Intelligent management method and system for bank seal cards

Abstract

The application provides an intelligent management method and system for bank seal cards, and relates to the field of article management.The method comprises the following steps: performing use homochromatic analysis on historical use data of seal cards of multiple enterprises, and generating use homochromatic analysis results; determining a regional storage scheme of the seal cards of the multiple enterprises based on the use homochromatic analysis results, and allocating an RFID tag to the seal cards of each enterprise; after storing the seal cards of the multiple enterprises into a seal card cabinet based on the regional storage scheme of the seal cards of the multiple enterprises, determining a storage position of the seal cards of an enterprise based on the RFID tag and an RFID positioning component of the seal card cabinet; determining an optimal inventory path based on use data and the storage position of the seal cards of the multiple enterprises in a current period; and controlling an RFID reader-writer to perform inventory of the seal cards of the multiple enterprises according to the optimal inventory path, and generating an automatic inventory result, so that the management level of the seal cards is improved, and the efficient and safe operation of bank business is ensured.

Description

Technical Field

[0001] This invention relates to the field of goods management, and in particular to an intelligent warehouse management method and system for bank signature cards. Background Technology

[0002] In the daily operations of banks, the signature card serves as the core credential for corporate identity verification and authorization of fund transactions. Its management efficiency and security directly affect business operations and risk control.

[0003] Currently, the number of bank signature cards is enormous and continues to grow. Traditional inventory management methods often rely on fixed-location storage, failing to consider the varying usage frequencies of different signature cards. Mixing frequently used signature cards with infrequently used ones forces staff to search frequently, wasting significant time, reducing operational efficiency, and potentially increasing the risk of card damage or loss due to frequent handling. Furthermore, while some existing signature card inventory management systems attempt to incorporate information technology, they suffer from significant functional deficiencies. Some systems only provide simple electronic registration, still relying on manual inventory checks and failing to automate inventory verification. This not only results in low efficiency but also makes them susceptible to errors due to human negligence.

[0004] Therefore, there is a need to provide intelligent inventory management methods and systems for bank signature cards to improve the management level of signature cards and ensure the efficient and secure operation of banking business. Summary of the Invention

[0005] This invention provides an intelligent inventory management method for bank signature cards, comprising: acquiring historical usage data of signature cards from multiple enterprises; performing usage frequency analysis on the historical usage data of signature cards from multiple enterprises to generate usage frequency analysis results; determining a regional storage scheme for signature cards from multiple enterprises based on the usage frequency analysis results, and assigning RFID tags to signature cards from each enterprise; after storing signature cards from multiple enterprises in a signature card cabinet based on the regional storage scheme, determining the storage location of the signature cards from each enterprise based on the RFID tags and the RFID positioning components of the signature card cabinet; acquiring usage data and storage location of signature cards from multiple enterprises in the current period; determining the optimal inventory path based on the usage data and storage location of signature cards from multiple enterprises in the current period; and controlling an RFID reader / writer to perform an inventory of signature cards from multiple enterprises according to the optimal inventory path, generating automatic inventory results.

[0006] Furthermore, a usage frequency analysis is performed on the historical usage data of seal cards from multiple enterprises to generate usage frequency analysis results. This includes: for each enterprise, calculating the average and fluctuation values ​​of seal card usage frequency based on the enterprise's historical usage data, and determining the enterprise's type based on the average and fluctuation values ​​of seal card usage frequency, wherein the enterprise type is either Category I or Category II; and performing usage frequency analysis on the historical usage data of seal cards from multiple Category I enterprises to generate usage frequency analysis results.

[0007] Furthermore, a usage frequency analysis is performed on the historical usage data of seal cards of multiple Class A enterprises to generate usage frequency analysis results. This includes: determining multiple historical time periods based on the historical usage data of seal cards of multiple enterprises; for each Class A enterprise, based on the multiple historical time periods, dividing the historical usage data of the seal cards of the Class A enterprise into multiple historical usage data units, where one historical usage data unit corresponds to one historical time period; for any two Class A enterprises, dividing each historical usage data unit of the two Class A enterprises into multiple historical usage data groups; calculating the usage frequency coefficient of the two Class A enterprises for the corresponding historical time period based on the multiple historical usage data groups included in each historical usage data unit of the two Class A enterprises; and calculating the frequency value of the two Class A enterprises based on the usage frequency coefficient of each historical time period of the two Class A enterprises. The usage frequency analysis results include the frequency values ​​of any two Class A enterprises.

[0008] Furthermore, based on the results of frequency analysis, a regional storage scheme for the signature cards of multiple enterprises is determined, including: for each enterprise, calculating the average usage frequency of the enterprise's signature cards based on the historical usage data of the enterprise's signature cards; determining the priority value of the storage cell based on the coordinates of each storage cell in the signature card cabinet; clustering enterprises in a class based on the frequency values ​​of any two enterprises in the same class using a clustering algorithm to determine multiple classes; and constructing a first multi-objective fitness function, wherein the regional storage scheme for the signature cards of multiple enterprises is determined based on the first multi-objective fitness function using a particle swarm optimization algorithm.

[0009] Furthermore, based on RFID tags and the RFID positioning component of the signature card cabinet, the storage location of the company's signature cards is determined, including: setting up multiple positioning readers on the signature card cabinet, wherein the RFID positioning component includes multiple positioning readers; and using multiple positioning readers to interact with RFID tags to determine the storage location of the company's signature cards.

[0010] Furthermore, multiple positioning readers are installed on the seal card cabinet, including: determining multiple positioning reader installation schemes; for each positioning reader installation scheme, sampling multiple storage compartments from the seal card cabinet, installing RFID tags in the sampled storage compartments, acquiring the signal interaction characteristics of the positioning reader installation scheme corresponding to the RFID tags installed in each sampled storage compartment, calculating the global difference of the signal interaction characteristics corresponding to the positioning reader installation scheme based on the signal interaction characteristics and priority value of the positioning reader installation scheme corresponding to the RFID tags installed in each sampled storage compartment; determining the optimal positioning reader installation scheme based on the global difference of the signal interaction characteristics corresponding to each positioning reader installation scheme; and installing multiple positioning readers on the seal card cabinet according to the optimal positioning reader installation scheme.

[0011] Furthermore, based on the usage data and storage location of the signature cards of multiple enterprises in the current period, the optimal inventory path is determined, including: determining the inventory priority of each enterprise based on the usage data and storage location of the signature cards of multiple enterprises in the current period; constructing a second multi-objective fitness function, wherein the second multi-objective fitness function is related to the inventory priority of each enterprise and the path length of the inventory path; and determining the optimal inventory path based on the second multi-objective fitness function using the particle swarm optimization algorithm.

[0012] Furthermore, the method also includes: determining multiple profiling factors; constructing enterprise profiles for multiple enterprises based on the multiple profiling factors; determining multiple influencing profiling factors based on the enterprise profiles of multiple enterprises and the historical usage data of the enterprise's seal card; constructing an enterprise profile for the enterprise that first stores its seal card based on the multiple influencing profiling factors; and determining the storage cell for the enterprise that first stores its seal card based on the enterprise profile of the enterprise that first stores its seal card.

[0013] Furthermore, based on the corporate profile of the company that first deposits its seal card, the storage space for that company is determined, including: predicting the type of the company based on its corporate profile; if the predicted type is Category II, predicting the average seal card usage frequency of the company based on its corporate profile, and determining the storage space for the company based on the predicted average seal card usage frequency; if the predicted type is Category I, predicting the average seal card usage frequency and the Category I group to which the company belongs based on its corporate profile, and determining the storage space for the company based on the predicted average seal card usage frequency and the Category I group to which the company belongs.

[0014] This invention provides an intelligent inventory management system for bank signature cards, applied to the aforementioned intelligent inventory management method for bank signature cards. The system includes: a data acquisition module for acquiring historical usage data of signature cards from multiple enterprises; a frequency analysis module for performing frequency analysis on the historical usage data of signature cards from multiple enterprises, generating frequency analysis results; a storage management module for determining a regional storage scheme for signature cards from multiple enterprises based on the frequency analysis results, and assigning RFID tags to signature cards from each enterprise; the storage management module is further used to, based on the regional storage scheme of signature cards from multiple enterprises, after storing the signature cards of multiple enterprises in a signature card cabinet, determine the storage location of the enterprise's signature cards based on the RFID tags and the RFID positioning components of the signature card cabinet; the data acquisition module is further used to acquire usage data and storage location of signature cards from multiple enterprises in the current period; and an intelligent inventory module for determining the optimal inventory path based on the usage data and storage location of signature cards from multiple enterprises in the current period, and controlling an RFID reader / writer to perform an inventory of signature cards from multiple enterprises according to the optimal inventory path, generating automatic inventory results.

[0015] Compared with existing technologies, the intelligent inventory management method and system for bank signature cards provided by this invention have at least the following beneficial effects:

[0016] 1. By analyzing the historical usage data of enterprise seal cards, the enterprise type is determined, and enterprises of the same type are further clustered. Combining multiple objectives such as the average usage frequency of seal cards, storage cell priority, and average storage distance within each group, a particle swarm optimization algorithm is used to determine a regional storage scheme. This makes seal card storage more scientific and rational, allowing frequently used seal cards to be stored in more suitable locations, reducing retrieval time and improving business processing efficiency. At the same time, a reasonable regional division facilitates centralized management, reduces management costs, and improves overall warehouse management efficiency.

[0017] 2. Multiple positioning readers are installed in the signature card cabinet. The optimal setup is determined through comparison of various methods, and the RFID positioning components interact with tag signals to accurately determine the storage location of the signature cards. Inventory priorities are determined based on current usage data and storage location. A multi-objective fitness function is constructed, and a particle swarm optimization algorithm is used to determine the optimal inventory path. This makes inventory work more accurate and efficient, enabling rapid and accurate monitoring of signature card status, timely detection of anomalies, and ensuring data accuracy, providing reliable data support for bank signature card management.

[0018] 3. Construct enterprise profiles and determine influencing factors. For enterprises storing signature cards for the first time, a profile is created. Based on the profile, the enterprise type, average signature card usage frequency, and group affiliation are predicted, thus determining the storage space. This method effectively adapts to the storage needs of new enterprise signature cards, ensuring their proper storage and management. It enhances the adaptability and scalability of the entire intelligent warehouse management system, providing strong support for the sustainable development of bank signature card management business. Attached Figure Description

[0019] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0020] Figure 1 This is a flowchart illustrating an intelligent inventory management method for bank signature cards according to some embodiments of this specification.

[0021] Figure 2 This is a flowchart illustrating a regional storage scheme for signature cards of multiple enterprises, as shown in some embodiments of this specification.

[0022] Figure 3 This is a flowchart illustrating the process of determining the storage compartment for the company that first stores the signature card, according to some embodiments of this specification.

[0023] Figure 4 This is a schematic diagram of a module for an intelligent inventory management system for bank signature cards, as shown in some embodiments of this specification. Detailed Implementation

[0024] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0025] Figure 1 This is a flowchart illustrating an intelligent inventory management method for bank signature cards, based on some embodiments of this specification. Figure 1 As shown, the intelligent inventory management method for bank signature cards may include the following steps:

[0026] Step 110: Obtain historical usage data of seal cards from multiple companies.

[0027] Specifically, when businesses conduct various transactions with banks, such as account fund inflows and outflows, bill transactions, and loan transactions, banks will access the signature card information for verification whenever the business's identity and authorization are required. The historical usage data of a business's signature card can include records of signature card accesses over a past period (e.g., 3 years, 5 years), including the time of each access.

[0028] Step 120: Perform usage frequency analysis on the historical usage data of seal cards of multiple enterprises and generate usage frequency analysis results.

[0029] Specifically, it includes:

[0030] For each enterprise, based on the historical usage data of the enterprise's seal card, the average and fluctuation values ​​of the enterprise's seal card usage frequency are calculated. Based on the average and fluctuation values ​​of the enterprise's seal card usage frequency, the enterprise type is determined, where the enterprise type is either Category I or Category II.

[0031] Perform usage frequency analysis on the historical usage data of seal cards of multiple companies of the same category, and generate usage frequency analysis results.

[0032] Specifically, for each enterprise, based on the historical usage data of the enterprise's seal card, the number of times the enterprise uses the seal card in multiple historical periods (e.g., one day, one week, one month, etc.) can be determined, the average value can be taken as the average frequency of the enterprise's seal card usage, and the variance can be taken as the fluctuation value of the enterprise's seal card usage frequency.

[0033] If the average frequency of a company's signature card usage is less than the average threshold (e.g., 1, 2, etc.), the company is classified as a Category II company.

[0034] If the average frequency of a company’s signature card usage is greater than or equal to the average threshold, and the fluctuation value of the signature card usage frequency is less than the fluctuation value threshold (e.g., 2, 3, etc.), then the company is classified as a Class II company.

[0035] Otherwise, the enterprise will be classified as a Class I enterprise.

[0036] In some embodiments, a usage frequency analysis is performed on the historical usage data of signature cards of multiple enterprises of the same type to generate usage frequency analysis results, including:

[0037] Based on the historical usage data of seal cards of multiple companies, multiple historical time periods are determined. For example, if the historical usage data of a company's seal cards corresponds to the past three years, then the past three years can be divided into multiple historical time periods, with each month as a historical time period.

[0038] For each type of enterprise, based on multiple historical time periods, the historical usage data of the seal card of the enterprise of type 1 is divided into multiple historical usage data units, where one historical usage data unit corresponds to one historical time period.

[0039] For any two companies of the same category, each historical usage data unit of the two companies of the same category is divided into multiple historical usage data groups. For example, according to a preset period (e.g., one day, one week, one month, etc.), each historical usage data unit of the two companies of the same category is divided into multiple historical usage data groups. Based on the multiple historical usage data groups included in each historical usage data unit of the two companies of the same category, the usage frequency coefficient of the two companies of the same category for the corresponding historical time period is calculated. Based on the usage frequency coefficient of the two companies of the same category for each historical time period, the frequency value of the two companies of the same category is calculated. The frequency analysis results include the frequency values ​​of any two companies of the same category.

[0040] Specifically, for each historical time period, the total number of times the seal cards of two companies of the same category were called in multiple historical periods of the corresponding historical time period can be used as two variables. These variables can be substituted into the formula for calculating the correlation coefficient (e.g., Pearson correlation coefficient, Spearman correlation coefficient, etc.) to calculate the frequency coefficient of use of the two companies of the same category in the corresponding historical time period.

[0041] The average of the frequency coefficients used by the two companies in the same category for each historical time period is calculated to obtain the frequency values ​​of the two companies in the same category.

[0042] In the scenario of intelligent inventory management of bank signature cards, the enterprise type is determined based on the average and fluctuation values ​​of the usage frequency of enterprise signature cards, and then a regional storage plan is formulated. This classification storage method makes the signature card storage layout more reasonable, makes full use of storage space, reduces search time, and improves overall inventory management efficiency. At the same time, it also helps to provide more accurate services for different types of enterprises, and enhances the bank's professionalism and responsiveness in signature card management.

[0043] Analyzing the frequency of signature card usage among a group of enterprises can identify closely related business groups. Storing signature cards of enterprises with high frequency values ​​in adjacent areas facilitates simultaneous processing of related business, further improving business collaboration efficiency. Moreover, based on the frequency analysis results, storage space allocation can be optimized, making the storage layout more aligned with the actual business needs of enterprises. This helps banks achieve refined management in signature card inventory control, reduce management costs, improve service quality, better adapt to the diverse business needs of enterprises, and ensure the scientific and efficient management of bank signature cards.

[0044] Step 130: Based on the results of the frequency analysis, determine the regional storage scheme for the signature cards of multiple enterprises, and assign RFID tags to the signature cards of each enterprise.

[0045] like Figure 2 As shown, it specifically includes:

[0046] For each enterprise, the average frequency of seal card usage is calculated based on the enterprise's historical seal card usage data.

[0047] Based on the coordinates of each storage compartment in the seal card cabinet, the priority value of the storage compartment is determined;

[0048] By using clustering algorithms, companies in a class are clustered based on the frequency values ​​of any two companies in the same class, thus identifying multiple groups of companies in the same class.

[0049] Construct the first multi-objective fitness function;

[0050] Using the particle swarm optimization algorithm and based on the first multi-objective fitness function, a regional storage scheme for the signature cards of multiple enterprises is determined.

[0051] Specifically, the priority value of each storage compartment in the signature card cabinet is determined based on its coordinates. The coordinates of a storage compartment may relate to its position within the cabinet, such as its distance from the cabinet exit and whether it is at an easily accessible height. The priority value is set to measure the suitability of each storage compartment for storing signature cards. For example, a storage compartment closer to the cabinet exit and at a suitable operating height has a higher priority value because such a location facilitates quick retrieval of signature cards, improving business processing efficiency. The priority values ​​of storage compartments can be preset manually or calculated by analyzing the differences between the distance of different storage compartments from the cabinet exit and their operating height.

[0052] Clustering algorithms (such as K-means clustering) are used to group companies within a group based on the frequency of their interactions with each other, thus creating multiple clusters. The frequency reflects the synchronicity of the timing of signature card usage between two companies in the same group. Clustering algorithms group companies with similar frequency values ​​together, indicating that companies within the same group tend to have similar timing patterns in their signature card usage. For example, some companies may use signature cards intensively during specific time periods each month; clustering can group these companies with similar usage patterns into one cluster.

[0053] The independent variables of the constructed first multi-objective fitness function are correlated with the enterprise's mean frequency, the priority of the storage cell, and the mean storage distance for each group. This function comprehensively considers multiple factors to find an optimal storage scheme. A high mean frequency for an enterprise means a more convenient storage location is needed; the priority of the storage cell reflects the quality of the storage cell itself; and the mean storage distance for each group considers the concentration of storage locations for the same group of enterprise seal cards, facilitating simultaneous processing of related business. For example, the first multi-objective fitness function tends to store enterprise seal cards with high mean frequency in storage cells with high priority and keeps enterprise seal cards in the same group closer together.

[0054] As an example only, the first multi-objective fitness function can be:

[0055]

[0056] in, For the first multi-objective fitness function, Let $\frac{i}{i}$ be the normalized mean frequency of signature card usage for the $i$-th company. Let be the priority value of the storage cell for the signature card of the i-th enterprise after normalization, where N is the total number of enterprises and M is the total number of groups. Let R be the average storage distance corresponding to the m-th group of first-class enterprises, and let R be the total number of first-class enterprises included in the m-th group of first-class enterprises. Let be the coordinate distance between the storage cell of the g-th enterprise in the m-th group and the storage cell of the h-th enterprise in the m-th group.

[0057] In the first multi-objective fitness function mentioned above, the average frequency of use of the company's signature card is... This reflects the frequency of the company's signature card usage. Companies with high usage frequency require more convenient storage locations for their signature cards for quick retrieval. Storage slot priority. The system measures the relative importance of each storage slot for storing signature cards, with higher priority slots offering better operational convenience. Multiplying the normalized mean usage frequency by the priority value and summing the results reflects the goal of allocating more efficient storage locations to frequently used company signature cards. For example, if a company signature card is used frequently and is assigned to a high-priority slot, this term will be larger, which is beneficial for optimizing the overall fitness function. Clusters are formed by clustering companies within a specific group based on their frequency of use, reflecting the synchronicity of signature card usage. Storing signature cards from the same group closer together facilitates simultaneous processing of related business. This represents the average storage distance for the m-th group, calculated by distributing the coordinate distances between all pairs of storage cells belonging to all companies within that group. It is derived from the average value. In the formula... The design ensures that this value is larger when the average storage distance is small, and smaller when the average storage distance is large. This encourages the centralized storage of corporate seal cards of the same type, facilitating business collaboration and improving overall warehouse management efficiency.

[0058] By combining these two parts, the first multi-objective fitness function comprehensively considers the usage frequency of corporate seal cards, the quality of storage cells, and the concentration of storage locations for corporate seal cards of the same type. During the particle swarm optimization process, the algorithm continuously adjusts the storage locations of corporate seal cards to maximize the value of the first multi-objective fitness function, thereby finding a regional storage scheme that is optimized in multiple aspects. For example, during optimization, the algorithm tends to allocate frequently used corporate seal cards to storage cells with higher priority values, while also storing corporate seal cards of the same type closer together to achieve the optimal value of the overall fitness function, thus realizing efficient storage and management of seal cards.

[0059] After determining the zoned storage scheme, RFID tags are assigned to each company's signature card. Each RFID tag uniquely identifies the signature card, and combined with the RFID positioning components of the signature card cabinet, precise management of the card's location is achieved, facilitating subsequent inventory checks and retrieval. For example, when a company's signature card needs to be retrieved, its location within the cabinet can be quickly determined using the RFID tag, improving work efficiency.

[0060] Step 140: Based on the regional storage scheme of multiple companies' signature cards, after storing the signature cards of multiple companies in the signature card cabinet, the storage location of the company's signature cards is determined based on the RFID tags and the RFID positioning components of the signature card cabinet.

[0061] Specifically, multiple positioning readers are installed on the signature card cabinet. The RFID positioning component includes multiple positioning readers, which interact with the RFID tags to determine the storage location of the company's signature cards.

[0062] Specifically, each company's signature card is assigned a unique RFID tag containing specific information about that signature card. When a signature card is stored in the card cabinet, the positioning reader actively emits a radio frequency signal. Upon receiving the signal, the RFID tag is activated and then sends a signal containing its own information back to the reader. Through this signal interaction, the positioning reader can obtain relevant data from the tag on the signature card. Using the signal strength information received by multiple positioning readers, triangulation or polygonal positioning algorithms are employed to determine the exact location of the signature card. After accurately determining the storage location of the signature cards, staff can quickly find the required signature cards during subsequent business processing, eliminating the need for blind searching in the card cabinet, significantly saving time and improving business processing efficiency. For example, when a signature card from a particular company is needed, its storage location can be found through the system, and staff can directly retrieve the card from the corresponding location. During regular inventory checks of the signature cards, the actual stored signature cards can be quickly verified against the recorded storage locations to promptly identify lost or misplaced signature cards, ensuring the secure management of signature cards.

[0063] In some embodiments, multiple positioning readers are installed on the signature card cabinet, including:

[0064] Determine multiple location reader setup schemes;

[0065] For each positioning reader setup scheme, multiple storage compartments are sampled from the seal card cabinet, and RFID tags are set in the sampled storage compartments. The signal interaction characteristics of the positioning reader setup scheme corresponding to the RFID tags set in each sampled storage compartment are obtained. Based on the signal interaction characteristics and priority values ​​of the positioning reader setup scheme corresponding to the RFID tags set in each sampled storage compartment, the global difference of the signal interaction characteristics corresponding to the positioning reader setup scheme is calculated.

[0066] Based on the global differences in signal interaction characteristics corresponding to each positioning reader setup scheme, the optimal positioning reader setup scheme is determined.

[0067] Based on the optimal positioning reader setup scheme, multiple positioning readers are installed on the seal card cabinet.

[0068] Specifically, due to differences in the structure, size, and expected positioning accuracy requirements of the signature card cabinet, a single positioning reader setup may not meet actual needs. Therefore, it is necessary to design multiple different setup schemes to select the optimal one, ensuring that the positioning reader can efficiently and accurately determine the location of the signature card.

[0069] Location reader installation schemes can cover multiple aspects, such as the number of readers, their installation location, and installation angle. For example, Scheme 1 might involve installing one location reader at each of the four corners of the signature card cabinet; Scheme 2 might involve evenly distributing multiple readers on the top of the cabinet; Scheme 3 might involve installing readers on both the sides and top of the cabinet. Different schemes will take into account factors such as the spatial layout of the signature card cabinet and the signal coverage area.

[0070] Signature card cabinets typically have a large number of storage compartments, and testing each compartment individually would be time-consuming and resource-intensive. Sampling a subset of compartments for testing can significantly reduce workload and improve efficiency while maintaining accuracy. A priority-based sampling method can be used to ensure that the sampled compartments are representative of the entire cabinet; higher priority values ​​result in a higher probability of being sampled. For example, all compartments in the cabinet can be numbered, and a random number generator can be used to select a certain number of compartments corresponding to those numbers as samples. RFID tags are then placed in each of the selected compartments. These tags will interact with a positioning reader to obtain relevant signal characteristic data. The tags must be securely fixed in the compartments, and the signals must be transmitted and received correctly.

[0071] The signal interaction characteristics of the RFID tags set in the sampled storage cells and the corresponding positioning reader settings can include the signal strength of the interaction between the RFID tags set in the sampled storage cells and each positioning reader in the positioning reader settings, forming a signal interaction feature vector. Here, one element of the signal interaction feature vector represents the signal strength of the interaction between the RFID tags set in the sampled storage cells and one positioning reader in the positioning reader settings.

[0072] For each sampled cell, calculate the Euclidean distance between the signal interaction feature vector corresponding to the sampled cell and the signal interaction feature vectors corresponding to all other sampled cells, and calculate the mean value to obtain the mean Euclidean distance for the sampled cell. Based on the mean Euclidean distance for the sampled cell, calculate the single-point difference of the signal interaction feature for the sampled cell. The larger the mean Euclidean distance, the larger the single-point difference of the signal interaction feature for the sampled cell.

[0073] Using priority values ​​as weights, a weighted average is calculated on the single-point differences in signal interaction features of each sampled storage cell to obtain the global differences in signal interaction features corresponding to the positioning reader setup scheme. The positioning reader setup scheme with the largest global differences in signal interaction features is taken as the optimal positioning reader setup scheme, and multiple positioning readers are set up on the seal card cabinet.

[0074] By sampling multiple storage compartments for each positioning reader setup and attaching RFID tags to acquire signal interaction characteristics, the signal performance of different setups in various areas of the seal card cabinet can be comprehensively assessed, avoiding result bias caused by testing a single area. Global differences are calculated based on signal interaction characteristics and priority values. Priority value sampling ensures sample representativeness, combining local signal conditions with overall importance, making the evaluation more aligned with actual needs. Calculating single-point and global differences in signal interaction characteristics accurately measures the uniformity and stability of signal interaction in each storage compartment under different setups. A larger global difference indicates a more significant differentiation in signal interaction across the entire setup, which is more conducive to the positioning reader accurately distinguishing seal cards in different locations. Finally, the optimal setup is determined based on global differences, selecting the setup that provides the most stable and accurate signal interaction effect within the seal card cabinet spatial layout. This ensures that the positioning reader efficiently and accurately determines the seal card location, improving the efficiency and accuracy of seal card management and reducing business risks and costs caused by inaccurate positioning.

[0075] Step 150: Obtain the usage data and storage location of the signature cards of multiple enterprises in the current period.

[0076] Step 160: Determine the optimal inventory path based on the usage data and storage location of the signature cards of multiple enterprises in the current period.

[0077] Specifically, it includes:

[0078] Based on the usage data and storage location of signature cards from multiple companies in the current period, the inventory priority for each company is determined. For example, for each company, the number of times its signature card is used in the current period is calculated. The higher the usage frequency, the higher the inventory priority for that company. The usage frequency of signature cards in the current period is an important indicator of their importance and activity. Companies with higher usage frequency are more likely to have their signature cards used frequently. Prioritizing the inventory of these companies' signature cards during the inventory process allows for timely monitoring of their status, ensuring that critical business operations are not affected. For instance, if a company frequently uses signature cards for important business transactions such as contract signing and fund transfers in the current period, failure to promptly inventory its signature cards could lead to serious economic losses and business risks should the signature cards be lost or damaged. Therefore, determining the inventory priority based on usage frequency makes the inventory work more targeted, improving the efficiency and value of the inventory process.

[0079] A second multi-objective fitness function is constructed, which is related to the inventory priority and path length of each enterprise. Specifically, enterprises with higher inventory priority are inventoried earlier, have shorter path lengths, and thus have larger second multi-objective fitness function values. The second multi-objective fitness function aims to comprehensively consider two key factors—inventory priority and path length—to find an optimal solution that satisfies both the need for prioritizing high-priority enterprises and minimizing path length. In actual inventory work, these two objectives are often mutually restrictive. For example, prioritizing high-priority enterprises may require longer detours; conversely, shortening path length may delay inventorying high-priority enterprises. Therefore, by constructing a multi-objective fitness function, a balance can be found between these two factors, achieving the optimal overall inventory results.

[0080] The optimal inventory path is determined using the particle swarm optimization (PSO) algorithm, based on a second multi-objective fitness function. Specifically, a set of initial inventory path schemes is randomly generated as a particle swarm, with each particle representing a possible inventory order and path. The fitness function value for each particle (i.e., each inventory path scheme) is calculated according to the constructed second multi-objective fitness function. A higher fitness function value indicates a better scheme. The particle's velocity and position are updated based on its own historical best position (individual optimum) and the best position of all particles in the swarm (global optimum). Particles move towards both individual and global optima while maintaining a degree of randomness to avoid getting trapped in local optima. For example, if a particle's current position corresponds to a low fitness function value, and it finds a particle in the swarm with a higher fitness function value, it will adjust its velocity and position to move closer to that global optimum position, repeating the process of calculating the fitness function value and updating the particle's velocity and position until the termination condition is met (such as reaching the maximum number of iterations or the fitness function value converging). Finally, the inventory path scheme corresponding to the global optimum position is the optimal inventory path.

[0081] Step 170: Based on the optimal inventory path, control the RFID reader / writer to conduct an inventory of the signature cards of multiple companies and generate automatic inventory results.

[0082] Specifically, RFID readers can be mounted on mobile components, which can adjust the position of the RFID readers according to the optimal inventory path to conduct inventory of signature cards for multiple companies.

[0083] like Figure 3 As shown, in some embodiments, it also includes:

[0084] Identify multiple profile factors;

[0085] Based on multiple profiling factors, construct enterprise profiles for multiple companies;

[0086] Based on the corporate profiles of multiple enterprises and the historical usage data of corporate seal cards, several influential profile factors were identified.

[0087] Based on multiple influencing profiling factors, construct a corporate profile for the company that stores its signature card for the first time.

[0088] Based on the corporate profile of the company that is storing the signature card for the first time, the storage compartment for the company that is storing the signature card for the first time is determined.

[0089] Specifically, profile factors are a series of key variables used to accurately characterize a company's features. They reflect the company's overall situation from different dimensions, providing basic data support for building a company profile. In the context of signature card management, these factors help reflect the characteristics of a company related to signature card usage. Multiple profile factors can include the company's size, typically expressed as the number of employees, asset size, business transaction volume, number of business partners, debt-to-equity ratio, and profitability.

[0090] For each enterprise, based on the enterprise's information, multiple profiling factors are quantified. For example, the enterprise size is divided into different levels according to certain standards and assigned corresponding scores, and industry attributes are coded.

[0091] We collected historical usage data of signature cards from multiple companies, including usage time, frequency, usage scenarios, and problems encountered during use. This data was then correlated with various profiling factors in the company profile. Using statistical analysis methods and machine learning algorithms, we identified profiling factors closely related to the frequency of signature card usage, which were then used as influencing profiling factors. For example, the analysis revealed that larger companies generally had higher signature card usage frequencies, companies in the financial industry had significantly higher signature card usage frequencies than other industries, and companies with lower credit ratings had more instances of irregular signature card usage, which could potentially affect their normal usage frequency.

[0092] The corporate profile of a company that is depositing its signature card for the first time can include scores of multiple influential profile factors based on information about the company that is depositing its signature card for the first time.

[0093] In some embodiments, determining the storage compartment for the company that first stores the signature card, based on the company profile, includes:

[0094] Based on the corporate profile of the company that stores its signature card for the first time, predict the type of company that stores its signature card for the first time;

[0095] If the type of enterprise that stores the signature card for the first time is predicted to be Category II, based on the enterprise profile of the enterprise that stores the signature card for the first time, the average frequency of signature card usage of the enterprise that stores the signature card for the first time is predicted. Based on the predicted average frequency of signature card usage of the enterprise that stores the signature card for the first time, the storage compartment of the enterprise that stores the signature card for the first time is determined.

[0096] If the type of enterprise that stores the signature card for the first time is predicted to be of category one, based on the enterprise profile of the enterprise that stores the signature card for the first time, predict the average frequency of signature card use and the category to which the enterprise belongs. Based on the predicted average frequency of signature card use and the category to which the enterprise belongs, determine the storage compartment of the enterprise that stores the signature card for the first time.

[0097] Specifically, the similarity between the company profile of the company that first stores its signature card and the company profile of each company that has already stored its signature card can be calculated. Companies that have already stored their signature cards with a similarity greater than a similarity threshold (e.g., 80%) are considered similar companies with stored signature cards. Based on the average usage frequency of signature cards of similar companies with stored signature cards, the average usage frequency of signature cards for the company that first stores its signature card can be predicted. For example, the average usage frequency of signature cards for similar companies with stored signature cards can be taken again as the predicted average usage frequency of signature cards for the company that first stores its signature card. The empty slots of companies with stored signature cards that are closest to the average usage frequency of signature cards are designated as the slots of the companies that first store their signature cards.

[0098] Based on the group affiliation of companies with similar existing signature cards, predict the group affiliation of companies that will be depositing signature cards for the first time. For example, the group affiliation of companies with the largest proportion of similar existing signature cards is used as the group affiliation of companies that will be depositing signature cards for the first time. The empty slots of companies with existing signature cards that have similar average signature card usage frequencies in the closest affiliation group are used as the slots for companies that will be depositing signature cards for the first time.

[0099] By calculating the similarity of enterprise profiles, similar enterprises are accurately screened. The average frequency of seal card usage among these similar enterprises is used to predict the initial placement of storage spaces for enterprises, significantly improving prediction accuracy. Because similar enterprises share similar business models, scale, and industry characteristics, their seal card usage frequency is highly valuable, avoiding blind prediction and ensuring that storage space selection better aligns with actual enterprise needs. Secondly, when determining which group a company belongs to, the classification is based on the highest proportion of similar enterprises, further refining enterprise categorization and providing a more precise basis for storage space selection. Furthermore, using vacant storage spaces near those of enterprises with the highest average seal card usage frequency as initial placement spaces for enterprises considers both usage frequency (ensuring high-frequency users can quickly access their spaces) and space utilization (avoiding resource waste). Simultaneously, for enterprises within a category, combining both group affiliation and average seal card usage frequency in storage space selection improves overall management efficiency.

[0100] Figure 4 These are schematic diagrams of modules for an intelligent inventory management system for bank signature cards, as shown in some embodiments of this specification. Figure 4 As shown, the intelligent inventory management system for bank signature cards may include a data acquisition module, a frequency analysis module, a storage management module, and an intelligent inventory module.

[0101] The data acquisition module is used to acquire historical usage data of seal cards from multiple companies;

[0102] The frequency analysis module is used to perform frequency analysis on the historical usage data of seal cards of multiple enterprises and generate frequency analysis results.

[0103] The storage management module is used to determine the regional storage scheme for the signature cards of multiple enterprises based on the results of frequency matching analysis, and to assign RFID tags to the signature cards of each enterprise.

[0104] The storage management module is also used to determine the storage location of the company's seal cards based on the RFID tags and the RFID positioning components of the seal card cabinet after storing the seal cards of multiple companies in the seal card cabinet.

[0105] The data acquisition module is also used to acquire the usage data and storage location of the signature cards of multiple enterprises in the current period;

[0106] The intelligent inventory module is used to determine the optimal inventory path based on the usage data and storage location of the signature cards of multiple companies in the current period, and to control the RFID reader to perform inventory of the signature cards of multiple companies according to the optimal inventory path, and generate automatic inventory results.

[0107] For a more detailed description of the intelligent inventory management system for bank signature cards, please refer to the relevant description of the intelligent inventory management method for bank signature cards, which will not be repeated here.

[0108] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An intelligent inventory management method for bank signature cards, characterized in that, include: Obtain historical usage data of signature cards from multiple companies; Perform usage frequency analysis on the historical usage data of seal cards from multiple enterprises to generate usage frequency analysis results; Based on the results of frequency co-frequency analysis, a regional storage scheme for the signature cards of multiple enterprises was determined, and RFID tags were assigned to the signature cards of each enterprise. Based on the regional storage scheme of multiple companies' signature cards, after storing the signature cards of multiple companies in the signature card cabinet, the storage location of the company's signature cards is determined based on RFID tags and the RFID positioning components of the signature card cabinet. Obtain usage data and storage location of signature cards from multiple companies in the current period; Based on the usage data and storage location of signature cards from multiple companies in the current period, the optimal inventory route is determined. Based on the optimal inventory path, control the RFID reader / writer to conduct inventory of signature cards of multiple enterprises and generate automatic inventory results; This includes conducting usage frequency analysis on historical usage data of signature cards from multiple enterprises, generating usage frequency analysis results, including: For each enterprise, based on the historical usage data of the enterprise's seal card, the average and fluctuation values ​​of the enterprise's seal card usage frequency are calculated. Based on the average and fluctuation values ​​of the enterprise's seal card usage frequency, the enterprise type is determined, where the enterprise type is either Category I or Category II. Based on the historical usage data of seal cards from multiple companies, multiple historical time periods were identified. For each type of enterprise, based on multiple historical time periods, the historical usage data of the seal card of the enterprise of type 1 is divided into multiple historical usage data units, where one historical usage data unit corresponds to one historical time period. For any two enterprises of the same category, each historical usage data unit of the two enterprises of the same category is divided into multiple historical usage data groups. Based on the multiple historical usage data groups included in each historical usage data unit of the two enterprises of the same category, the usage frequency coefficient of the two enterprises of the same category for the corresponding historical time period is calculated. Based on the usage frequency coefficient of the two enterprises of the same category for each historical time period, the frequency value of the two enterprises of the same category is calculated. The usage frequency analysis results include the frequency value of any two enterprises of the same category. Based on the results of frequency co-occurrence analysis, a regional storage scheme for signature cards of multiple enterprises was determined, including: For each enterprise, the average frequency of seal card usage is calculated based on the enterprise's historical seal card usage data. Based on the coordinates of each storage compartment in the seal card cabinet, the priority value of the storage compartment is determined; By using clustering algorithms, companies in a class are clustered based on the frequency values ​​of any two companies in the same class, thus identifying multiple groups of companies in the same class. Construct the first multi-objective fitness function; Using the particle swarm optimization algorithm and based on the first multi-objective fitness function, a regional storage scheme for the signature cards of multiple enterprises is determined.

2. The intelligent inventory management method for bank signature cards according to claim 1, characterized in that, Based on RFID tags and RFID positioning components in the signature card cabinet, the location of a company's signature cards is determined, including: Multiple positioning readers are installed on the seal card cabinet, and the RFID positioning component includes multiple positioning readers; By using multiple positioning readers to interact with RFID tags, the location of the company's signature card can be determined.

3. The intelligent inventory management method for bank signature cards according to claim 2, characterized in that, Multiple positioning readers are installed on the signature card cabinet, including: Determine multiple location reader setup schemes; For each positioning reader setup scheme, multiple storage compartments are sampled from the seal card cabinet, and RFID tags are set in the sampled storage compartments. The signal interaction characteristics of the positioning reader setup scheme corresponding to the RFID tags set in each sampled storage compartment are obtained. Based on the signal interaction characteristics and priority values ​​of the positioning reader setup scheme corresponding to the RFID tags set in each sampled storage compartment, the global difference of the signal interaction characteristics corresponding to the positioning reader setup scheme is calculated. Based on the global differences in signal interaction characteristics corresponding to each positioning reader setup scheme, the optimal positioning reader setup scheme is determined. Based on the optimal positioning reader setup scheme, multiple positioning readers are installed on the seal card cabinet.

4. The intelligent inventory management method for bank signature cards according to claim 1, characterized in that, Based on the usage data and storage location of signature cards from multiple companies in the current period, the optimal inventory counting path is determined, including: Based on the usage data and storage location of signature cards of multiple companies in the current period, determine the inventory priority of each company; Construct a second multi-objective fitness function, which is related to the inventory priority and the path length of the inventory path for each enterprise; The optimal inventory path is determined using the particle swarm optimization algorithm based on the second multi-objective fitness function.

5. The intelligent inventory management method for bank signature cards according to claim 1, characterized in that, Also includes: Identify multiple profile factors; Based on multiple profiling factors, construct enterprise profiles for multiple companies; Based on the corporate profiles of multiple enterprises and the historical usage data of corporate seal cards, several influential profile factors were identified. Based on multiple influencing profiling factors, construct a corporate profile for the company that stores its signature card for the first time. Based on the corporate profile of the company that is storing the signature card for the first time, the storage compartment for the company that is storing the signature card for the first time is determined.

6. The intelligent inventory management method for bank signature cards according to claim 5, characterized in that, Based on the corporate profile of the company that is storing its signature card for the first time, the storage compartment for that company is determined, including: Based on the corporate profile of the company that stores its signature card for the first time, predict the type of company that stores its signature card for the first time; If the type of enterprise that stores the signature card for the first time is predicted to be Category II, based on the enterprise profile of the enterprise that stores the signature card for the first time, the average frequency of signature card usage of the enterprise that stores the signature card for the first time is predicted. Based on the predicted average frequency of signature card usage of the enterprise that stores the signature card for the first time, the storage compartment of the enterprise that stores the signature card for the first time is determined. If the type of enterprise that stores the signature card for the first time is predicted to be of category one, based on the enterprise profile of the enterprise that stores the signature card for the first time, predict the average frequency of signature card use and the category to which the enterprise belongs. Based on the predicted average frequency of signature card use and the category to which the enterprise belongs, determine the storage compartment of the enterprise that stores the signature card for the first time.

7. An intelligent inventory management system for bank signature cards, characterized in that, The intelligent inventory management method for bank signature cards as described in claim 1 includes: The data acquisition module is used to acquire historical usage data of seal cards from multiple companies; The frequency analysis module is used to perform frequency analysis on the historical usage data of seal cards of multiple enterprises and generate frequency analysis results. The storage management module is used to determine the regional storage scheme for the signature cards of multiple enterprises based on the results of frequency matching analysis, and to assign RFID tags to the signature cards of each enterprise. The storage management module is also used to determine the storage location of the enterprise's seal cards based on the RFID tags and the RFID positioning components of the seal card cabinet after storing the seal cards of multiple enterprises in the seal card cabinet according to the regional storage scheme of multiple enterprise seal cards. The data acquisition module is also used to acquire the usage data and storage location of the seal cards of multiple enterprises in the current period; The intelligent inventory module is used to determine the optimal inventory path based on the usage data and storage location of the signature cards of multiple companies in the current period. Based on the optimal inventory path, it controls the RFID reader to perform inventory of the signature cards of multiple companies and generate automatic inventory results.

Images