set according to the historical article outbound information set and the feature set, which can improve the accuracy and stability of article flow prediction, thereby improving the article flow control level and avoiding problems such as untimely article flow or excessive article flow, and further avoiding waste of flow transportation resources.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, specifically to methods, apparatus, terminal devices, and computer media for controlling the flow of goods. Background Technology

[0002] As a crucial link in the supply chain, logistics warehouses connect upstream to the source of goods and support the flow of goods to numerous offline and online stores. During the digital transformation and online-offline integration phase of large supermarkets, logistics warehouses commonly experience resource allocation issues (such as high inventory and stockouts), leading to high inventory costs and significant resource waste. Effective forecasting of outbound goods flow in logistics warehouses can guide warehouse staff to more accurately order from suppliers and manage inventory, significantly reducing the risks of high inventory and stockouts, thereby improving the overall turnover and resource consumption levels of stores and merchants.

[0003] However, the following technical problems often arise during the process of controlling the flow of goods in logistics warehouses:

[0004] First, existing technologies predict future goods circulation based on historical data of goods circulation in logistics warehouses. This prediction does not take into account various temporary influencing factors, such as weather, seasonal changes, and the characteristics of the goods themselves. In addition, this prediction has poor interpretability, cannot take into account other influencing factors, and has a low level of predictive control over goods circulation, resulting in a significant waste of resources.

[0005] Second, existing technologies generally use predictive models to predict the flow of goods. However, the applicable scenarios for a single model are different, and the accuracy and stability of the prediction are poor, resulting in a low level of control over the flow of goods. Summary of the Invention

[0006] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion that follows. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0007] Some embodiments of this disclosure provide methods, apparatus, terminal devices, and computer media for controlling the flow of goods, in order to solve one or more of the technical problems mentioned in the background section above.

[0008] In a first aspect, some embodiments of this disclosure provide a method for controlling the flow of goods. The method includes: acquiring a set of historical goods outbound information, wherein the set of historical goods outbound information includes a first number of historical goods outbound information sets, and the historical goods outbound information in each set includes an item identifier, a quantity of goods transferred, a time of goods transfer, an in-stock status identifier, and a transfer level index; generating a set of feature sets based on the set of historical goods outbound information, wherein the set of feature sets includes a transfer volume feature set, a special transfer feature set, a date feature set, and other feature sets; and generating a set of goods transfer result data pairs based on the set of historical goods outbound information and the set of feature sets, wherein the set of goods transfer result data pairs includes a first number of items. The data pairs for item circulation results are: an item circulation result data pair consisting of an item identifier and an item circulation result; based on the historical item outbound information set and the item circulation result data pair set, an item circulation prediction result data pair set is generated, wherein the item circulation prediction result data pair set includes a first number of item circulation prediction result data pairs and a second number of item circulation prediction result data pairs, each consisting of an item identifier and an item circulation prediction result; the item circulation prediction result data pair set is sent to the target terminal device, wherein the target terminal device controls the front-end device of the communication connection to complete the item circulation-related operations according to the item circulation prediction result data pair set.

[0009] Secondly, some embodiments of this disclosure provide a goods circulation control device, which includes: an acquisition unit configured to acquire a set of historical goods outbound information, wherein the set of historical goods outbound information includes a first number of historical goods outbound information sets, and the historical goods outbound information in the historical goods outbound information sets includes goods identifier, goods circulation quantity, goods circulation time, goods in-stock status identifier, and circulation level index; a first generation unit configured to generate a feature set based on the set of historical goods outbound information, wherein the feature set includes a circulation quantity feature set, a special circulation feature set, a date feature set, and other feature sets; and a second generation unit configured to generate a set of goods circulation result data pairs based on the set of historical goods outbound information and the feature set, wherein the set of goods circulation result data pairs includes a first The system comprises: a set of item flow result data pairs, each consisting of an item identifier and an item flow result; a third generation unit, configured to generate a set of item flow prediction result data pairs based on a set of historical item outbound information and a set of item flow result data pairs, wherein the set of item flow prediction result data pairs includes a first set of item flow prediction result data pairs and a second set of item flow prediction result data pairs, each consisting of an item identifier and an item flow prediction result; and a control unit, configured to send the set of item flow prediction result data pairs to a target terminal device, wherein the target terminal device controls the front-end device connected to the communication link to complete item flow-related operations based on the set of item flow prediction result data pairs.

[0010] Thirdly, some embodiments of this disclosure provide a terminal device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods in the first aspect.

[0011] Fourthly, some embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements any of the methods in the first aspect.

[0012] The above-described embodiments of this disclosure have the following beneficial effects: the item flow control method of some embodiments of this disclosure can improve the accuracy and stability of item flow prediction, thereby improving the item flow control level, avoiding problems such as untimely item flow or excessive item flow, and thus avoiding waste of flow and transportation resources. Specifically, the inventors have found that the reason for the current low level of item flow control is that the existing technology predicts the future item flow based on historical data of item flow in logistics warehouses. This prediction does not consider various temporary influencing factors, such as weather factors, seasonal changes, and the characteristics of the items themselves. In addition, this prediction has poor interpretability and cannot consider other influencing factors, resulting in a low level of predicted item flow control and causing significant resource waste. Based on this, firstly, some embodiments of this disclosure obtain a set of historical item outbound information. The set of historical item outbound information includes a first number of historical item outbound information sets, and the historical item outbound information in the historical item outbound information set includes item identification, item flow quantity, item flow time, item in-warehouse status identification, and flow level index. Specifically, the historical goods outbound information set contains historical goods circulation data corresponding to specific stores. This data can be cleaned and processed according to prediction needs. Secondly, a feature set is generated based on the historical goods outbound information set. This feature set includes circulation volume feature sets, special circulation feature sets, date feature sets, and other feature sets. Specifically, the feature set can be features obtained through processing according to prediction needs for goods circulation prediction. Thirdly, a goods circulation result data pair set is generated based on the historical goods outbound information set and the feature set. This set includes a first number of goods circulation result data pairs, each consisting of an item identifier and a goods circulation result. Finally, a goods circulation prediction result data pair set is generated based on the historical goods outbound information set and the goods circulation result data pair set. The set of predicted item circulation results includes a first set of predicted item circulation results data pairs, and a second set of predicted item circulation results data pairs, each consisting of an item identifier and a predicted item circulation result. Specifically, the set of predicted item circulation results data pairs may contain predicted item circulation information broken down to a specific date. Finally, the set of predicted item circulation results data pairs is sent to the target terminal device. The target terminal device then controls the front-end device connected to the communication link to complete the item circulation-related operations based on the set of predicted item circulation results data pairs.This method processes historical outbound information and extracts features based on prediction needs. It also makes predictions based on the historical outbound information and feature sets, which improves the accuracy of the final prediction. It can fully consider the impact of different factors on the prediction results, thereby improving the level of goods circulation control, avoiding waste of circulation and transportation resources, and reducing the consumption of goods inventory. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0014] Figure 1 This is an architecture diagram of an exemplary system to which some embodiments of this disclosure can be applied;

[0015] Figure 2 This is a flowchart of some embodiments of the article flow control method according to the present disclosure;

[0016] Figure 3 This is a flowchart of some embodiments of the article flow control device according to the present disclosure;

[0017] Figure 4 This is a schematic diagram of the structure of a terminal device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0018] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0019] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0020] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0021] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0022] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] Figure 1 An exemplary system architecture 100 is shown, to which embodiments of the article flow control method of this disclosure can be applied.

[0024] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0025] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as information processing applications, goods circulation control applications, and data analysis applications.

[0026] Terminal devices 101, 102, and 103 can be either hardware or software. When terminal devices 101, 102, and 103 are hardware, they can be various terminal devices with displays, including but not limited to smartphones, tablets, laptops, and desktop computers. When terminal devices 101, 102, and 103 are software, they can be installed on the terminal devices listed above. They can be implemented as multiple software programs or software modules (e.g., used to provide input of historical item outbound information sets), or as a single software program or software module. No specific limitations are imposed here.

[0027] Server 105 can be a server that provides various services, such as a server that stores the historical item outbound information set input by terminal devices 101, 102, and 103. The server can process the received historical item outbound information set and feed back the processing results (such as the item circulation prediction result data set) to the terminal devices.

[0028] It should be noted that the item flow control method provided in this embodiment can be executed by the server 105 or by the terminal device.

[0029] It should be noted that the server 105 can also directly store the set of historical item outbound information locally. The server 105 can directly extract the set of historical item outbound information locally and process it to obtain the set of item flow prediction results data. In this case, the exemplary system architecture 100 may not include terminal devices 101, 102, 103 and network 104.

[0030] It should also be noted that the terminal devices 101, 102, and 103 may also have an item flow control application installed, in which case the processing method can also be executed by the terminal devices 101, 102, and 103. In this case, the exemplary system architecture 100 may also exclude the server 105 and the network 104.

[0031] It should be noted that server 105 can be either hardware or software. When server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When server 105 is software, it can be implemented as multiple software programs or software modules (for example, used to provide item flow control services), or as a single software program or software module. No specific limitations are made here.

[0032] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0033] Continue to refer to Figure 2 The diagram illustrates a flow 200 of some embodiments of a method for controlling the flow of goods according to the present disclosure. This method includes the following steps:

[0034] Step 201: Obtain the set of historical item outbound information.

[0035] In some embodiments, the entity executing the item flow control method (e.g. Figure 1 The server shown retrieves a set of historical item outbound information. This set includes a first set of historical item outbound information. The historical item outbound information in each set includes item identifier, item circulation quantity, item circulation time, item in-warehouse status identifier, and circulation level index. Specifically, the set of historical item outbound information can be a collection of historical data representing the outbound status of a first set of items of a specific logistics warehouse.

[0036] Optionally, outlier removal processing is performed on the historical item outbound information set to update the set. Specifically, for each historical item outbound information set, the mean and standard deviation of the item turnover data for that product category are calculated based on the item turnover volume corresponding to that product category over the three months preceding the current time. Mean and standard deviation thresholds are determined; these thresholds can be estimated based on historical item turnover data from the logistics warehouse. For each historical item outbound information set, outlier data is removed based on the mean and standard deviation thresholds. Specifically, for each historical item outbound information in the historical item outbound information set, in response to the result that the item turnover quantity of that historical item outbound information is greater than n times the mean threshold plus the standard deviation threshold, the item turnover quantity of that historical item outbound information is replaced with the sum of the mean threshold and n times the standard deviation threshold, where n is a positive integer set according to historical data, and n can be 2, to update the historical item outbound information set. Specifically, for each historical item outbound information set in the historical item outbound information set, based on the item turnover quantity corresponding to that type of product in the historical item outbound information set from the current time point back fifteen months, the historical mean and historical standard deviation of the item turnover data quantity of that type of product are calculated. The historical mean threshold and historical standard deviation threshold are determined; specifically, the historical mean threshold and historical standard deviation threshold can be estimated based on the historical item turnover data of the logistics warehouse. For each historical item outbound information set in the historical item outbound information set, abnormal data in that historical item outbound information set is removed based on the historical mean threshold and historical standard deviation threshold. Specifically, for each historical item outbound information in the historical item outbound information set, in response to the result that the item circulation quantity of the historical item outbound information is greater than n times the mean threshold plus the standard deviation threshold, the item circulation quantity of the historical item outbound information is replaced with the sum of the mean threshold and n times the standard deviation threshold, where n is a positive integer set according to historical conditions, and n can be 2, in order to update the historical item outbound information set.

[0037] For each historical item outbound information set in the historical item outbound information set, the item inventory status identifier for that historical item outbound information set is updated according to the historical item outbound information set. The inventory status identifier includes both inventory and non-in-stock identifiers, and all historical item outbound information in the historical item outbound information set has the same inventory status identifier. Specifically, the inventory status identifier of each historical item outbound information in each historical item outbound information set is determined as an inventory identifier. Specifically, for each historical item outbound information set in the historical item outbound information set, an inventory status update process is performed on that historical item outbound information set. Specifically, for each historical item outbound information in the historical item outbound information set, in response to the item circulation quantity value being empty, the item inventory status identifier of that historical item outbound information is updated. Specifically, in response to the inventory quantity of that historical item outbound information being 0, the item circulation quantity of that historical item outbound information is set to "0", and the item inventory status identifier of that historical item outbound information is determined to be "0". Specifically, for each historical item outbound information set in the historical item outbound information set, in response to the fact that there is a fourth consecutive number of item outbound information entries in the historical item outbound information set with an in-stock status of "0", the in-stock status of the fourth consecutive number of item outbound information entries is determined to be a non-in-stock status. In response to the fact that there is no fourth consecutive number of item outbound information entries in the historical item outbound information set with an in-stock status of "0", the in-stock status of the fourth consecutive number of item outbound information entries is determined to be an in-stock status.

[0038] For each historical item outbound information set in the historical item outbound information set, a circulation level index is determined based on the item's inventory status identifier. The circulation level index characterizes the item's circulation capacity, and all historical item outbound information sets in the set share the same circulation level index. Specifically, for each historical item outbound information set, outbound information with the item's inventory status identifier set as "in stock" is identified as candidate historical item outbound information, resulting in a candidate historical item outbound information set, and ultimately, a set of historical item outbound information sets. Further, for the candidate historical item outbound information sets, data from three months prior to the current time point and from 12 months to 9 months prior to the current time point are filtered to obtain a preferred candidate historical item outbound information set. Specifically, for each set of preferred candidate historical item outbound information in the preferred candidate historical item outbound information set, the weekly average outbound quantity, denoted as N, is generated based on the item turnover quantity value of each preferred candidate historical item outbound information in the preferred candidate historical item outbound information set. The weekly average outbound frequency, denoted as F, is generated based on the item turnover quantity value of each preferred candidate historical item outbound information in the preferred candidate historical item outbound information set over a seven-day period. The quartiles of the weekly average outbound quantity (N) and outbound frequency (F) are calculated, from smallest to largest as N1, N2, N3 and F1, F2, F3, respectively. Then, the turnover level index is determined for each item according to the rules in Table 1 below:

[0039] Serial Number Weekly average outbound volume Outbound frequency Circulation grade indicators 1 N <= N1 F ≤ F1 Low 2 N1 < N <= N2 F1 <F<=F2 middle 3 N2 <N<=N3 F2 <F<=F3 high 4 N3 <N F3 <F Super high

[0040] Table 1

[0041] Specifically, for each historical item outbound information set in the historical item outbound information set, for historical item outbound information in that set whose corresponding historical data records are only three weeks or less, the circulation level index of that historical item outbound information is determined to be low.

[0042] Step 202: Generate a feature set based on the historical item outbound information set.

[0043] In some embodiments, the aforementioned executing entity generates a feature set based on a historical set of outbound goods information. This feature set includes a turnover volume feature set, a special turnover feature set, a date feature set, and other feature sets. Specifically, generating the feature set allows for better prediction of goods turnover through data extraction and reprocessing. Therefore, it is necessary not only to develop universally applicable feature indicators for time-series data but also to analyze the business scenarios and data patterns of logistics warehouses to extract specific indicators. Specifically, based on autocorrelation analysis, time-series data from the past 35 days has a significant impact on future predictions, generating feature sets for 7, 14, 21, 28, and 35 days prior to the current time point.

[0044] Specifically, the turnover characteristic set includes the total outbound volume and average outbound volume of the past i days, where the value of i can be 7, 14, 21, 28, or 35; the total outbound volume, quantity, and average outbound volume of the past nth week, where the value of n can be 1, 2, 3, 4, or 5, representing the first, second, third, fourth, and fifth weeks; the growth trend between the past nth weeks, specifically the growth difference between the past second week and the past first week; the difference between the past nth week and the week with the highest sales volume last year; and the difference between the past nth week and the weekly average of the month with the highest sales volume last year. The special circulation feature set includes the largest number of stores in the early stage of the special circulation period in the past i days, where i can be 7, 14, 21, 28, or 35; the number of days in the special circulation period in the past n-th week, where n can be 1, 2, 3, 4, or 5, representing the first, second, third, fourth, and fifth weeks; and the average weekly outbound volume of the number of stores in the early stage of the special circulation period in the predicted week compared to last year's average weekly outbound volume. The date feature set includes the number of statutory holidays in the past i-th day, where i can be 7, 14, 21, 28, or 35; the number of specific holidays in the past i-th day (number of statutory holidays - number of weekends), where i can be 7, 14, 21, 28, or 35; the number of statutory holidays in the next 7, 14, or 21 days; and the number of specific holidays in the next 7, 14, or 21 days (number of statutory holidays - number of weekends). The remaining feature sets include the time difference with the week with the highest sales last year and the time difference with the month with the highest sales last year.

[0045] Step 203: Based on the historical item outbound information set and feature set, generate a set of item circulation result data pairs.

[0046] In some embodiments, the aforementioned executing entity generates a set of item circulation result data pairs based on a set of historical item outbound information and a set of feature sets. The set of item circulation result data pairs includes a first number of item circulation result data pairs, each consisting of an item identifier and an item circulation result.

[0047] Optionally, for each historical item outbound information set in the historical item outbound information set, in response to the circulation level index of that historical item outbound information set being designated as a first-class index, the moving average method is used to generate the item circulation result data for that historical item outbound information set, thus obtaining the item circulation result dataset. Specifically, circulation level indices with a circulation level of "low" are determined as first-class indices. Moving average is a widely used forecasting algorithm. The forecast results are relatively stable, highly interpretable, and very suitable for industrial production. The moving average method calculates the average value of time series data over a specified period. Specifically, for each historical item outbound information set in the historical item outbound information set, the predicted quantity for the next 7 days can be calculated by multiplying the daily average of the item turnover quantity marked as "in stock" over the past 28 days by 7. Specifically, the predicted quantity for the next 7 days corresponds to the predicted quantities for days 1 to 7. Similarly, the predicted quantity for the next 14 days can be calculated by multiplying the daily average of the item turnover quantity marked as "in stock" over the past 28 days by 14. Specifically, the predicted quantity for the next 14 days corresponds to the predicted quantities for days 8 to 15. Finally, the predicted quantity for the next 21 days can be calculated by multiplying the daily average of the item turnover quantity marked as "in stock" over the past 28 days by 21. Specifically, the predicted quantity for the next 21 days corresponds to the predicted quantities for days 15 to 21. The average value is obtained by combining the predicted quantities for the next 7 days, the next 14 days, and the next 21 days. Based on the item turnover result dataset and the historical item outbound information set, an item turnover result data pair set is generated. Specifically, the item circulation result dataset is added to the historical item outbound set according to the corresponding historical item outbound set to obtain the item circulation result data pair set.

[0048] For each historical item outbound information set in the historical item outbound information set, in response to the circulation level index of that historical item outbound information set being classified as a second-category index, the feature set corresponding to that historical item outbound information set is input into the prediction model to generate an item circulation result dataset for that historical item outbound information set, thus obtaining a set of item circulation result datasets. The prediction model includes a gradient enhancement module and a linear regression module. Specifically, circulation level indices of "medium," "high," and "very high" are determined as first-category indicators. Based on the historical item outbound information set and the feature set, a target feature set is generated. Specifically, data processing and calculations are performed on the data included in the historical item outbound information set to obtain the target feature set. The target feature set includes the feature of the maximum daily average outbound volume in the previous year, the feature of the difference in the maximum daily average outbound volume, the feature of the predicted difference in the maximum weekly outbound volume in the previous year, the feature of the daily average outbound volume of the previous year by container value, the feature of the first average outbound volume, the feature of the second average outbound volume, the feature of the standard deviation of the first average outbound volume, the feature of the first category label, the feature of the second category label, the feature of the third category label, and the feature of the fourth category label.

[0049] Optionally, the target feature set is input into the gradient enhancement module to obtain the first item circulation result dataset. Specifically, the gradient enhancement module can be a Light Gradient Boosting Machine (LightGBM), whose main idea is to use a weak classifier (decision tree) for iterative training to obtain the optimal model. LightGBM has advantages such as good training effect and low overfitting. The target feature set is input into the linear regression module to obtain the second item circulation result dataset. Specifically, linear regression (LR) is a regression analysis that uses the least squares function of the linear regression equation to model the relationship between one or more independent variables and the dependent variable. Based on the first and second item circulation result datasets, an item circulation result dataset for this historical item outbound information set is generated. For each first item circulation result data in the first item circulation result dataset, the following formula is used to generate item circulation result data based on the corresponding second item circulation result data in the second item circulation result dataset, to obtain the item circulation result dataset:

[0050] predict = α * predict Lgbm +β*predict Lr

[0051] Where α and β are predetermined control parameters, predict Lgbm For the flow result data of the first item, predictLr For the second item circulation result data corresponding to the first item circulation result data, Lgbm is used to represent the result obtained by the gradient enhancement module, and Lr is used to represent the result obtained by the linear regression module.

[0052] Based on the dataset of goods circulation results and the dataset of historical goods outbound information, a set of goods circulation result data pairs is generated. Specifically, the dataset of goods circulation results is added to the set of historical goods outbound information according to the corresponding set of historical goods outbound information to obtain the set of goods circulation result data pairs.

[0053] The optional content in step 203 above, namely: "Predicting the flow of goods information based on the fusion scheme of flow classification is an inventive point of this disclosure, which solves the second technical problem mentioned in the background art: "The prior art generally uses a prediction model to predict the flow of goods. The applicable scenarios of a single model are different, and the accuracy and stability of the prediction are poor, resulting in a low level of flow control of goods." The factors contributing to the current poor accuracy and stability of goods circulation forecasting are often as follows: A single model yields different forecast results for goods with high and low circulation levels; using only one model cannot adapt to the changing circulation needs of different goods, resulting in poor forecast accuracy and stability. Addressing these factors would improve the level of goods circulation forecasting. To achieve this, this disclosure proposes a method that integrates multiple models for goods circulation forecasting. For retail goods with high outbound demand in logistics warehouses, whose circulation levels are low, a moving average model that is simple to implement, consumes fewer resources, and has the fastest forecasting speed is used. For goods with higher circulation levels, the accuracy requirement is higher; inaccurate forecasts can significantly impact business operations, causing problems such as poor goods circulation or wasted warehousing resources. Linear Regression and Light GBM models can be selected to address the poor accuracy and stability of single-model forecasts, thereby improving the accuracy and stability of goods circulation forecasting in logistics warehouses, enhancing the level of goods circulation forecasting, and thus solving technical problem two.

[0054] Step 204: Based on the historical item outbound information set and the item circulation result data pair set, generate the item circulation prediction result data pair set.

[0055] In some embodiments, the aforementioned executing entity generates a set of predicted item flow results based on a set of historical item outbound information and a set of item flow result data pairs. The set of predicted item flow results includes a first number of predicted item flow result data pairs and a second number of predicted item flow result data pairs, each consisting of an item identifier and an item flow prediction result. Specifically, the set of predicted item flow results includes predictions of item flow for three time periods: days 1-7, days 8-14, and days 15-21. To facilitate application in actual replenishment scenarios, the outbound volume needs to be broken down into daily outbound volumes. Data from the previous two months and the same period last year (two months prior) are retrieved, and the percentage of outbound volume for each item on Sundays (Monday to Sunday) is calculated for each period, denoted as A and B. Because seasonal factors need to be considered, the final breakdown ratio for each item is a weighted fusion of A and B, resulting in S = αA + βB (α + β = 1). Specifically, Table 2 provides an example of the results as follows:

[0056] Item label Sunday Split ratio 49001 on Monday 0 49001 Tuesday 0 49001 Wednesday 0.18 49001 Thursday 0 49001 Friday 0.31 49001 Saturday 0 49001 Zhou Tian 0.49

[0057] Table 2

[0058] Specifically, referring to the example results given in Table 2, predict the day of the week for each day of the next 7 days and take the corresponding proportion from the table (for example, Wednesday, the corresponding proportion is 0.18), and denot it as split_ratio. The predicted amount for that day = split_ratio * total outbound volume for the next 7 days.

[0059] Step 205: Send the predicted data set of goods circulation results to the target terminal device.

[0060] In some embodiments, the aforementioned executing entity sends the goods circulation prediction result data to the target terminal device. The target terminal device, based on the goods circulation prediction result data, controls the front-end device connected to the control group to complete the goods circulation-related operations. The target terminal device can be a device connected to the aforementioned executing entity. The target terminal device can be a mobile phone or a computer. Specifically, the target terminal device can use the goods prediction result data to complete the allocation and storage of goods in the logistics warehouse based on the predicted information on the circulation of each type of goods provided in the control group, thereby meeting the warehouse's outbound goods needs while reducing inventory and saving inventory resources.

[0061] Figure 2One embodiment provides the following advantages: It acquires a set of historical outbound goods information; generates a set of feature sets based on this set; generates a set of goods circulation result data pairs; generates a set of goods circulation prediction result data pairs; and sends the set of goods circulation prediction result data pairs to a target terminal device. The target terminal device then controls the front-end device connected to the communication network to complete goods circulation-related operations based on the set of goods circulation prediction result data pairs. This implementation simultaneously generates the set of goods circulation prediction result data pairs based on both the historical outbound goods information set and the feature set, improving the accuracy and stability of goods circulation prediction, thereby enhancing goods circulation control and preventing problems such as untimely or excessive goods circulation, ultimately avoiding waste of transportation resources.

[0062] Further reference Figure 3 As an implementation of the above figures and methods, this disclosure provides some embodiments of an article flow control device, which are similar to... Figure 2 Corresponding to the above-described method embodiments, the device can be specifically applied to various terminal devices.

[0063] like Figure 3As shown, some embodiments of the goods circulation control device 300 include: an acquisition unit 301, a first generation unit 302, a second generation unit 303, a third generation unit 304, and a control unit 305. The acquisition unit 301 is configured to acquire a set of historical goods outbound information, which includes a first number of historical goods outbound information sets. The historical goods outbound information in each set includes goods identifier, goods circulation quantity, goods circulation time, goods in-stock status identifier, and circulation level index. The first generation unit 302 is configured to generate a set of feature sets based on the historical goods outbound information set, which includes a circulation quantity feature set, a special circulation feature set, a date feature set, and other feature sets. The second generation unit 303 is configured to generate a set of goods circulation result data pairs based on the historical goods outbound information set and the feature set, which includes a first number of goods circulation result data pairs, each consisting of a goods identifier and a goods circulation result. The third generation unit 304 is configured to generate a set of predicted item flow results data pairs based on a set of historical item outbound information and a set of item flow result data pairs. The set of predicted item flow results data pairs includes a first number of predicted item flow results data pairs and a second number of predicted item flow results data pairs. Each predicted item flow results data pair is a data pair consisting of an item identifier and an item flow prediction result. The control unit 305 is configured to send the set of predicted item flow results data pairs to a target terminal device. The target terminal device controls the front-end device with which it has a communication connection to complete item flow-related operations based on the set of predicted item flow results data pairs.

[0064] It is understandable that the units described in the device 300 are related to the reference. Figure 2 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 300 and the units contained therein, and will not be repeated here.

[0065] The following is for reference. Figure 4 It shows a schematic diagram of the structure of a computer system 400 suitable for implementing a terminal device according to embodiments of the present disclosure. Figure 4 The terminal device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0066] like Figure 4As shown, the computer system 400 includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 402 or programs loaded from storage section 406 into Random Access Memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of the system 400. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0067] The following components are connected to I / O interface 405: storage section 406, including hard disks, etc.; and communication section 407, including network interface cards such as LAN (Local Area Network) cards, modems, etc. Communication section 407 performs communication processing via a network such as the Internet. Drive 408 is also connected to I / O interface 405 as needed. Removable media 409, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on drive 408 as needed so that computer programs read from them can be installed into storage section 406 as needed.

[0068] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 407, and / or installed from removable medium 409. When the computer program is executed by central processing unit (CPU) 401, it performs the functions defined in the methods of this disclosure. It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0069] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone 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 remote computers, the remote computer can 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 can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0070] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0071] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A method for controlling the flow of goods, comprising: Obtain a set of historical item outbound information, wherein the set of historical item outbound information includes a first number of historical item outbound information sets, and the historical item outbound information in the historical item outbound information set includes item identifier, item circulation quantity, item circulation time, item in-stock status identifier, and circulation level index; Based on the set of historical item outbound information, a set of feature sets is generated, wherein the set of feature sets includes a turnover volume feature set, a special turnover feature set, a date feature set, and other feature sets; Based on the set of historical item outbound information and the set of features, a set of item circulation result data pairs is generated, wherein the set of item circulation result data pairs includes a first number of item circulation result data pairs, and each item circulation result data pair is a data pair composed of the item identifier and the item circulation result; Based on the historical item outbound information set and the item circulation result data pair set, an item circulation prediction result data pair set is generated. The item circulation prediction result data pair set includes a first number of item circulation prediction result data pair sets and a second number of item circulation prediction result data pairs. Each item circulation prediction result data pair is a data pair composed of the item identifier and the item circulation prediction result. The predicted data set of goods circulation is sent to the target terminal device, wherein the target terminal device controls the front-end device with which the predicted data set of goods circulation is communicated to complete the relevant operations of goods circulation based on the predicted data set of goods circulation.

2. The method according to claim 1, wherein, The acquisition of the historical item outbound information set also includes: Outlier removal is performed on the historical item outbound information set in order to update the historical item outbound information set; For each historical item outbound information set in the historical item outbound information set, update the item in-stock status identifier of the historical item outbound information set according to the historical item outbound information set, wherein the value of the in-stock status identifier includes in-stock identifier and non-in-stock identifier, and each historical item outbound information set in the historical item outbound information set has the same item in-stock status identifier. For each historical item outbound information set in the set of historical item outbound information sets, based on the item in-stock status identifier, the circulation level index of the historical item outbound information set is determined, wherein the circulation level index characterizes the circulation capacity of the item, and each historical item outbound information set in the set of historical item outbound information sets has the same circulation level index.

3. The method according to claim 2, wherein, The process of generating a set of item circulation result data pairs based on the historical item outbound information set and the feature set includes: For each historical item outbound information set in the historical item outbound information set, in response to the circulation level index of the historical item outbound information set being a first-class index, the item circulation result data of the historical item outbound information set is generated using the moving average method to obtain the item circulation result dataset. Based on the item circulation result dataset and the historical item outbound information set, the item circulation result data pair set is generated.

4. The method according to claim 3, wherein, The process of generating a set of item circulation result data pairs based on the historical item outbound information set and the feature set further includes: For each historical item outbound information set in the historical item outbound information set, in response to the circulation level index of the historical item outbound information set being a second type of index, the feature set corresponding to the historical item outbound information set is input into the prediction model to generate the item circulation result dataset of the historical item outbound information set, so as to obtain the set of item circulation result datasets. The set of item circulation result data pairs is generated based on the set of the item circulation result dataset and the set of the historical item outbound information dataset.

5. The method according to claim 4, wherein, The prediction model includes a gradient enhancement module and a linear regression module, and The step of inputting the feature set corresponding to the historical item outbound information set into the prediction model to generate an item circulation result dataset for the historical item outbound information set includes: Based on the historical item outbound information set and the aforementioned feature set set, a target feature set set is generated; The target feature set is input into the gradient enhancement module to obtain the first item circulation result dataset; The target feature set is input into the linear regression module to obtain the second item circulation result dataset; Based on the first item circulation result dataset and the second item circulation result dataset, an item circulation result dataset of the historical item outbound information set is generated.

6. The method according to claim 5, wherein, The target feature set includes the feature of the maximum daily average outbound volume in the previous year, the feature of the difference in the maximum daily average outbound volume, the feature of the predicted difference in the maximum weekly outbound volume in the previous year, the feature of the daily average outbound volume of the previous year by container value, the feature of the first average outbound volume, the feature of the second average outbound volume, the feature of the standard deviation of the first average outbound volume, the feature of the first category label, the feature of the second category label, the feature of the third category label, and the feature of the fourth category label.

7. The method according to claim 6, wherein, The step of generating the historical item outbound information set, based on the first item flow result dataset and the second item flow result dataset, includes: For each first item circulation result data in the first item circulation result dataset, the following formula is used to generate item circulation result data based on the second item circulation result data corresponding to that first item circulation result data in the second item circulation result dataset, so as to obtain the item circulation result dataset: predict=α*predict Lgbm +β*predict Lr Where α and β are predetermined control parameters, predict Lgbm For the flow result data of the first item, predict Lr For the second item circulation result data corresponding to the first item circulation result data, Lgbm is used to characterize the result obtained by the gradient enhancement module, and Lr is used to characterize the result obtained by the linear regression module.

8. A goods circulation control device, comprising: The acquisition unit is configured to acquire a set of historical item outbound information, wherein the set of historical item outbound information includes a first number of historical item outbound information sets, and the historical item outbound information in the historical item outbound information set includes item identifier, item circulation quantity, item circulation time, item in-stock status identifier, and circulation level index. The first generation unit is configured to generate a feature set based on the set of historical item outbound information, wherein the feature set includes a turnover feature set, a special turnover feature set, a date feature set, and other feature sets; The second generation unit is configured to generate a set of item circulation result data pairs based on the set of historical item outbound information and the set of feature sets, wherein the set of item circulation result data pairs includes a first number of item circulation result data pairs, and each item circulation result data pair is a data pair composed of the item identifier and the item circulation result; The third generation unit is configured to generate a set of item flow prediction result data pairs based on the set of historical item outbound information and the set of item flow result data pairs. The set of item flow prediction result data pairs includes a first number of item flow prediction result data pairs and a second number of item flow prediction result data pairs. Each item flow prediction result data pair is a data pair composed of an item identifier and an item flow prediction result. The control unit is configured to send the item circulation prediction result data set to the target terminal device, wherein the target terminal device controls the front-end device with which the item circulation prediction result data set is communicated to complete the item circulation-related operations.

9. A terminal device, comprising: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-8.

Images