Retail store warehouse in and out number of items time-sharing forecast system

Abstract

The present invention relates to a time-sharing prediction system for the number of in-and-out items in a warehouse of a retail store, comprising: a prediction execution mechanism, which is used to adopt a logistics prediction model to intelligently predict the predicted value of the number of in-warehouse items and the predicted value of the number of out-warehouse items in the warehouse of the retail store in the current time interval according to the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of in-warehouse items and the number of out-warehouse items corresponding to each time interval before the current time interval of the warehouse of the retail store, and multiple configuration information of the warehouse of the retail store; a data transmission mechanism, which is used to transmit the intelligent prediction result. Through the present invention, it is possible to simultaneously predict the number of in-warehouse items and the number of out-warehouse items of the warehouse of the retail store in the current time interval, design a logistics prediction model with a customized structure, and selectively screen multiple basic data, thereby providing valuable reference information for the resource allocation strategy and space utilization of the warehouse of the retail store.

Description

Technical Field

[0001] The present invention relates to the field of warehouse management of retail stores, and in particular to a time-sharing forecasting system for the number of items entering and leaving the warehouse of a retail store. Background Art

[0002] At present, when offline retail stores conduct online sales, they need to connect to multiple Internet sales platforms, including Meituan, Ele.me, JD.com, Taoxianda, Douyin, etc. Each third-party sales platform has its own operation management platform and corresponding interfaces. If retail companies directly use the backend of the third-party platform to manage online sales, a lot of manpower will be required. If they connect separately through the system, a lot of production and research resources will also be required.

[0003] Therefore, there is a need to meet the needs of merchants to connect to multiple Internet sales platforms in one operation through a unified platform. The applicant's Bailian Business Operating System unifies the content of different third-party platforms into a set of standards. Merchants only need to perform one operation on the Bailian Business Operating System to meet the sales preparation and order processing work of all three-party platforms. For example, a commodity maintenance synchronizes multiple platforms, and orders from different platforms can be processed uniformly through standardized operating methods. This reduces the problems of repeated maintenance, repeated development of docking, and multiple order processing forms for retail merchants. This platform has brought specific benefits in practice, which is specifically manifested in the fact that sales volume has been greatly improved, so the pressure is transferred to the management of warehouses in retail stores. In this case, the management of warehouses in retail stores needs to be more intelligent to meet this demand.

[0004] Currently, it is necessary to simultaneously predict the number of incoming and outgoing items of a retail store's warehouse in the current time interval. The current time interval starts at the current moment, so it actually belongs to the future time interval. On the basis of the successful prediction of the above two item counts, it can provide valuable reference information for the resource allocation strategy of the retail store's warehouse in the current time interval and the space utilization of the retail store's warehouse. Obviously, the prior art lacks a prediction mechanism that needs to simultaneously predict the number of incoming and outgoing items of a retail store's warehouse in the current time interval. Summary of the invention

[0005] In order to solve the technical problems in the prior art, the present invention provides a time-sharing prediction system for the number of items entering and leaving the warehouse of a retail store, which can design a logistics prediction model with a customized structure to simultaneously predict the number of items entering and leaving the warehouse of the retail store in the current time period, and selectively screen multiple basic data to simultaneously predict the number of items entering and leaving the warehouse of the retail store in the current time period, thereby ensuring the stability and reliability of the prediction results, and providing valuable reference information for the resource allocation strategy of the retail store's warehouse in the current time period and the space utilization of the retail store's warehouse.

[0006] According to the present invention, a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store is provided, the system comprising:

[0007] A configuration parsing device is used to obtain multiple configuration information of a warehouse of a retail store, wherein the multiple configuration information of the warehouse of the retail store is the number of city blocks served by the warehouse of the retail store, the number of permanent urban population, and the total geographical area occupied by each city block served by the warehouse of the retail store;

[0008] A content collection device is used to obtain the number of incoming items and the number of outgoing items of the warehouse of the retail store corresponding to each time interval before the current time interval;

[0009] A target assembly device is used to perform training operations on a single-layer neural network each time to obtain a single-layer neural network after each training operation is completed, and output the single-layer neural network after each training as a logistics prediction model, wherein the number of times the single-layer neural network is trained is positively correlated with the number of city blocks served by the warehouse of the retail store;

[0010] A prediction execution mechanism is connected to the configuration analysis device, the content acquisition device and the target assembly device respectively, and is used to use the logistics prediction model to intelligently predict the predicted value of the number of inbound pieces and the predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval according to the maximum number of pieces stored in the warehouse of the retail store, the duration of each time interval, the number of inbound pieces and the number of outbound pieces of the warehouse of the retail store corresponding to each time interval before the current time interval, and multiple configuration information of the warehouse of the retail store;

[0011] A data transmission mechanism, connected to the prediction execution mechanism, for wirelessly transmitting the predicted value of the number of inbound pieces and the predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval to a remote smart logistics service node;

[0012] Wherein, the logistics prediction model is used to intelligently predict the predicted value of the number of inbound items and the predicted value of the number of outbound items of the warehouse of the retail store in the current time interval according to the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store, including: synchronously inputting the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store into the logistics prediction model;

[0013] Among them, each training operation is performed on the single-layer neural network to obtain the single-layer neural network after each training operation, and the single-layer neural network after each training is output as a logistics prediction model, and the number of times the single-layer neural network is trained is positively correlated with the number of urban blocks served by the warehouse of the retail store, including: using a numerical conversion function to represent the numerical conversion relationship between the number of times the single-layer neural network is trained and the number of urban blocks served by the warehouse of the retail store.

[0014] It can be seen that the present invention has at least the following three important inventive features:

[0015] Firstly: in order to perform simultaneous prediction of the number of inbound items and outbound items of the warehouse of the retail store in the current time interval, a logistics prediction model with a customized structure is designed. Specifically, the logistics prediction model is a single-layer neural network after completing each training operation, and the number of trainings of the single-layer neural network is positively correlated with the number of urban blocks served by the warehouse of the retail store, so that logistics prediction models with different structures are customized for the warehouses of different retail stores;

[0016] Secondly: In order to perform simultaneous prediction of the number of incoming and outgoing items of the retail store's warehouse in the current time interval, multiple basic data are targeted and screened, including the number of incoming items and the number of outgoing items of the retail store's warehouse in each time interval before the current time interval, as well as multiple configuration information of the retail store's warehouse, thereby ensuring the stability and reliability of the prediction results;

[0017] Again: In the specific design of the positive correlation between the number of training times of the single-layer neural network and the number of city blocks served by the warehouse of the retail store, a numerical conversion function is used to express the numerical conversion relationship of the positive correlation, wherein the number of city blocks served by the warehouse of the retail store is the input parameter and the number of training times of the single-layer neural network is the output parameter. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0019] Figure 1 Schematic diagram of the internal structure of a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment A of the present invention.

[0020] Figure 2 Schematic diagram of the internal structure of a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment B of the present invention.

[0021] Figure 3 Schematic diagram of the internal structure of a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment C of the present invention. DETAILED DESCRIPTION

[0022] The implementation scheme of the time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 The internal structure diagram of the time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment A of the present invention is shown, and the system comprises:

[0024] A configuration parsing device is used to obtain multiple configuration information of a warehouse of a retail store, wherein the multiple configuration information of the warehouse of the retail store is the number of city blocks served by the warehouse of the retail store, the number of permanent urban population, and the total geographical area occupied by each city block served by the warehouse of the retail store;

[0025] Specifically, the configuration parsing device is used to obtain multiple configuration information of the warehouse of the retail store, and the multiple configuration information of the warehouse of the retail store is the number of urban blocks served by the warehouse of the retail store, the number of permanent urban population, and the total geographical area occupied by each urban block served by the warehouse of the retail store. The configuration parsing device is selected to use a programmable logic device to implement the configuration parsing device, which is used to obtain multiple configuration information of the warehouse of the retail store, and the multiple configuration information of the warehouse of the retail store is the number of urban blocks served by the warehouse of the retail store, the number of permanent urban population, and the total geographical area occupied by each urban block served by the warehouse of the retail store;

[0026] A content collection device is used to obtain the number of incoming items and the number of outgoing items of the warehouse of the retail store corresponding to each time interval before the current time interval;

[0027] A target assembly device is used to perform training operations on a single-layer neural network each time to obtain a single-layer neural network after each training operation is completed, and output the single-layer neural network after each training as a logistics prediction model, wherein the number of times the single-layer neural network is trained is positively correlated with the number of city blocks served by the warehouse of the retail store;

[0028] Specifically, the target assembly device is used to perform training operations on the single-layer neural network to obtain the single-layer neural network after each training operation, and output the single-layer neural network after each training as a logistics prediction model, and the number of training times of the single-layer neural network is positively correlated with the number of city blocks served by the warehouse of the retail store, including: the more the number of city blocks served by the warehouse of the retail store, the more the number of training times of the single-layer neural network;

[0029] A prediction execution mechanism is connected to the configuration analysis device, the content acquisition device and the target assembly device respectively, and is used to use the logistics prediction model to intelligently predict the predicted value of the number of inbound pieces and the predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval according to the maximum number of pieces stored in the warehouse of the retail store, the duration of each time interval, the number of inbound pieces and the number of outbound pieces of the warehouse of the retail store corresponding to each time interval before the current time interval, and multiple configuration information of the warehouse of the retail store;

[0030] A data transmission mechanism, connected to the prediction execution mechanism, for wirelessly transmitting the predicted value of the number of inbound pieces and the predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval to a remote smart logistics service node;

[0031] Wherein, the logistics prediction model is used to intelligently predict the predicted value of the number of inbound items and the predicted value of the number of outbound items of the warehouse of the retail store in the current time interval according to the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store, including: synchronously inputting the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store into the logistics prediction model;

[0032] The single-layer neural network is trained for each time to obtain a single-layer neural network after each training operation, and the single-layer neural network after each training is output as a logistics prediction model, and the number of times the single-layer neural network is trained is positively correlated with the number of city blocks served by the warehouse of the retail store, including: using a numerical conversion function to represent the numerical conversion relationship of the positive correlation between the number of times the single-layer neural network is trained and the number of city blocks served by the warehouse of the retail store;

[0033] Among them, each training operation is performed on the single-layer neural network to obtain the single-layer neural network after each training operation, and the single-layer neural network after each training is output as a logistics prediction model, and the number of training times of the single-layer neural network is positively correlated with the number of city blocks served by the warehouse of the retail store, and also includes: in the numerical conversion function, the number of city blocks served by the warehouse of the retail store is an input parameter, and the number of training times of the single-layer neural network is an output parameter.

[0034] In the present invention, multiple configuration information can increase the number of street traffic and street consumption data. The consumption data described in the present invention can be consumption data generated by merchants in the applicant's system, or can be purchased external data. The data source can be selected according to demand, which can be understood by those skilled in the art.

[0035] Figure 2 Schematic diagram of the internal structure of a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment B of the present invention.

[0036] Compared with implementation scheme A, the retail store warehouse in-and-out item count time-sharing forecasting system shown in implementation scheme B may also include the following components:

[0037] A positioning server component is connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, respectively, and is used to provide the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device with their respective current instant positioning data;

[0038] Wherein, the positioning server is respectively connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, and is used to provide the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device with their respective current instant positioning data, including: the positioning server includes a plurality of positioning service units, which are respectively connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, so as to complete the respective supply of the current instant positioning data of the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device;

[0039] Wherein, the positioning server device includes a plurality of positioning service units, which are used to respectively connect with the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device to complete the respective supply of the current instant positioning data of the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, including: the plurality of positioning service units are a plurality of positioning sensors, which are used to respectively connect with the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device to complete the respective supply of the current instant positioning data of the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device;

[0040] Wherein, the plurality of positioning service units are a plurality of positioning sensors, which are used to respectively connect with the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, so as to respectively supply the current instant positioning data of the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, including: the structures of the plurality of positioning service units are the same;

[0041] Among them, the multiple positioning service units are multiple positioning sensors, which are used to connect with the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, so as to complete the separate supply of current and instant positioning data to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, and also include: the multiple positioning sensors have the same positioning range.

[0042] Figure 3 Schematic diagram of the internal structure of a time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store according to Embodiment C of the present invention.

[0043] Compared with implementation scheme A, the retail store warehouse in-and-out item count time-sharing forecasting system shown in implementation scheme C may also include the following components:

[0044] A user control interface is connected to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the multiple positioning service units of the target assembly device, respectively, and is used to synchronously control the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device;

[0045] Wherein, the user control interface is respectively connected to the multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, and is used to synchronously control the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, including: the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device is a sleep working mode or a running working mode;

[0046] And wherein, the user control interface is respectively connected to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the multiple positioning service units of the target assembly device, and is used to synchronously control the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, including: the multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device are all based on the Beidou positioning mechanism;

[0047] Alternatively, the prediction execution mechanism, the configuration analysis device, the content acquisition device and the multiple positioning service units of the target assembly device are all based on the Beidou positioning mechanism, including: the Galileo navigation positioning mechanism can also be used to replace the Beidou positioning mechanism.

[0048] In addition, in the time-sharing prediction system for the number of items entering and leaving the warehouse of the retail store, the logistics prediction model is used to intelligently predict the predicted value of the number of items entering and the predicted value of the number of items leaving the warehouse of the retail store in the current time interval based on the maximum number of items that can be stored in the warehouse of the retail store, the duration of each time interval, the number of items entering and the number of items leaving the warehouse of the retail store corresponding to each time interval before the current time interval, and multiple configuration information of the warehouse of the retail store, and also includes: running the logistics prediction model to obtain the predicted value of the number of items entering and the predicted value of the number of items leaving the warehouse of the retail store in the current time interval output by the logistics prediction model.

[0049] The retail store warehouse inbound and outbound piece count time-sharing prediction system of the present invention aims to solve the technical problem in the prior art that it is difficult to simultaneously predict the number of inbound and outbound pieces of a retail store warehouse in the current time period. By simultaneously predicting the number of inbound and outbound pieces of a retail store warehouse in the current time period, a customized logistics prediction model is designed, and multiple basic data are targeted for screening, thereby providing valuable reference information for the resource allocation strategy and space utilization of the retail store warehouse, thereby solving the above technical problem.

[0050] It is understood by those skilled in the art that the warehouse of the retail store described in the present invention may also be a forward warehouse of the retail store, which refers to a small storage center set up close to consumers. It can quickly respond to consumers' order needs and provide faster delivery services. This model combines the advantages of online and offline, aiming to improve shopping experience and efficiency. Forward warehouses usually store high-frequency consumer goods or hot-selling products for fast delivery. The advantages of forward warehouses include:

[0051] 1. Quick response to orders: The forward warehouse is located in a geographical location close to consumers, which can quickly process online or offline order requests, significantly shortening the time cycle from order placement to delivery. This immediacy is crucial to meeting the speed needs of modern consumers.

[0052] 2. Flexible inventory adjustment: By analyzing sales data and forecasting models, companies can dynamically allocate inventory between forward warehouses in different regions to ensure that hot-selling products are always in sufficient supply while avoiding the cost increase caused by excessive stockpiling. In addition, during special promotions, the inventory of specific products can be temporarily increased as needed.

[0053] 3. Support diversified service models: In addition to conventional door-to-door delivery, the forward warehouse can also provide diversified service options such as self-pickup points and instant delivery, giving customers more choices. Especially for time-sensitive products such as fresh food, this method can effectively ensure freshness.

[0054] The time-sharing prediction system for the number of items entering and leaving the warehouse of the retail store of the present invention is also an important link in the digital operation and supply chain construction of the Bailian business operating system. In addition to connecting to multiple third-party Internet sales platforms, such as Meituan, Ele.me, JD.com, Taoxianda, Douyin, etc., the actual application scenarios can also include the online and offline linkage of its own department store format and supermarket format. For the department store format, it supports the collection of data based on the passenger flow data and consumption data collection industry of the business district, evaluates the subsequent user consumption trend, and prepares goods in advance to cope with the changes brought about by holidays and seasonal changes; in the department store industry with limited space, the display of goods in the store and the sales circulation of inventory goods have a vital impact on the overall sales performance; the present invention can also help merchants break the offline flow restrictions of stores, such as stores that are not popular shops on the first floor or along the street in the business district, as long as they are connected to online shops, accurate drainage, and sales are realized in the form of forward warehouses. The time-sharing prediction system of the present invention will also prepare beauty, clothing and other department store products that may be sold in the future in advance in the corresponding business district, and quickly deliver them to consumers in the form of online cloud stores. In the supermarket industry, when holidays or promotional activities are approaching, the system can accurately predict inventory changes during the upcoming peak demand period based on historical sales data and current trends. In this way, supermarkets can not only ensure that there are sufficient goods on the shelves, but also avoid capital occupation and increased storage costs caused by excessive stockpiling. In addition, the data transmission mechanism of the present invention transmits the prediction results to the relevant management system in real time, which can help supermarkets respond quickly to market changes and optimize resource allocation strategies, thereby improving overall operational efficiency.

[0055] While the present invention has been described in conjunction with the specific exemplary embodiments, it should be understood that the invention is not limited to the disclosed embodiments, but on the contrary, the invention is intended to cover various modifications and equivalent structures included within the spirit and scope of the appended claims and their equivalents.

Claims

1. A time-sharing forecasting system for the number of items entering and leaving a warehouse of a retail store, characterized in that: The system comprises: A configuration parsing device is used to obtain multiple configuration information of a warehouse of a retail store, wherein the multiple configuration information of the warehouse of the retail store is the number of city blocks served by the warehouse of the retail store, the number of permanent urban population, and the total geographical area occupied by each city block served by the warehouse of the retail store; A content collection device is used to obtain the number of incoming items and the number of outgoing items of the warehouse of the retail store corresponding to each time interval before the current time interval; The target assembly device is used to perform training operations on a single-layer neural network for each time to obtain a single-layer neural network after each training operation is completed, and output the single-layer neural network after each training as a logistics prediction model, wherein the number of times the single-layer neural network is trained is positively correlated with the number of city blocks served by the warehouse of the retail store, including: using a numerical conversion function to represent a numerical conversion relationship that the number of times the single-layer neural network is trained is positively correlated with the number of city blocks served by the warehouse of the retail store; A prediction execution mechanism is connected to the configuration analysis device, the content acquisition device and the target assembly device respectively, and is used to adopt the logistics prediction model to intelligently predict the predicted value of the number of inbound items and the predicted value of the number of outbound items of the warehouse of the retail store in the current time interval according to the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store, including: synchronously inputting the maximum number of items stored in the warehouse of the retail store, the duration of each time interval, the number of inbound items and the number of outbound items corresponding to each time interval of the warehouse of the retail store before the current time interval, and multiple configuration information of the warehouse of the retail store into the logistics prediction model; Running the logistics prediction model to obtain a predicted value of the number of inbound pieces and a predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval output by the logistics prediction model; A data transmission mechanism, connected to the prediction execution mechanism, for wirelessly transmitting the predicted value of the number of inbound pieces and the predicted value of the number of outbound pieces of the warehouse of the retail store in the current time interval to a remote smart logistics service node; In the numerical conversion function, the number of city blocks served by the warehouse of the retail store is an input parameter, and the number of times the single-layer neural network is trained is an output parameter; The system also includes a positioning server component; The positioning server is respectively connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, and is used to provide the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device with their respective current instant positioning data; Among them, the positioning server is respectively connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device, and is used to provide the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device with their respective current and real-time positioning data, including: the positioning server includes multiple positioning service units, which are respectively connected to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device to complete the respective supply of current and real-time positioning data to the prediction execution mechanism, the configuration analysis device, the content collection device and the target assembly device.

2. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 1, characterized in that: The positioning server device includes multiple positioning service units, which are used to connect with the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, so as to complete the separate supply of current and real-time positioning data of each of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device. The multiple positioning service units are multiple positioning sensors, which are used to connect with the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, so as to complete the separate supply of current and real-time positioning data of each of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device.

3. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 2, characterized in that: The multiple positioning service units are multiple positioning sensors, which are used to connect to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, so as to complete the respective supply of current instant positioning data to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, including: the structures of the multiple positioning service units are the same.

4. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 3, characterized in that: The multiple positioning service units are multiple positioning sensors, which are used to connect to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, so as to complete the separate supply of current instant positioning data to the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device respectively, and also include: the multiple positioning sensors have the same positioning range.

5. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 1, characterized in that: The system further comprises: The user control interface is respectively connected to the multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, and is used to synchronously control the current working mode of the respective positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device.

6. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 5, characterized in that: The user control interface is respectively connected to multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, and is used to synchronously control the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, including: the current working mode of the positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device is a sleep working mode or a running working mode.

7. The system for predicting the number of items entering and leaving a warehouse of a retail store as claimed in claim 5, characterized in that: A user control interface is respectively connected to the multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, and is used to synchronously control the current working mode of the respective positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device, including: the multiple positioning service units of the prediction execution mechanism, the configuration analysis device, the content acquisition device and the target assembly device are all based on the Beidou positioning mechanism.

Images