Hardware building material warehouse management system and method
Through deep learning technology, analyzing the inventory data of hardware accessories, generating a dynamic inventory change feature map, realizing intelligent early warning of the hardware building materials warehousing management system, solving the problem of insufficient inventory in traditional management, and improving the accuracy and timeliness of inventory management.
Patent Information
- Application Number
- CN202411156237.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional hardware and building materials warehousing management relies on manual inventory and simple inventory recording systems, and cannot monitor inventory changes in time, resulting in insufficient inventory, resulting in supply chain disruptions and decreased customer satisfaction.
Using intelligent data analysis technology based on deep learning, we generate a dynamic inventory change feature map to achieve future inventory intelligent early warning by analyzing the procurement and inventory inventories, sales return and exchange of hardware accessories, sales and out-of-stock quantity.
It significantly improves the accuracy and timeliness of hardware accessories inventory management, helps enterprises replenish inventory in a timely manner, and avoids supply chain disruptions and declining customer satisfaction.
Smart Images

Figure CN118966974A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent warehouse management, and more specifically, to a hardware and building materials warehouse management system and method. Background Art
[0002] For enterprises, effectively managing items in the warehouse can not only improve work efficiency, but also save costs. Warehouses in the hardware industry usually store a large number of hardware accessories, which are of various types and large quantities. In traditional hardware and building materials storage management, inventory is often monitored by manual regular inventory or a simple inventory record system, which will lead to the inability to timely discover future inventory shortages. Insufficient inventory in the future is prone to failure to supply goods in time, which can easily lead to a decline in customer satisfaction, thereby affecting the cooperative relationship with customers.
[0003] Therefore, an optimized hardware and building materials storage management solution is needed. Summary of the invention
[0004] In view of the shortcomings of the prior art, this application provides a hardware and building materials warehouse management system and method, which adopts intelligent data analysis technology based on deep learning, and realizes intelligent early warning of future inventory of hardware accessories by analyzing the purchase and warehousing quantity, sales return and exchange quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different hardware accessories in the warehouse every day. In this way, the accuracy and timeliness of hardware accessories inventory management are significantly improved, helping enterprises to understand the situation in time before the inventory is insufficient, so as to replenish the inventory, effectively solving the problems of possible supply chain disruption and reduced customer satisfaction.
[0005] According to one aspect of the present application, a hardware and building materials storage management system is provided, which includes: The hardware accessories related warehouse data collection module is used to obtain the daily purchase quantity of different types of hardware accessories in the warehouse, the quantity of incoming goods for sales returns and exchanges, the quantity of outgoing goods for sales, the quantity of outgoing goods for sales returns and exchanges, and the quantity of existing inventory; The hardware accessories related warehouse data structuring module is used to arrange the daily purchase inbound quantity, sales return inbound quantity, sales outbound quantity, sales return outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension into a hardware accessories inventory related indicator time series input tensor; A hardware accessories related storage data feature encoding module, used for performing feature encoding on the hardware accessories inventory related index time series input tensor to obtain a hardware accessories inventory time series dynamic change feature graph; A hardware accessories related storage data feature emphasis module is used to highlight the characteristics of the hardware accessories inventory time series dynamic change feature graph to obtain a hardware accessories type inventory time series dynamic change emphasized feature graph; The hardware accessories inventory analysis result generation module is used to obtain the results of the hardware accessories types that need to increase the inventory based on the time series dynamic change of the hardware accessories inventory type and emphasize the characteristic diagram.
[0006] According to another aspect of the present application, a method for storage management of hardware and building materials is provided, which includes: Obtain the daily purchase quantity of different types of hardware accessories in the warehouse, the quantity of incoming sales returns and exchanges, the quantity of outgoing sales, the quantity of outgoing sales returns and exchanges, and the quantity of existing inventory; Arrange the daily purchase inbound quantity, sales return inbound quantity, sales outbound quantity, sales return outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension into a hardware accessories inventory-related indicator time series input tensor; Performing feature encoding on the hardware accessories inventory related index time series input tensor to obtain a hardware accessories inventory time series dynamic change feature graph; Performing feature emphasis on the hardware accessories inventory time series dynamic change feature graph to obtain a hardware accessories inventory time series dynamic change emphasized feature graph; Based on the hardware accessories inventory time series dynamic change emphasized feature diagram, the hardware accessories types that need to increase inventory are obtained.
[0007] This application has significant technical effects due to the adoption of the above technical solutions: The hardware and building materials storage management system and method provided by the present application adopts intelligent data analysis technology based on deep learning, and realizes intelligent early warning of future inventory of hardware accessories by analyzing the purchase and warehousing quantity, sales return and exchange quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different hardware accessories in the warehouse every day. In this way, the accuracy and timeliness of hardware accessories inventory management are significantly improved, helping enterprises to understand the situation in time before the inventory is insufficient, so as to replenish the inventory, and effectively solve the problems of possible supply chain disruption and reduced customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0009] Figure 1 It is a flow chart of the hardware and building materials storage management method according to an embodiment of the present application.
[0010] Figure 2 It is a schematic diagram of the architecture of the hardware and building materials storage management method according to an embodiment of the present application.
[0011] Figure 3 In the hardware and building materials warehouse management method according to an embodiment of the present application, the characteristic diagram of the dynamic change of the inventory time series of hardware accessories is highlighted to obtain a flowchart of the emphasized characteristic diagram of the dynamic change of the inventory time series of hardware accessories types.
[0012] Figure 4 In the hardware and building materials warehouse management method according to an embodiment of the present application, a flowchart is provided for obtaining the results of hardware accessories types that need to increase inventory based on the hardware accessories type inventory time series dynamic change emphasis feature diagram.
[0013] Figure 5 This is a system block diagram of a hardware and building materials storage management system according to an embodiment of the present application. DETAILED DESCRIPTION
[0014] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.
[0015] As mentioned in the above background technology, the importance of warehouse management in the hardware industry is self-evident. Effective warehouse management can not only significantly improve work efficiency, but also greatly reduce operating costs. Warehouses in the hardware industry usually store a large number of accessories and materials. These items are of various types and large in quantity, which makes management quite challenging. However, traditional warehouse management often relies on manual inventory and basic inventory record systems. This method has obvious limitations: it cannot monitor inventory status in real time, making it difficult for companies to respond to inventory changes in a timely manner. When inventory is insufficient, companies often cannot quickly detect and take action, which will not only delay delivery, but may also damage customer satisfaction, and thus affect the company's market reputation and customer relationships. Therefore, an optimized hardware and building materials warehouse management solution is expected.
[0016] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for the storage management of hardware and building materials.
[0017] Figure 1It is a flow chart of the hardware and building materials storage management method according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the structure of the hardware and building materials storage management method according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the hardware and building materials warehousing management method according to the embodiment of the present application includes: S110, obtaining the daily purchase warehousing quantity, sales return and exchange warehousing quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse; S120, arranging the daily purchase warehousing quantity, sales return and exchange warehousing quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension as a hardware accessories inventory related indicator time series input tensor; S130, feature encoding the hardware accessories inventory related indicator time series input tensor to obtain a hardware accessories inventory time series dynamic change feature graph; S140, feature emphasizing the hardware accessories inventory time series dynamic change feature graph to obtain a hardware accessories type inventory time series dynamic change emphasized feature graph; S150, based on the hardware accessories type inventory time series dynamic change emphasized feature graph, obtain the hardware accessories type result that needs to increase inventory.
[0018] Specifically, in the technical solution of the present application, it is first necessary to obtain the daily purchase warehousing quantity, sales return and exchange warehousing quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse. It should be understood that the purchase warehousing quantity, sales outbound quantity, sales return and exchange warehousing and outbound quantity, etc., together constitute the dynamic change process of inventory. These data can fully reflect the inflow and outflow of hardware accessories and are the cornerstone of warehouse management. By continuously collecting daily data of different hardware accessories in the warehouse, it is easier to identify the trend and pattern of changes in hardware accessories inventory and thus predict the demand for hardware accessories in the future. This prediction ability is the key to determining the need to increase the inventory of hardware accessories. That is, by analyzing the inventory data of different types of hardware accessories in the warehouse, it is possible to identify which hardware accessories are rapidly declining in inventory level, and the information based on the rapid decline in inventory is the basis for judging the need to increase the inventory of hardware accessories. In particular, in one implementable embodiment of the present application, automatic identification technologies such as barcodes and RFID can be used to realize automated warehousing and outbound management of hardware accessories, and relevant data can be uploaded to the inventory management system, thereby obtaining relevant inventory data and analyzing it, thereby realizing accessory inventory early warning.
[0019] Hardware and building materials warehouses usually contain many different types of accessories. In order to facilitate the processing of data of various accessories, and because the inventory of different hardware accessories changes over time, in order to clearly show the increase and decrease of each type of accessory inventory, thereby providing a basis for inventory forecasting and strategy formulation, in the technical solution of this application, the daily purchase and warehousing quantity, sales return and exchange warehousing quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse are arranged according to the time dimension and the hardware accessories sample dimension as a hardware accessories inventory-related indicator time series input tensor. The spatial dimension of the hardware accessories inventory-related indicator time series input tensor formed by the arrangement respectively expresses different hardware accessories and different inventory indicators of hardware accessories, and the channel dimension represents the hardware accessories inventory data at different times.
[0020] The inventory data of hardware accessories has obvious time series characteristics, that is, the inventory quantity will change over time. In order to effectively capture this time series feature and extract the time dependence and change trend in the hardware accessories inventory data, it is necessary to pass the hardware accessories inventory-related indicator time series input tensor through the inventory change encoder based on the three-dimensional convolution kernel to obtain the hardware accessories inventory time series dynamic change feature map. Through the inventory change encoder based on the three-dimensional convolution kernel, the hardware accessories inventory-related indicator time series input tensor can be converted into a dynamic change feature map. This feature map can intuitively show the trend and pattern of inventory data over time, and provide strong data support for subsequent inventory forecasts. The essence of the inventory change encoder based on the three-dimensional convolution kernel in this application is a convolutional neural network model using a three-dimensional convolution kernel. Compared with the traditional convolutional neural network using a two-dimensional convolution kernel, the three-dimensional convolution kernel can perform convolution operations on the three dimensions of time, hardware accessories type and inventory index. This feature extraction method can capture the complex patterns and structures in the original arranged inventory data, thereby providing a rich feature representation and realizing efficient feature extraction of inventory data. It is worth mentioning that the inventory change encoder based on the three-dimensional convolution kernel can also be flexibly adjusted according to actual needs. It can adapt to hardware accessories inventory data of different types and time scales by changing parameters such as the size, step size and number of the convolution kernel.
[0021] In an embodiment of the present application, a possible implementation method of passing the hardware accessories inventory-related indicator time series input tensor through an inventory change encoder based on a three-dimensional convolution kernel to obtain a hardware accessories inventory time series dynamic change feature map can be: using each layer of the inventory change encoder to perform the following on the input data in the forward pass of the layer: using the convolution units of each layer of the inventory change encoder to perform convolution processing on the input data based on a three-dimensional convolution kernel to obtain a convolution feature map; using the pooling units of each layer of the inventory change encoder to perform pooling processing on the convolution feature map based on a local feature matrix to obtain a pooling feature map; using the activation units of each layer of the inventory change encoder to perform nonlinear activation on the feature values of each position in the pooling feature map to obtain an activation feature map; wherein the output of the last layer of the convolutional neural network model is the hardware accessories inventory time series dynamic change feature map.
[0022] Figure 3 In the hardware and building materials storage management method according to the embodiment of the present application, the characteristic diagram of the dynamic change of the inventory time series of hardware accessories is highlighted to obtain a flowchart of the dynamic change of the inventory time series of hardware accessories. Figure 3 As shown, the feature emphasis is performed on the hardware accessories inventory time series dynamic change characteristic graph to obtain the hardware accessories type inventory time series dynamic change emphasized characteristic graph, including: S141, the feature emphasis is performed on the time dimension of the hardware accessories inventory time series dynamic change characteristic graph to obtain the hardware accessories inventory time series dynamic change emphasized characteristic graph; S142, the feature emphasis is performed on the hardware accessories sample dimension of the hardware accessories inventory time series dynamic change emphasized characteristic graph to obtain the hardware accessories type inventory time series dynamic change emphasized characteristic graph.
[0023] Next, the hardware accessories inventory time series dynamic change feature map is passed through the inventory change time series feature emphasizer based on the channel attention mechanism to obtain the hardware accessories inventory time series dynamic change emphasized feature map. The inventory change time series feature emphasizer based on the channel attention mechanism in this application is a convolutional neural network model using the channel attention mechanism. Those of ordinary skill in the art should understand that the channel attention mechanism allows the convolutional neural network model to automatically learn the weights of different channels on the channel dimension of the feature map, so that the model can pay more attention to those features that are more important to the current task. In this application, the channel dimension refers to the time dimension, that is, the model can identify which time points of inventory changes are more important for subsequent analysis, and thus pay more attention to these time points. By emphasizing the inventory change characteristics at key time points, the model can more accurately capture the trends and patterns of inventory changes, thereby improving the accuracy of inventory forecasts.
[0024] In an embodiment of the present application, a possible implementation method of passing the hardware accessories inventory time series dynamic change feature map through an inventory change time series feature emphasizer based on a channel attention mechanism to obtain a hardware accessories inventory time series dynamic change emphasized feature map can be: each layer of the inventory change time series feature emphasizer based on a channel attention mechanism performs the following on the input data in the forward transfer of the layer: convolution processing on the input data based on a two-dimensional convolution kernel to generate a convolution feature map; pooling processing on the convolution feature map to generate a pooling feature map; activation processing on the pooling feature map to generate an activation feature map; calculating the quotient of the eigenvalue mean of the feature matrix corresponding to each channel in the activation feature map and the sum of the eigenvalue means of the feature matrices corresponding to all channels as the weighting coefficient of the feature matrix corresponding to each channel; weighting the feature matrix of each channel with the weighting coefficient of each channel in the activation feature map to generate a channel attention feature map; wherein the input of the first layer of the inventory change time series feature emphasizer is the hardware accessories inventory time series dynamic change feature map, and the output of the last layer of the inventory change time series feature emphasizer is the hardware accessories inventory time series dynamic change emphasized feature map.
[0025] Next, the hardware accessories inventory time series dynamic change emphasis feature map is passed through the accessory type feature emphasizer based on the spatial attention mechanism to obtain the hardware accessories type inventory time series dynamic change emphasis feature map. The accessory type feature emphasizer based on the spatial attention mechanism in this application is a convolutional neural network model using the spatial attention mechanism. The space in this application refers to the two-dimensional space formed by different inventory accessory types and inventory indicators. The spatial attention mechanism allows the convolutional neural network model to automatically learn the weights of different accessory types and inventory indicators on the hardware accessories inventory time series dynamic change emphasis feature map, which means that the model can identify which accessory types and inventory indicators are more important for analyzing the types of hardware accessories that need to increase inventory, thereby paying more attention to these accessory types and indicators. Through the processing of the accessory type feature emphasizer, we can focus on those accessory types and inventory indicators that have a greater impact. This targeted feature extraction helps to more accurately capture the inventory change rules and trends of different accessory types, thereby formulating a more reasonable inventory management strategy.
[0026] Figure 4 In the hardware and building materials storage management method according to the embodiment of the present application, based on the hardware accessories inventory time series dynamic change emphasis feature diagram, a flow chart of the hardware accessories types that need to increase inventory is obtained. Figure 4As shown, the result of the types of hardware accessories that need to increase inventory based on the hardware accessories inventory time series dynamic change emphasis feature graph is obtained, including: S151, performing feature distribution deviation back-propagation correction on the hardware accessories inventory time series dynamic change emphasis feature graph to obtain an optimized hardware accessories inventory time series dynamic change emphasis feature graph; S152, passing the optimized hardware accessories inventory time series dynamic change emphasis feature graph through an accessory type classifier to obtain a classification result, and the classification result is used to represent the types of hardware accessories that need to increase inventory.
[0027] In particular, in the technical solution of the present application, an accessory category feature emphasizer based on a spatial attention mechanism is used to emphasize important category features. In the feature emphasis process, the features of certain types of hardware accessories may be over-emphasized, resulting in some parts of the feature map being too prominent, thereby generating a shift deviation in the feature distribution. The feature emphasizer may not balance the feature importance of different types of hardware accessories well, resulting in some types of features being over-amplified, while other types of features are relatively ignored. If the hardware accessories category inventory time series dynamic change emphasize feature map produces an extreme shift in the emphasis process, it may cause the classifier to over-rely on these features during the training process, thereby generating an extreme shift in classification judgment. That is, the hardware accessories category inventory time series dynamic change emphasize feature map has a feature distribution shift deviation relative to the hardware accessories inventory time series dynamic change emphasize feature map, which causes the hardware accessories category inventory time series dynamic change emphasize feature map to have an extreme category shift when the classification judgment is performed by the classifier, resulting in a decrease in the accuracy of the classification result. In order to solve this problem, in the technical solution of the present application, the characteristic distribution deviation of the hardware accessories inventory time series dynamic change emphasis feature map is back-propagated to obtain an optimized hardware accessories inventory time series dynamic change emphasis feature map.
[0028] Specifically, in an embodiment of the present application, the feature distribution deviation of the hardware accessories inventory time series dynamic change emphasis feature map is corrected by back-propagation to obtain an optimized hardware accessories inventory time series dynamic change emphasis feature map, including: calculating the shift difference representation map between the hardware accessories inventory time series dynamic change emphasis feature map and the hardware accessories inventory time series dynamic change emphasis feature map; probabilizing each eigenvalue of the shift difference representation map based on a probability activation function to obtain a probabilistic shift difference representation map; inputting the hardware accessories inventory time series dynamic change emphasis feature map into the accessory type classifier to obtain an initial classification probability value; based on the The backpropagation compensation representation diagram is determined by comparing the characteristic values of each position of the probabilistic shifted difference representation diagram with the initial classification probability value and the weight characteristic diagram of the accessory type classifier; the exponential function value of each characteristic value of the backpropagation compensation representation diagram is calculated to obtain the backpropagation compensation class representation diagram; the positional dot multiplication between the backpropagation compensation class representation diagram and the hardware accessory type inventory time series dynamic change emphasis feature diagram is calculated to obtain the dot multiplication result diagram, and the weighted sum between the dot multiplication result diagram and the hardware accessory type inventory time series dynamic change emphasis feature diagram is calculated to obtain the optimized hardware accessory type inventory time series dynamic change emphasis feature diagram.
[0029] In an embodiment of the present application, another expression method of performing feature distribution deviation back propagation correction on the hardware accessories inventory time series dynamic change emphasis feature map to obtain an optimized hardware accessories inventory time series dynamic change emphasis feature map can be: processing the hardware accessories inventory time series dynamic change emphasis feature map with the following formula to obtain the optimized hardware accessories inventory time series dynamic change emphasis feature map; wherein, the formula is:
[0030] in, represents the probability activation function, The hardware accessories inventory time series dynamic change emphasizes the characteristic diagram, The characteristic diagram showing the dynamic changes in the inventory of the hardware accessories type is emphasized. represents the probabilistic shifted difference representation graph, The probabilistic shift difference representation is shown in FIG The eigenvalues of the positions, represents the symbolic function, and denote the first weighted hyperparameter and the second weighted hyperparameter respectively, represents the weight feature map of the accessory type classifier, represents the initial classification probability value, It means point multiplication by position. represents matrix multiplication, The diagram represents the emphasized feature of the dynamic changes in the inventory time series of the optimized hardware accessories.
[0031] That is, in view of the distribution shift deviation problem between feature graphs, the feature deviation information between the hardware accessories inventory time series dynamic change emphasis feature graph and the hardware accessories type inventory time series dynamic change emphasis feature graph is used for feature distribution deviation back-propagation correction. The difference between the hardware accessories inventory time series dynamic change emphasis feature graph and the hardware accessories type inventory time series dynamic change emphasis feature graph is calculated to obtain a shift difference representation graph, so as to quantify the deviation of feature distribution. The shift difference representation graph reveals the alignment of the two feature graphs in the feature space, providing a basis for subsequent compensation. Then, the probability activation function is applied to each feature value of the shift difference representation graph for probabilistic processing, so that the difference value is converted into a probability distribution, so that the model can understand the deviation between features in a probabilistic form. Then, the original hardware accessories type inventory time series dynamic change emphasis feature graph is input into the accessory type classifier to obtain an initial classification probability value, wherein the initial classification probability value provides the model with a basic classification performance indicator before feature compensation. Next, the back propagation compensation representation is determined by combining the probabilistic shifted differential representation and the initial classification probability value, as well as the weighted feature map of the accessory category classifier. That is, the feature map is compensated using the model's own weight and bias information to reduce the classification error. Next, the exponential function value is calculated for each eigenvalue of the back propagation compensation representation, thereby enhancing the expression range of the eigenvalue and providing a mechanism for enlarging or reducing the feature compensation. Finally, the dot product result between the back propagation compensation class representation and the hardware accessories category inventory time series dynamic change emphasized feature map is calculated, and then the weighted sum is calculated with the original hardware accessories category inventory time series dynamic change emphasized feature map to obtain the optimized hardware accessories category inventory time series dynamic change emphasized feature map. It should be understood that the dot product and weighted sum operations combine the compensation information and the original feature information to generate an adjusted feature map, which aims to reduce the extreme deviation of the category and improve the accuracy of classification. The entire data processing process achieves the goal of optimizing the feature representation from identifying the feature distribution shift deviation to compensating for these deviations. This process not only improves the adaptability of the accessory type classifier to changes in feature distribution, but also enhances the model's ability to distinguish different categories. Through the feature distribution deviation backpropagation correction mechanism, the model can more accurately capture the subtle differences between feature maps, thereby achieving higher accuracy and better generalization performance in classification tasks.
[0032] Finally, the optimized hardware accessories inventory time series dynamic change emphasis feature map is passed through the accessory type classifier to obtain a classification result, and the classification result is used to indicate the hardware accessories that need to increase inventory. It should be understood that the optimized hardware accessories inventory time series dynamic change emphasis feature map contains the time series dynamic change information such as inventory, sales, demand forecast, etc. of different hardware accessories over a period of time, and this information has been weighted and emphasized by the model according to its importance to inventory changes. Although the feature map already contains rich information, it is still a relatively primitive data representation form and is difficult to be directly used to guide the inventory management decision of the enterprise. The role of the accessory type classifier is to convert this complex information into a concise and clear classification result, that is, which hardware accessories need to increase inventory. The accessory type classifier is a machine learning model that can analyze and judge based on the input data and map it to different categories. The classification result directly points out which hardware accessories may face shortage risks in the future, so it is an important basis for enterprises to make inventory management decisions. Enterprises can adjust procurement plans, optimize inventory structures, and improve inventory turnover rates based on the classification results, thereby ensuring the smooth progress of production and sales.
[0033] In summary, the hardware and building materials storage management method based on the embodiment of the present application is explained, which adopts intelligent data analysis technology based on deep learning, and realizes intelligent early warning of future inventory of hardware accessories by analyzing the purchase warehousing quantity, sales return and exchange warehousing quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different hardware accessories in the warehouse every day. In this way, the accuracy and timeliness of hardware accessories inventory management are significantly improved, helping enterprises to understand the situation in time before the inventory is insufficient, so as to replenish the inventory, and effectively solve the problems of possible supply chain disruptions and reduced customer satisfaction.
[0034] Figure 5 FIG. 1 is a system block diagram of a hardware and building materials storage management system according to an embodiment of the present application. Figure 5As shown, according to the hardware and building materials warehouse management system 100 of the embodiment of the present application, it includes: a hardware accessories related warehouse data acquisition module 110, which is used to obtain the daily purchase warehouse quantity, sales return and exchange warehouse quantity, sales warehouse quantity, sales return and exchange warehouse quantity, and existing inventory quantity of different types of hardware accessories in the warehouse; a hardware accessories related warehouse data structuring module 120, which is used to arrange the daily purchase warehouse quantity, sales return and exchange warehouse quantity, sales warehouse quantity, sales return and exchange warehouse quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension as hardware accessories inventory Relevant indicator time series input tensor; hardware accessories related storage data feature encoding module 130, used for feature encoding the hardware accessories inventory related indicator time series input tensor to obtain a hardware accessories inventory time series dynamic change feature diagram; hardware accessories related storage data feature emphasis module 140, used for feature emphasis on the hardware accessories inventory time series dynamic change feature diagram to obtain a hardware accessories type inventory time series dynamic change emphasis feature diagram; hardware accessories inventory analysis result generation module 150, used to obtain the hardware accessories type results that need to increase inventory based on the hardware accessories type inventory time series dynamic change emphasis feature diagram.
[0035] Here, those skilled in the art will appreciate that the specific functions and operations of the various units and modules in the hardware and building materials storage management system 100 have been described in detail above. Figures 1 to 4 The description of the hardware and building materials warehouse management method has been introduced in detail, and therefore, its repeated description will be omitted.
[0036] In summary, the hardware and building materials warehouse management system 100 based on the embodiment of the present application is explained, which adopts intelligent data analysis technology based on deep learning, and realizes intelligent early warning of future inventory of hardware accessories by analyzing the purchase and warehousing quantity, sales return and exchange quantity, sales outbound quantity, sales return and exchange outbound quantity, and existing inventory quantity of different hardware accessories in the warehouse every day. In this way, the accuracy and timeliness of hardware accessories inventory management are significantly improved, helping enterprises to understand the situation in time before the inventory is insufficient, so as to replenish the inventory, and effectively solve the problems of possible supply chain disruptions and reduced customer satisfaction.
[0037] As described above, the hardware and building materials warehouse management system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for hardware and building materials warehouse management. In one example, the hardware and building materials warehouse management system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the hardware and building materials warehouse management system 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the hardware and building materials warehouse management system 100 can also be one of the many hardware modules of the wireless terminal.
[0038] Alternatively, in another example, the hardware and building materials warehouse management system 100 and the wireless terminal may also be separate devices, and the hardware and building materials warehouse management system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.
[0039] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details of the above application are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present application to the use of the above specific details to be implemented.
[0040] In this application, words such as "including", "comprising", "having", etc. are open words, meaning "including but not limited to", and can be used interchangeably therewith. The words "or" and "and" used herein mean the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably therewith.
[0041] It should also be noted that in the system and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.
[0042] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features of the present application.
[0043] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. A method for storage management of hardware and building materials, characterized in that: include: Obtain the daily purchase quantity of different types of hardware accessories in the warehouse, the quantity of incoming sales returns and exchanges, the quantity of outgoing sales, the quantity of outgoing sales returns and exchanges, and the quantity of existing inventory; Arrange the daily purchase inbound quantity, sales return inbound quantity, sales outbound quantity, sales return outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension into a hardware accessories inventory-related indicator time series input tensor; Performing feature encoding on the hardware accessories inventory related index time series input tensor to obtain a hardware accessories inventory time series dynamic change feature graph; Performing feature emphasis on the hardware accessories inventory time series dynamic change feature graph to obtain a hardware accessories inventory time series dynamic change emphasized feature graph; Based on the hardware accessories inventory time series dynamic change emphasized feature diagram, the hardware accessories types that need to increase inventory are obtained.
2. The method for storage management of hardware and building materials according to claim 1, characterized in that: The hardware accessories inventory-related indicator time series input tensor is feature encoded to obtain a hardware accessories inventory time series dynamic change feature map, including: the hardware accessories inventory-related indicator time series input tensor is passed through an inventory change encoder based on a three-dimensional convolution kernel to obtain the hardware accessories inventory time series dynamic change feature map.
3. The method for storage management of hardware and building materials according to claim 2 is characterized in that: The hardware accessories inventory time series dynamic change feature graph is subjected to feature emphasis to obtain a hardware accessories inventory time series dynamic change emphasized feature graph, including: Emphasizing the characteristics of the hardware accessories inventory time series dynamic change characteristic graph in the time dimension to obtain a hardware accessories inventory time series dynamic change emphasized characteristic graph; The hardware accessories inventory time series dynamic change emphasis feature graph is subjected to feature emphasis of the hardware accessories sample dimension to obtain the hardware accessories type inventory time series dynamic change emphasis feature graph.
4. The method for storage management of hardware and building materials according to claim 3 is characterized in that: The feature of the hardware accessories inventory time series dynamic change feature graph is emphasized in the time dimension to obtain the hardware accessories inventory time series dynamic change emphasized feature graph, including: passing the hardware accessories inventory time series dynamic change feature graph through an inventory change time series feature emphasizer based on a channel attention mechanism to obtain the hardware accessories inventory time series dynamic change emphasized feature graph.
5. The method for storage management of hardware and building materials according to claim 4 is characterized in that: The hardware accessories inventory time series dynamic change emphasis feature map is subjected to feature emphasis on the hardware accessories sample dimension to obtain the hardware accessories type inventory time series dynamic change emphasis feature map, including: the hardware accessories inventory time series dynamic change emphasis feature map is passed through an accessory type feature emphasizer based on a spatial attention mechanism to obtain the hardware accessories type inventory time series dynamic change emphasis feature map.
6. The method for storage management of hardware and building materials according to claim 5, characterized in that: The inventory change encoder based on three-dimensional convolution kernel is a convolutional neural network model using three-dimensional convolution kernel, the inventory change timing feature emphasizer based on channel attention mechanism is a convolutional neural network model using channel attention mechanism, and the accessory type feature emphasizer based on spatial attention mechanism is a convolutional neural network model using spatial attention mechanism.
7. The method for storage management of hardware and building materials according to claim 6, characterized in that: Based on the hardware accessories inventory time series dynamic change emphasized feature diagram, the hardware accessories types that need to increase inventory are obtained, including: Performing feature distribution deviation back propagation correction on the hardware accessories inventory time series dynamic change emphasized feature graph to obtain an optimized hardware accessories inventory time series dynamic change emphasized feature graph; The optimized hardware accessories inventory time series dynamic change emphasized feature diagram is passed through an accessories category classifier to obtain a classification result, and the classification result is used to indicate the types of hardware accessories that need to increase inventory.
8. The method for storage management of hardware and building materials according to claim 7, characterized in that: The hardware accessories inventory time series dynamic change emphasized feature map is corrected by feature distribution deviation back propagation to obtain an optimized hardware accessories inventory time series dynamic change emphasized feature map, including: Calculate a shift difference representation diagram between the hardware accessories inventory time series dynamic change emphasized feature diagram and the hardware accessories type inventory time series dynamic change emphasized feature diagram; Probabilize each eigenvalue of the shifted difference representation map based on a probability activation function to obtain a probabilized shifted difference representation map; Input the hardware accessories inventory time series dynamic change emphasized feature graph into the accessories category classifier to obtain an initial classification probability value; Determining a back-propagation compensation representation map based on a comparison between the feature values of each position of the probabilistic shifted difference representation map and the initial classification probability value and a weight feature map of the accessory type classifier; Calculating the exponential function value of each eigenvalue of the back-propagation compensation representation graph to obtain a back-propagation compensation class representation graph; Calculate the positional point multiplication between the back-propagation compensation class representation graph and the hardware accessories category inventory time series dynamic change emphasis feature graph to obtain a point multiplication result graph, and calculate the weighted sum between the point multiplication result graph and the hardware accessories category inventory time series dynamic change emphasis feature graph to obtain the optimized hardware accessories category inventory time series dynamic change emphasis feature graph.
9. A hardware and building materials storage management system, characterized in that: include: The hardware accessories related warehouse data collection module is used to obtain the daily purchase quantity of different types of hardware accessories in the warehouse, the quantity of incoming goods for sales returns and exchanges, the quantity of outgoing goods for sales, the quantity of outgoing goods for sales returns and exchanges, and the quantity of existing inventory; The hardware accessories related warehouse data structuring module is used to arrange the daily purchase inbound quantity, sales return inbound quantity, sales outbound quantity, sales return outbound quantity, and existing inventory quantity of different types of hardware accessories in the warehouse according to the time dimension and the hardware accessories sample dimension into a hardware accessories inventory related indicator time series input tensor; A hardware accessories related storage data feature encoding module, used for performing feature encoding on the hardware accessories inventory related index time series input tensor to obtain a hardware accessories inventory time series dynamic change feature graph; A hardware accessories related storage data feature emphasis module is used to highlight the characteristics of the hardware accessories inventory time series dynamic change feature graph to obtain a hardware accessories type inventory time series dynamic change emphasized feature graph; The hardware accessories inventory analysis result generation module is used to obtain the results of the hardware accessories types that need to increase the inventory based on the time series dynamic change of the hardware accessories inventory type and emphasize the characteristic diagram.
10. The hardware and building materials storage management system according to claim 9 is characterized in that: The hardware accessories related warehouse data feature encoding module is used to: input the hardware accessories inventory related indicator time series into the tensor through the inventory change encoder based on the three-dimensional convolution kernel to obtain the hardware accessories inventory time series dynamic change feature map.
