A bank business intelligent flow transfer distribution method and system based on big data analysis

Abstract

The application discloses a bank business intelligent flow transfer distribution method and system based on big data analysis, relates to the field of data processing methods specially used for finance, and divides each network point into a first-level network point, a second-level network point and a third-level network point in the method by calculating a fund business safety coefficient and an actual available transaction limit of each network point. A business amount is split into a plurality of sub-business packages; candidate network points are determined, and a fund flow transfer path set is determined; a risk value is calculated, a path combination with the minimum risk value is selected, and an initial fund distribution scheme is generated; the fund processing pressure of each network point is calculated; when the fund processing pressure of any network point exceeds a preset threshold, a standby path is selected from the fund flow transfer path set of the network point, the sub-business package corresponding to the network point is re-distributed, and the fund processing pressure of all network points is lower than the preset threshold. The application realizes optimal allocation of cross-regional resources, and further improves the business processing efficiency and service quality of the bank.

Description

Technical Field

[0001] This application belongs to the field of data processing methods specifically applicable to finance, and in particular relates to a smart circulation and allocation method and system for banking business based on big data analysis. Background Technology

[0002] With the continuous increase in the types of banking services and the rapid growth in business volume, the traditional manual allocation model can no longer meet the needs of intelligent and efficient banking operations. Under the traditional model, business allocation mainly relies on human experience and fixed allocation rules. This approach is not only inefficient but also prone to problems such as uneven task allocation and waste of professional resources, thus reducing the processing efficiency of banking operations.

[0003] In related technologies, banking services can be automatically classified and allocated through preset business rules. The system allocates services to the appropriate personnel according to factors such as service type, priority, and the professional skills of the personnel handling the service, thereby improving the automation level and accuracy of service allocation.

[0004] However, when processing a large transaction, it may require the collaboration of personnel from multiple branches and different departments. The aforementioned technologies cannot comprehensively consider multiple factors such as the real-time business volume of each branch, the matching degree of personnel professionalism, and geographical distribution, thus making it difficult to achieve optimal allocation of cross-regional resources and reducing the bank's business processing efficiency and service quality. Summary of the Invention

[0005] This application provides a method and system for intelligent allocation of banking business based on big data analysis, which is used to achieve optimal allocation of cross-regional resources, thereby improving the efficiency of bank business processing and service quality.

[0006] Firstly, this application provides a smart transfer and allocation method for banking business based on big data analysis. The method calculates the security coefficient of each branch's fund business based on the historical processing data of large-value fund business at each branch. The historical processing data of large-value fund business includes the maximum limit, historical transaction success rate and large-value fund error rate. The security coefficient of fund business is obtained by weighted calculation of historical transaction success rate and large-value fund error rate.

[0007] The actual available transaction limit for each branch is calculated based on the maximum limit and the security coefficient of the funds business. Based on the actual available transaction limit of each branch, the branches are divided into first-level branches, second-level branches and third-level branches. The actual available transaction limit of the first-level branch is greater than that of the second-level branch, and the actual available transaction limit of the second-level branch is greater than that of the third-level branch.

[0008] Receive and extract the transaction amount, customer level, and original branch information from large-value transaction requests;

[0009] The transaction amount is divided into multiple sub-transaction packages, and the amount of each sub-transaction package is less than the actual available transaction limit of the first-level outlet.

[0010] Candidate outlets within a preset radius are determined with the original outlet as the center, and the risk path of fund transfer between each candidate outlet is calculated based on the fund business security coefficient of each candidate outlet, so as to obtain a set of fund transfer paths;

[0011] Calculate the risk value of each sub-business package in the set of fund transfer paths, select the path combination with the lowest risk value, and generate an initial fund allocation plan. The initial fund allocation plan includes the outlets and fund transfer paths corresponding to each sub-business package.

[0012] Based on the initial funding allocation plan, the funding processing pressure of each branch is calculated. The funding processing pressure is determined by the ratio of the amount of the sub-business package to the actual available transaction limit of the branch.

[0013] When the fund processing pressure of any branch exceeds the preset threshold, an alternative path is selected from the fund flow path set of the branch, and the sub-business package corresponding to the branch is redistributed until the fund processing pressure of all branches is lower than the preset threshold.

[0014] By adopting the above technical solutions, a scientific branch tiering mechanism is established by calculating the security coefficient of fund transactions through analysis of the historical transaction success rate and large-value fund error rate of branches, and determining the actual available transaction limit in conjunction with the maximum limit. When processing large-value fund transactions, the transaction amount is divided into sub-business packages adapted to the processing capabilities of different branch levels, and the optimal fund transfer scheme is selected based on the risk paths among candidate branches. A fund processing pressure index is introduced for dynamic adjustment. When the branch pressure is too high, the system automatically switches to backup paths to reallocate the business, improving the security and efficiency of large-value fund transactions. Through business splitting and multi-path diversion, the system can make full use of branch resources and avoid excessive pressure on a single point; through risk path optimization and dynamic adjustment mechanisms, efficient business transfer can be achieved while ensuring fund security, reducing operational and systemic risks in the process of processing large-value fund transactions.

[0015] In conjunction with some embodiments of the first aspect, in some embodiments, the actual available transaction limit of each branch is calculated based on the maximum limit of each branch and the fund transaction security coefficient, specifically including:

[0016] The transaction reliability of each branch is calculated based on the ratio of its historical transaction success rate to the preset benchmark success rate.

[0017] The ratio of the error rate of large-amount funds at each branch to the preset benchmark error rate is used as the risk attenuation coefficient for each branch.

[0018] The actual available transaction limit for each branch is calculated by multiplying the maximum limit of each branch by the corresponding transaction reliability and risk attenuation coefficient.

[0019] By adopting the above technical solution, the ratio of the branch's historical transaction success rate to a preset benchmark success rate is used as the transaction reliability, and the ratio of the large-amount fund error rate to a preset benchmark error rate is used as the risk attenuation coefficient. The product of the maximum limit and the transaction reliability and risk attenuation coefficient is used as the actual available transaction limit, ensuring that the branch's transaction capacity matches its actual business level. This allows the system to more accurately assess and control the business capacity of each branch. By introducing two dimensions—transaction reliability and risk attenuation coefficient—it reflects both the branch's transaction success capability and risk control level, resulting in a more reasonable and practical actual available transaction limit. This can reduce the risk of branches taking on business beyond their capacity.

[0020] In conjunction with some embodiments of the first aspect, in some embodiments, the risk path of fund transfer between candidate outlets is calculated based on the fund business security coefficient of each candidate outlet, resulting in a set of fund transfer paths, specifically including:

[0021] Calculate the difference in the security coefficient of fund transactions between each pair of candidate outlets, and use the difference in the security coefficient of fund transactions as the basic risk value for fund transfers between outlets;

[0022] A risk-weighted graph is constructed among candidate sites based on the basic risk value. In the risk-weighted graph, sites are nodes, and the basic risk value is the weight of the edge.

[0023] Establish a branch path mapping relationship based on the risk weighting diagram. The branch path mapping relationship includes the starting branch, intermediate branch and target branch.

[0024] An improved Dijkstra algorithm is used to search for multiple paths with the minimum risk weight in a risk-weighted graph, and the node sequence of the path with the minimum risk weight is converted into the corresponding network point sequence.

[0025] Based on the network sequence, the shortest physical path between adjacent network points is combined with the network path mapping relationship to generate the actual path of fund transfer;

[0026] The actual paths are classified and organized to obtain a set of fund transfer paths. The overlap between any two paths in the set of fund transfer paths does not exceed a preset ratio.

[0027] By adopting the above technical solution, a risk-weighted graph is constructed by calculating the difference in the security coefficient of fund transactions between candidate branches. An improved Dijkstra algorithm is used to search for multiple paths with the minimum risk weights. This is combined with physical paths to generate a set of actual fund transfer paths. The difference in business security between branches is used as the path weight, ensuring that path risk is minimized while also considering actual physical distance constraints. By controlling the overlap between any two paths in the path set to not exceed a preset ratio, the system's risk resistance is improved, the security of fund transfers is optimized, and the feasibility and diversity of paths are guaranteed.

[0028] In conjunction with some embodiments of the first aspect, in some embodiments, the risk value of each sub-business package in each path of the set of fund transfer paths is calculated, specifically including:

[0029] The basic risk coefficient of a path is obtained by summing the reciprocals of the security coefficients of the funds transactions of each point on each path in the set of fund transfer paths.

[0030] The initial risk value is obtained by multiplying the amount of each sub-business package by the basic risk coefficient of the corresponding optional path.

[0031] The probability of business congestion is calculated based on the historical business data of each branch, and the product of the probability of business congestion and the initial risk value is used as the final risk value of each path.

[0032] By adopting the above technical solution, a basic risk coefficient is obtained by summing the reciprocals of the fund transaction security coefficients of each branch along the path. This coefficient is then multiplied by the amount of the sub-business package to obtain an initial risk value. Finally, the path risk value is obtained by combining the probability of business congestion. This approach considers the security of the path itself, the impact of business scale, and the actual operational status of the branches. By combining static path risk with dynamic business pressure for comprehensive evaluation, the risk assessment results are more objective and accurate. This helps the system select the optimal solution from multiple alternative paths, improving the security of fund transfers.

[0033] In conjunction with some embodiments of the first aspect, in some embodiments, the probability of service congestion is calculated based on the historical service data of each branch, specifically including:

[0034] The business volume of each branch at different time periods is statistically analyzed, and the ratio of the business volume per unit time to the maximum processing capacity of the branch is calculated to obtain the business saturation.

[0035] Extract the peak and average business volume from the historical business data of each branch, and calculate the business volatility, which is the ratio of peak business volume to average business volume.

[0036] The backlog risk value is calculated based on business saturation and business volatility. The backlog risk value increases as business saturation and business volatility increase.

[0037] The standardized congestion index is obtained by dividing the business backlog risk value by the standard processing capacity of the branch.

[0038] Congestion level classification criteria are set based on a standardized congestion index, and the probability value corresponding to the current congestion level is used as the business congestion probability.

[0039] By adopting the above technical solution, and calculating the business volume of each branch at different time periods, and combining this with the peak and average business volumes from historical data to calculate the business volatility, the business processing status of each branch can be comprehensively reflected. By comprehensively considering business saturation and business volatility to calculate the business backlog risk value, the current business processing pressure of each branch can be accurately assessed. Dividing the business backlog risk value by the standard processing capacity of each branch yields a standardized congestion index, reducing the impact of differences between branches of different sizes. Based on the standardized congestion index, congestion level classification standards are set, and the probability of business congestion is determined, making the assessment of branch business congestion status more objective and accurate, and dynamically reflecting the actual business processing capacity and pressure status of each branch.

[0040] In conjunction with some embodiments of the first aspect, in some embodiments, after redistributing the sub-business packages corresponding to the outlets until the fund processing pressure of all outlets is lower than a preset threshold, the method further includes:

[0041] Calculate the capital utilization rate of each branch. The capital utilization rate is the ratio of the branch's average daily business capital to its capital reserves.

[0042] The fund early warning level is set based on the fund utilization rate. When the fund utilization rate is greater than the first preset value, it is the first warning. When the fund utilization rate is greater than the second preset value and less than the first preset value, it is the second warning. When the fund utilization rate is less than the second preset value, it is the third warning.

[0043] Establish mutual aid groups based on the geographical distance between each branch, with each mutual aid group containing at least three geographically adjacent branches;

[0044] When any branch is in the first warning level, funds will be allocated from the branches in the third warning level within the mutual aid fund group to which the branch belongs.

[0045] By adopting the above technical solution, a real-time monitoring mechanism for branch fund status was established by calculating the fund utilization rate of branches and setting a three-level fund early warning system. Fund mutual aid groups were established based on geographical distance. When a branch is found to be in the first early warning state, funds can be allocated from branches in the same mutual aid group that are in the third early warning state, selecting the branch with the lowest fund utilization rate. This achieves dynamic balance of fund reserves among branches and improves overall fund utilization efficiency. Since fund allocation only occurs between geographically adjacent branches, the time cost and security risks in the fund allocation process are reduced. By selecting the branch with the lowest fund utilization rate as the source of allocation, the feasibility of fund allocation is ensured, while avoiding impact on the normal business operations of the allocated branch, thus improving the overall fund utilization efficiency of the banking system.

[0046] In conjunction with some embodiments of the first aspect, in some embodiments, the branch with the lowest fund utilization rate is selected from the branches in the third warning zone within the mutual aid fund group to which the branch belongs for fund allocation, specifically including:

[0047] The fund utilization rate of each branch is calculated based on the average daily business fund volume and the branch's fund reserve volume within the mutual fund group.

[0048] The capital utilization rate is compared with the first and second preset values ​​to identify branches in the first warning state and branches in the third warning state.

[0049] Calculate the distance between each branch based on their geographical location, and determine at least three geographically adjacent branches within the same mutual aid group;

[0050] The network points in the third warning state were selected from the mutual aid groups and sorted from low to high according to the fund utilization rate;

[0051] Select the branch with the lowest capital utilization rate as the source branch for capital allocation, and allocate the funds to the target branch that is in the first warning state.

[0052] By adopting the above technical solution and specifying in detail the selection method for fund allocation outlets, the standardization and automation of the fund allocation process were achieved. Outlets were categorized based on fund utilization rates, and geographically based mutual aid groups were established, ensuring the accuracy and operability of fund allocation. When selecting source outlets for fund allocation, the optimal outlets were chosen by ranking them according to fund utilization rates, guaranteeing the rationality of fund allocation. This precise fund allocation mechanism not only responds quickly to the funding needs of outlets but also reduces the impact of fund allocation on the normal business operations of outlets. Because the allocation process considers the geographical distance between outlets, the actual execution efficiency of fund allocation is improved, while also reducing the risk of funds being in transit, thus increasing the efficiency and accuracy of fund allocation and making the use of funds throughout the entire bank's branch system more rational and efficient.

[0053] Secondly, embodiments of this application provide a smart banking business allocation system based on big data analysis. This smart banking business allocation system based on big data analysis includes: one or more processors and a memory; the memory is coupled to one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and one or more processors call the computer instructions to cause the system to perform the method described in the first aspect and any possible implementation of the first aspect.

[0054] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof.

[0055] Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the method described in any possible implementation of the first aspect.

[0056] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0057] 1. This application provides a smart transaction allocation method for banking operations based on big data analysis. It calculates a security coefficient for fund transactions by analyzing the historical transaction success rate and large-value error rate of branches, and determines the actual available transaction limit based on the maximum limit, establishing a scientific branch tiering mechanism. When processing large-value transactions, the transaction amount is divided into sub-transaction packages adapted to the processing capabilities of different branch levels, and the optimal fund transfer scheme is selected based on the risk paths among candidate branches. A fund processing pressure index is introduced for dynamic adjustment; when the branch pressure is too high, the system automatically switches to a backup path to reallocate the transaction, improving the security and efficiency of large-value transaction processing. Through transaction splitting and multi-path diversion, the system can fully utilize branch resources and avoid excessive pressure on single points; through risk path optimization and dynamic adjustment mechanisms, efficient transaction transfer can be achieved while ensuring fund security, reducing operational and systemic risks in the process of processing large-value transactions.

[0058] 2. This application provides a smart transaction allocation method for banking operations based on big data analysis. By statistically analyzing the transaction volume of each branch at different time periods and calculating the transaction saturation, and combining this with peak and average transaction volumes from historical data to calculate transaction volatility, the method can comprehensively reflect the business processing status of each branch. By comprehensively considering transaction saturation and transaction volatility to calculate the transaction backlog risk value, the method can accurately assess the current business processing pressure of each branch. A standardized congestion index is obtained by dividing the transaction backlog risk value by the standard processing capacity of each branch, reducing the impact of differences between branches of different sizes. Based on the standardized congestion index, a congestion level classification standard is set, and the probability of business congestion is determined, making the assessment of branch business congestion status more objective and accurate, and dynamically reflecting the actual business processing capacity and pressure status of each branch.

[0059] 3. This application provides a smart fund allocation method for banking operations based on big data analysis. By calculating the fund utilization rate of branches and setting a three-level fund warning system, a real-time monitoring mechanism for branch fund status is established. Fund mutual aid groups are established based on geographical distance. When a branch is found to be in the first warning state, funds can be allocated from branches in the same mutual aid group that are in the third warning state, selecting the branch with the lowest fund utilization rate. This achieves dynamic balance of fund surplus among branches and improves overall fund utilization efficiency. Since fund allocation only occurs between geographically adjacent branches, the time cost and security risks in the fund allocation process are reduced. By selecting the branch with the lowest fund utilization rate as the allocation source, the feasibility of fund allocation is ensured, and the impact on the normal business of the allocated branch is avoided, thereby improving the overall fund utilization efficiency of the banking system. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating a smart transfer and allocation method for banking business based on big data analysis, as described in an embodiment of this application.

[0061] Figure 2 This is a flowchart illustrating a fund early warning and allocation method in an embodiment of this application.

[0062] Figure 3 This is a schematic diagram of the physical device structure of a smart banking business transfer and allocation system based on big data analysis, provided in an embodiment of this application. Detailed Implementation

[0063] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0064] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0065] The following example is used in conjunction with Figure 1 This application describes a smart transfer and allocation method for banking business based on big data analysis in its embodiments:

[0066] Please see Figure 1 This is a flowchart illustrating a smart transfer and allocation method for banking business based on big data analysis, as described in this application.

[0067] S101. Calculate the security coefficient of fund transactions for each branch based on the historical data of large-value fund transactions at each branch.

[0068] The system calculates the security coefficient of each branch's large-value transaction history based on historical data from each branch. This historical data includes the maximum limit, historical transaction success rate, and large-value error rate. The security coefficient is calculated by weighting the historical transaction success rate and the large-value error rate. This step calculates the security coefficient for each branch's large-value transactions. The system obtains historical data from each branch, including the maximum limit, historical transaction success rate, and large-value error rate, and uses this data to calculate the security coefficient for each branch. The purpose of this step is to assess the security and reliability of each branch in handling large-value transactions, providing a basis for subsequent branch classification and fund allocation. In addition to the maximum limit, historical transaction success rate, and large-value error rate mentioned in the text, other relevant historical data, such as transaction processing time and customer complaint rate, can be considered to more comprehensively assess the security of each branch's large-value transactions.

[0069] In practical implementation, the system can use a weighted calculation method to determine the security factor for fund transactions. First, the system can assign different weights to historical transaction success rates and large-scale fund error rates, and then sum these weighted averages to obtain the final security factor. The weighting can be determined based on actual circumstances and business needs. For example, a higher weight can be given to historical transaction success rates, as they directly reflect the success rate of branch transactions; conversely, a higher weight can be given to large-scale fund error rates, as branches with high error rates may pose greater risks. Furthermore, the system can incorporate other factors, such as transaction volume and customer satisfaction, into the weighted calculation to obtain a more comprehensive and accurate security factor.

[0070] S102. Calculate the actual available transaction limit of each branch based on the maximum limit and the fund business security coefficient of each branch, and divide each branch into first-level branches, second-level branches and third-level branches according to the actual available transaction limit of each branch.

[0071] The system calculates the actual available transaction limit for each branch based on its maximum limit and fund security coefficient. Specifically, it calculates the transaction reliability of each branch by comparing its historical transaction success rate with a preset benchmark success rate; it uses the ratio of its large-amount fund error rate to a preset benchmark error rate as its risk attenuation coefficient; and it calculates the actual available transaction limit for each branch by multiplying its maximum limit by its corresponding transaction reliability and risk attenuation coefficient. Based on these actual available transaction limits, branches are categorized into Tier 1, Tier 2, and Tier 3 branches, with Tier 1 branches having a higher actual available transaction limit than Tier 2 branches, and Tier 2 branches having a higher actual available transaction limit than Tier 3 branches.

[0072] This step calculates the actual available transaction limit based on each branch's maximum limit and fund security coefficient, and then classifies the branches accordingly. The system first calculates transaction reliability based on the ratio of each branch's historical transaction success rate to a preset benchmark success rate, used to assess the branch's actual transaction success capability. Then, it uses the ratio of the large-amount fund error rate to the preset benchmark error rate as a risk attenuation coefficient to assess the branch's error risk. Finally, the maximum limit is multiplied by the transaction reliability and risk attenuation coefficient to obtain the actual available transaction limit. This limit comprehensively considers the branch's transaction capacity and risk level, more accurately reflecting its actual transaction capacity. Based on the size of the actual available transaction limit, the system divides branches into three levels; the higher the level, the larger the actual available transaction limit, allowing for the handling of larger transaction amounts. In addition to the maximum limit, transaction success rate, and error rate, other factors such as the branch's business volume and staffing can be considered when calculating the actual available transaction limit to more comprehensively assess the branch's actual transaction capacity.

[0073] In practical implementation, the system can set different thresholds to classify branch levels. For example, branches with the top 20% of actual available transaction limits can be classified as Level 1, those between 20% and 50% as Level 2, and the rest as Level 3. The thresholds can be determined based on actual circumstances and business needs, or automatically classified using algorithms such as clustering. Furthermore, the system can periodically update the actual available transaction limits and levels of branches to adapt to changes in business and branch capabilities. The level classification can also be more granular, for example, divided into five levels to better differentiate the transaction capabilities of branches.

[0074] S103. Receive and extract the transaction amount, customer level and original branch information from the large-amount fund transaction request, and split the transaction amount into multiple sub-business packages.

[0075] The system receives and extracts the transaction amount, customer level, and original branch information from large-amount transaction requests, and splits the transaction amount into multiple sub-transaction packages. The amount of each sub-transaction package is less than the actual available transaction limit of the first-level branch.

[0076] This step involves the system receiving large-value transaction requests and performing necessary information extraction and transaction breakdown. When a customer submits a large-value transaction request, the system first receives the request and extracts key information such as the transaction amount, customer level, and original branch. This information forms the basis for subsequent transaction processing. Due to the large transaction amount, to reduce the risk of fund processing, the system breaks down the transaction amount into multiple smaller sub-transaction packages, each with an amount not exceeding the actual available transaction limit of the first-level branch. This allows large transactions to be distributed across multiple branches for processing, reducing the financial pressure and risk for individual branches. In addition to splitting the amount, the system can also break down transaction requests based on other factors such as customer level and transaction type to achieve more granular transaction allocation and processing.

[0077] In practical implementation, the system can split the business amount into sub-business packages according to preset splitting rules. For example, the business amount can be split according to a fixed ratio or a fixed amount, or the splitting scheme can be dynamically adjusted according to the actual available transaction limit of the branch. During splitting, it is necessary to ensure that the sum of the amounts in the sub-business packages equals the original business amount to avoid mismatch issues. At the same time, to improve the efficiency of splitting, the system can pre-set splitting scheme templates and automatically select an appropriate splitting scheme based on the business amount. After splitting, the system will generate multiple sub-business packages, each containing relevant information about the original business request, such as customer level and original branch, for subsequent business allocation and processing.

[0078] S104. Determine candidate outlets within a preset radius centered on the original outlets, and calculate the risk path of fund transfer between candidate outlets based on the fund business security coefficient of each candidate outlet, to obtain a set of fund transfer paths.

[0079] The system determines candidate outlets within a preset radius centered on the original outlet, and calculates the risk path of fund transfer between each candidate outlet based on the fund business security coefficient of each candidate outlet, resulting in a set of fund transfer paths. Specifically, it calculates the difference in fund business security coefficient between each pair of candidate outlets, using this difference as the basic risk value for fund transfer between outlets. Based on the basic risk value, it constructs a risk weighted graph between candidate outlets, where outlets are nodes and the basic risk value is the edge weight. It establishes an outlet path mapping relationship based on the risk weighted graph, including the starting outlet, intermediate outlets, and target outlets. Using an improved Dijkstra algorithm, it searches for multiple paths with the minimum risk weight in the risk weighted graph and converts the node sequence of the path with the minimum risk weight into the corresponding outlet sequence. Based on the outlet sequence, it combines the shortest physical path between adjacent outlets with the outlet path mapping relationship to generate the actual fund transfer path. The actual paths are then classified and organized to obtain a set of fund transfer paths, where the overlap between any two paths in the set does not exceed a preset ratio.

[0080] This step involves identifying candidate outlets based on the original outlets and calculating the risk paths for fund transfers between them. The system uses the original outlet as the center and determines candidate outlets within a preset radius; these candidate outlets are those potentially involved in fund transfers. Then, the system calculates the risk paths for fund transfers between candidate outlets based on their fund security coefficients. Specifically, the system first calculates the difference in fund security coefficients between each pair of candidate outlets as a basic risk value. Then, based on this basic risk value, it constructs a risk-weighted graph between candidate outlets, where outlets are nodes and the basic risk value is the edge weight. Next, the system establishes an outlet path mapping relationship based on the risk-weighted graph, including starting outlets, intermediate outlets, and target outlets. Finally, the system uses an improved Dijkstra algorithm to search for multiple paths with the minimum risk weights in the risk-weighted graph, converts the node sequence of the path with the minimum risk weight into a corresponding outlet sequence, and then combines this with the shortest physical path between outlets to generate the actual fund transfer paths. These actual paths are then categorized and organized to obtain a set of fund transfer paths, where the overlap between any two paths in the set does not exceed a preset ratio. The purpose of this step is to identify the lowest-risk fund transfer path, providing a basis for subsequent fund allocation. Besides the methods mentioned in the text, other algorithms can be used to calculate risk paths, such as the Floyd algorithm and the A* algorithm.

[0081] In practical implementation, the system allows for flexible settings of the candidate branch selection range and path generation methods. The preset radius can be determined based on business needs and branch distribution; a larger radius results in more candidate branches and more generated paths. When constructing the risk-weighted graph, in addition to using the difference in financial business safety coefficients as edge weights, other factors such as distance between branches and business volume can be considered to comprehensively assess risk. When searching for the path with the minimum risk weight, the system can set restrictions on the number and length of paths according to actual needs to control the size of the path set. When generating actual paths, the system can consider factors such as physical distance between branches and traffic conditions to select the optimal physical path. When classifying and organizing actual paths, the system can use clustering algorithms such as K-means and DBSCAN to automatically group similar paths into one category.

[0082] When generating a set of fund transfer paths, issues such as an excessive number of paths and high redundancy may arise, increasing the complexity of subsequent fund allocation and impacting system performance. To address this, the system can set upper limits on the number of paths and redundancy. When the number of generated paths or the redundancy exceeds these limits, the system can automatically filter and optimize the paths. During filtering, path quality can be evaluated based on indicators such as risk weight, length, and redundancy, prioritizing the retention of higher-quality paths. During optimization, highly redundant paths can be merged, or redundant nodes and edges can be removed to reduce the number and complexity of paths. Through path filtering and optimization, a moderately sized and high-quality set of fund transfer paths can be obtained, improving the efficiency and accuracy of subsequent fund allocation.

[0083] S105. Calculate the risk value of each sub-business package in the set of fund transfer paths, select the path combination with the lowest risk value, and generate an initial fund allocation plan.

[0084] The system calculates the risk value of each sub-business package across all paths in the set of fund transfer paths. Specifically, this includes: calculating the sum of the reciprocals of the fund transaction security coefficients of each branch along each path to obtain the basic risk coefficient for that path; multiplying the amount of each sub-business package by the basic risk coefficient of the corresponding optional path to obtain the initial risk value; calculating the business congestion probability based on the historical business data of each branch, and using the product of the business congestion probability and the initial risk value as the final risk value for each path. The calculation of the business congestion probability based on the historical business data of each branch specifically includes: statistically analyzing the business processing at each branch during different time periods. The system calculates the business volume and the ratio of the business volume processed per unit time to the maximum processing capacity of each branch to obtain the business saturation. It extracts the peak and average business volumes from the historical business data of each branch and calculates the business volatility, which is the ratio of peak to average business volume. Based on the business saturation and volatility, it calculates the business backlog risk value, which increases with both factors. The standardized congestion index is obtained by dividing the business backlog risk value by the standard processing capacity of each branch. A congestion level classification standard is set based on the standardized congestion index, and the probability value corresponding to the current congestion level is used as the business congestion probability. The system then selects the path combination with the lowest risk value and generates an initial funding allocation plan, which includes the branch and funding flow path corresponding to each sub-business package.

[0085] This step further details how to calculate the risk value of each sub-business package within the set of fund transfer paths. A crucial detail is the calculation of the business congestion probability. The system first statistically analyzes the business volume of each branch at different time periods, calculating the ratio of the business volume per unit time to the branch's maximum processing capacity to obtain the business saturation, reflecting the branch's workload. Then, the system extracts the peak and average business volumes from each branch's historical business data, calculating the business volatility, i.e., the ratio of peak to average business volume, reflecting the branch's business volume fluctuations. Next, the system calculates the business backlog risk value based on the business saturation and business volatility. The business backlog risk value increases with increasing business saturation and business volatility, indicating that the busier the branch and the greater the business volume fluctuation, the higher the backlog risk. To facilitate comparison of backlog risks across different branches, the system divides the business backlog risk value by the branch's standard processing capacity to obtain a standardized congestion index. Finally, the system sets a congestion level classification standard based on the standardized congestion index, using the probability value corresponding to the current congestion level as the business congestion probability. This calculation method fully considers the dynamic changes in branch traffic and the risk of backlog, and can more accurately assess the congestion of the route.

[0086] In calculating the probability of business congestion, the system can employ various data processing and analysis techniques. When calculating business volume, the system can use a sliding time window to dynamically calculate the volume within a certain time range, reflecting real-time changes in volume. When calculating business saturation, the system can set different maximum processing capacity thresholds based on the business type and processing capacity of each branch, to more accurately assess branch saturation. When calculating business volatility, the system can use more complex statistical indicators, such as standard deviation and coefficient of variation, to more comprehensively characterize the volatility of business volume. When calculating the risk value of business backlog, the system can use different risk functions, such as linear and exponential functions, to accommodate different risk preferences. When setting congestion level classification criteria, the system can use clustering algorithms to automatically classify standardized congestion indices into different levels and calculate the historical congestion probability for each level. Furthermore, the system can introduce machine learning algorithms, such as decision trees and random forests, to automatically learn and optimize the congestion probability calculation model based on historical data, improving the accuracy and adaptability of the calculation.

[0087] S106. Calculate the cash handling pressure of each branch based on the initial cash allocation plan;

[0088] The system calculates the fund processing pressure for each branch based on the initial fund allocation plan. The fund processing pressure is determined by the ratio of the amount of the sub-business package to the actual available transaction limit of the branch.

[0089] This step calculates the processing pressure for each branch based on the initial funding allocation plan. Processing pressure reflects the difficulty and risk of a branch handling the allocated funding transactions; higher pressure indicates a strained processing capacity and greater risk. The system calculates processing pressure by dividing the sum of the sub-business package amounts allocated to each branch by the branch's actual available transaction limit. A higher ratio indicates greater processing pressure for the branch. The purpose of calculating processing pressure is to assess the feasibility and balance of the initial funding allocation plan, providing a basis for subsequent fund reallocation. Besides using the ratio of sub-business package amounts to the actual available transaction limit to measure pressure, other factors such as the branch's historical business volume and staffing levels can be considered to more comprehensively evaluate the branch's processing capacity.

[0090] In practical implementation, the system compares the fund processing pressure with a preset pressure threshold to determine whether a branch can handle the allocated fund transactions. The pressure threshold can be set based on business needs and risk appetite; a lower threshold indicates a higher requirement for the branch's processing capacity and a lower risk tolerance. When a branch's fund processing pressure exceeds the threshold, the fund allocation plan needs to be adjusted to reduce the branch's pressure. Adjustments can be made by reallocating some sub-business packages to other branches or adjusting the branch's actual available transaction limit. The system can determine the priority and magnitude of adjustments based on the magnitude of the fund processing pressure; branches with greater pressure receive higher priority and larger adjustments.

[0091] When calculating fund processing pressure, some branches may experience significantly higher pressure than others, leading to an imbalance in fund allocation and impacting overall processing efficiency and risk control. To address this, the system can introduce a pressure balancing mechanism. This mechanism aims to bring the pressure of each branch as close as possible to the average, while ensuring that the fund processing pressure of each branch remains within a threshold. Specifically, the system calculates the mean and variance of branch pressure. When the variance exceeds a certain range, branches with higher pressure are prioritized for adjustment, reducing their pressure to near the mean. This adjustment can be achieved by transferring their assigned sub-business packages to branches with lower pressure or by increasing their actual available transaction limits. Through this pressure balancing mechanism, fund allocation becomes more balanced, improving overall branch processing efficiency and reducing single-point risk.

[0092] S107. When the fund processing pressure of any branch exceeds the preset threshold, select an alternate path from the fund flow path set of the branch, and redistribute the sub-business package corresponding to the branch until the fund processing pressure of all branches is lower than the preset threshold.

[0093] This step dynamically adjusts the fund allocation scheme based on the fund processing pressure of each branch. When the system detects that the fund processing pressure of a branch exceeds a preset threshold, it indicates that the current fund allocation scheme places excessive demands on the branch's processing capacity and needs adjustment. The adjustment involves selecting backup paths from the branch's fund flow path set and then reallocating some of the sub-business packages allocated to the branch to these backup paths to reduce the branch's fund processing pressure. The number and amount of reassigned sub-business packages can be determined based on the degree to which the threshold is exceeded; the greater the exceedance, the more sub-business packages need to be reassigned. Adjustments continue until the fund processing pressure of all branches is reduced below the threshold. The purpose of this step is to dynamically optimize the fund allocation scheme, ensuring that the processing pressure of each branch is within an acceptable range. Besides backup paths, other adjustment methods can be considered, such as increasing the actual available transaction limit of branches and reallocating personnel between branches, to improve the branch's processing capacity.

[0094] In practical implementation, the system needs to pre-generate a set of fund transfer paths for each branch as a source of backup paths. Backup paths can be paths not selected in the initial fund allocation plan, or new paths dynamically generated based on real-time network conditions and business needs. When selecting backup paths, the system needs to comprehensively consider factors such as the path's risk value, the branch's processing capacity, and the timeliness of the business, prioritizing paths with lower risk, stronger branch processing capacity, and lower timeliness requirements. After selecting a backup path, the system sends the information of the sub-business package to the relevant branches along the path, notifying them to process it. Simultaneously, the system also needs to update the branch's fund processing pressure to promptly detect situations where pressure exceeds limits.

[0095] In the above embodiments, a scientific branch tiering mechanism is established by calculating the fund transaction security coefficient by analyzing the historical transaction success rate and large-value fund error rate of the branches, and determining the actual available transaction limit in conjunction with the maximum limit. When processing large-value fund transactions, the transaction amount is divided into sub-business packages adapted to the processing capabilities of different branch levels, and the optimal fund transfer scheme is selected based on the risk paths among candidate branches. A fund processing pressure index is introduced for dynamic adjustment. When the branch pressure is too high, the backup path is automatically switched to reallocate the business, improving the security and efficiency of large-value fund transactions. Through business splitting and multi-path diversion, the system can make full use of branch resources and avoid excessive pressure on a single point; through risk path optimization and dynamic adjustment mechanisms, efficient business transfer can be achieved while ensuring fund security, reducing operational and systemic risks in the process of processing large-value fund transactions.

[0096] The above embodiments describe the basic process of an intelligent flow allocation method for banking business based on big data analysis. This method achieves safe and efficient processing of large-value transactions through methods such as calculating the security coefficient of branch fund transactions, business segmentation, and risk path optimization. However, to further improve the system's operating efficiency and ensure the rational utilization of fund reserves in each branch, this application also provides a fund early warning and allocation method. This method monitors the fund utilization status of branches in real time and makes timely allocations when fund pressure occurs, thereby maintaining the fund balance of the entire bank branch system. The following is a combination of... Figure 2 The following describes a method for early warning and allocation of funds in an embodiment of this application:

[0097] Please see Figure 2 This is a flowchart illustrating a fund early warning and allocation method in an embodiment of this application.

[0098] S201. Calculate the capital utilization rate of each branch;

[0099] The system calculates the capital utilization rate of each branch, which is the ratio of the branch's average daily business capital to its capital reserves.

[0100] In this step, the system calculates the capital utilization rate of each branch to understand how its funds are used. Capital utilization rate reflects the efficiency of fund utilization at each branch and is a crucial indicator for measuring its financial operations. The system can calculate capital utilization rate in various ways, such as using the ratio of average daily business funds to branch cash reserves, or turnover days. Furthermore, the system can analyze and compare capital utilization rates based on different periods and business types to comprehensively assess the branch's capital utilization.

[0101] Specifically, the system calculates capital utilization rate as follows: First, it obtains the average daily business capital volume and branch cash reserves data for each branch; then, it divides the average daily business capital volume by the branch cash reserves to obtain the capital utilization rate. The average daily business capital volume can be obtained by statistically analyzing the total business capital volume of the branch over a certain period (e.g., one month) and then dividing it by the number of days in that period. Branch cash reserves refer to the total amount of funds available for business operations at the branch, including cash on hand and funds in transit.

[0102] S202. Set early warning levels for funds based on fund utilization rate;

[0103] The system sets early warning levels based on the capital utilization rate. When the capital utilization rate is greater than the first preset value, it is the first early warning; when the capital utilization rate is greater than the second preset value but less than the first preset value, it is the second early warning; and when the capital utilization rate is less than the second preset value, it is the third early warning.

[0104] In this step, the system sets different early warning levels based on the calculated capital utilization rate to reflect the health of the branch's capital situation. Setting these early warning levels helps managers promptly identify problems in capital utilization and take corresponding measures for adjustment and optimization. The system can set multiple early warning levels according to actual conditions, such as normal, attention, warning, and high risk, with different levels corresponding to different capital utilization rate thresholds.

[0105] In practice, the system can preset two capital utilization thresholds: a first preset value and a second preset value. When the capital utilization rate is greater than the first preset value, it is considered that the capital utilization rate is too high, and the branch's funds are under considerable pressure, so the warning level is set to the first warning level. When the capital utilization rate is between the first and second preset values, it is considered that the capital utilization rate is relatively high, and the branch's funds are under some pressure, so the warning level is set to the second warning level. When the capital utilization rate is lower than the second preset value, it is considered that the capital utilization rate is normal, and the branch's funds are relatively abundant, so the warning level is set to the third warning level.

[0106] When setting early warning levels, the system may encounter the problem of unreasonable preset values, leading to distorted warning levels that fail to accurately reflect the financial status of branches. To address this, the system can employ a dynamic adjustment mechanism, automatically optimizing preset values ​​based on historical data and trends in branch fund utilization rates. For example, the system can calculate the average and standard deviation of fund utilization rates for all branches over a certain period, using the average plus or minus different multiples of the standard deviation as the preset value. Furthermore, the system can set a fluctuation range for the preset values, allowing for timely adjustments when actual conditions change, thus maintaining the effectiveness of the early warning levels.

[0107] S203. Establish mutual aid groups based on the geographical distance between each branch;

[0108] The system establishes mutual aid groups based on the geographical distance between each branch, and each mutual aid group contains at least three geographically adjacent branches.

[0109] In this step, the system groups nearby branches into mutual aid groups based on their geographical location, providing a foundation for subsequent fund allocation. The establishment of these mutual aid groups facilitates fund transfers between branches and improves the efficiency of fund utilization. The system can use various methods to divide mutual aid groups, such as grouping them based on factors like administrative divisions or economic zones where the branches are located.

[0110] In practice, the system can obtain the geographical coordinates of each location, calculate the distance between them, and group locations with a distance less than or equal to a certain threshold (e.g., 50 kilometers) into the same mutual aid group. To ensure the feasibility and effectiveness of the mutual aid, the system can require each mutual aid group to contain at least three locations. When calculating the distance between locations, the system can use various distance metrics, such as Euclidean distance and Manhattan distance, and can also consider factors such as road distance and travel time.

[0111] When establishing mutual aid groups, the system may face problems such as uneven distribution of service points and large differences in group size, affecting the effectiveness of mutual aid. To address this, the system can employ a dynamic adjustment mechanism, adjusting the group division method and parameter settings in a timely manner based on changes in service point distribution. For example, when a new service point is added in a certain area, the system can recalculate the distance between service points and assign the new service point to the nearest group; when the number of service points in a group is too large or too small, the system can adjust the distance threshold to split or merge groups, maintaining a relatively balanced group size.

[0112] S204. When any branch is under the first warning, select the branch with the lowest fund utilization rate from the branches under the third warning in the fund mutual aid group to which the branch belongs for fund allocation.

[0113] When any branch is in the first warning state, funds are allocated from the branches in the third warning state within the mutual aid group to which the branch belongs. Specifically, the fund utilization rate of each branch is calculated based on the average daily business fund volume and the branch's fund reserve in the mutual aid group; the fund utilization rate is compared with the first and second preset values ​​to identify branches in the first warning state and branches in the third warning state; the distance between branches is calculated based on their geographical location to determine at least three geographically adjacent branches within the same mutual aid group; branches in the third warning state are screened from the mutual aid group and sorted from low to high fund utilization rate; the branch with the lowest fund utilization rate is selected as the source branch for fund allocation, and funds are allocated to the target branch in the first warning state.

[0114] In this step, the system allocates funds based on the branch's funding warning level and its affiliated mutual aid group to alleviate financial pressure. When a branch triggers the first warning, it indicates that the branch is experiencing a funding shortage and needs to draw funds from other branches. The system first filters out branches with a warning level of the third warning within the mutual aid group to which the branch belongs. These branches have relatively abundant funds and are capable of providing financial support. Then, the system selects the branch with the lowest fund utilization rate as the fund supplier and allocates funds to the branch experiencing a funding shortage. This allocation method ensures an adequate supply of funds while minimizing the impact on the fund supplier.

[0115] In practice, the system can execute the following steps: First, obtain the real-time fund utilization rate and warning level of each branch within the mutual aid group; then, identify the branch with the highest warning level (Level 1) as the target branch for fund allocation; next, within the group to which this branch belongs, select the branch with the highest warning level (Level 3) as the alternative fund supplier; finally, sort the alternative branches according to their fund utilization rate from low to high, select the branch with the lowest fund utilization rate as the final fund supplier, calculate the amount of funds to be allocated, and complete the fund allocation. The amount of funds allocated can be determined comprehensively based on factors such as the target branch's funding gap and the supplier's surplus funds.

[0116] In the above embodiments, a real-time monitoring mechanism for branch fund status was established by calculating the fund utilization rate of branches and setting a three-level fund warning level. Fund mutual aid groups were established based on geographical distance. When a branch is found to be in the first warning state, funds can be allocated from branches in the same mutual aid group that are in the third warning state, selecting the branch with the lowest fund utilization rate. This achieves dynamic balance of fund reserves among branches and improves overall fund utilization efficiency. Since fund allocation only occurs between geographically adjacent branches, the time cost and security risks in the fund allocation process are reduced. By selecting the branch with the lowest fund utilization rate as the source of allocation, the feasibility of fund allocation is ensured, while avoiding impact on the normal business operations of the allocated branch, thus improving the overall fund utilization efficiency of the banking system.

[0117] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a smart banking business transfer and allocation system based on big data analysis, provided in an embodiment of this application.

[0118] It should be noted that, Figure 3 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0119] like Figure 3 As shown, the system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 302 or a program loaded from storage portion 308 into Random Access Memory (RAM) 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0120] The following components are connected to I / O interface 305: input section 306 including a camera, infrared sensor, etc.; output section 307 including a liquid crystal display (LCD) and speakers, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0121] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0122] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0124] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0125] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0126] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0127] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0128] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A smart transaction allocation method for banking operations based on big data analytics, characterized in that, include: The security coefficient of fund transactions for each branch is calculated based on the historical processing data of large-value fund transactions for each branch. The historical processing data of large-value fund transactions includes the maximum limit, historical transaction success rate and large-value fund error rate. The security coefficient of fund transactions is obtained by weighted calculation of the historical transaction success rate and the large-value fund error rate. The actual available transaction limit of each branch is calculated based on the maximum limit and the fund business security coefficient of each branch. The branches are then divided into first-level branches, second-level branches, and third-level branches according to the actual available transaction limit of each branch. The actual available transaction limit of the first-level branch is greater than that of the second-level branch, and the actual available transaction limit of the second-level branch is greater than that of the third-level branch. Receive and extract the transaction amount, customer level, and original branch information from large-value transaction requests; The transaction amount is divided into multiple sub-transaction packages, and the amount of each sub-transaction package is less than the actual available transaction limit of the first-level outlet. Candidate outlets within a preset radius are determined with the original outlet as the center, and the risk path of fund transfer between the candidate outlets is calculated based on the fund business security coefficient of each candidate outlet, so as to obtain a set of fund transfer paths; Calculate the risk value of each of the sub-business packages in the set of fund transfer paths, select the path combination with the lowest risk value, and generate an initial fund allocation plan. The initial fund allocation plan includes the branch and fund transfer path corresponding to each sub-business package. The risk value of each path is obtained by the fund business security coefficient of each branch on the path, the amount of the sub-business package, and the business congestion probability obtained based on the historical business data of each branch. According to the initial fund allocation plan, the fund processing pressure of each of the outlets is calculated, and the fund processing pressure is determined based on the ratio of the amount of the sub-business package to the actual available transaction limit of the outlet. When the fund processing pressure of any of the outlets exceeds a preset threshold, an alternate path is selected from the set of fund transfer paths of the outlets, and the sub-business packages corresponding to the outlets are redistributed until the fund processing pressure of all the outlets is lower than the preset threshold.

2. The method according to claim 1, characterized in that, The calculation of the actual available transaction limit for each branch based on the maximum limit and the fund security coefficient of each branch specifically includes: The transaction reliability of each branch is calculated based on the ratio of its historical transaction success rate to a preset benchmark success rate. The ratio of the large-amount fund error rate of each branch to the preset benchmark error rate is used as the risk attenuation coefficient of each branch. The product of the maximum limit of each branch and the corresponding transaction reliability and risk attenuation coefficient is taken as the actual available transaction limit of each branch.

3. The method according to claim 1, characterized in that, The process of calculating the risk path of fund transfer between the candidate outlets based on the fund business security coefficient of each candidate outlet, and obtaining a set of fund transfer paths, specifically includes: Calculate the difference in the security coefficient of fund transactions between each pair of the candidate outlets, and use the difference in the security coefficient of fund transactions as the basic risk value of fund transfer between the outlets; A risk-weighted graph is constructed among candidate sites based on the basic risk value, where the sites are nodes and the basic risk value is the weight of the edges. Establish a branch path mapping relationship based on the risk weighting diagram, wherein the branch path mapping relationship includes the starting branch, intermediate branch and target branch; The improved Dijkstra algorithm is used to search for multiple paths with the minimum risk weight in the risk-weighted graph, and the node sequence of the path with the minimum risk weight is converted into the corresponding network point sequence. Based on the network sequence, the shortest physical path between adjacent network points is combined with the network path mapping relationship to generate the actual path of fund transfer; The actual paths are classified and organized to obtain a set of fund transfer paths, and the overlap between any two paths in the set of fund transfer paths does not exceed a preset ratio.

4. The method according to claim 1, characterized in that, The calculation of the risk value of each of the sub-business packages in each path of the set of fund transfer paths specifically includes: The basic risk coefficient of a path is obtained by summing the reciprocals of the security coefficients of the funds transactions of each point on each path in the set of fund transfer paths. The amount of each sub-business package is multiplied by the basic risk coefficient of the corresponding optional path to obtain the initial risk value; The probability of service congestion is calculated based on the historical service data of each of the aforementioned outlets, and the product of the probability of service congestion and the initial risk value is used as the final risk value of each path.

5. The method according to claim 4, characterized in that, The calculation of the service congestion probability based on the historical service data of each of the aforementioned outlets specifically includes: The business volume of each of the aforementioned outlets is statistically analyzed at different time periods, and the ratio of the business volume per unit time to the maximum processing capacity of the outlets is calculated to obtain the business saturation. Extract the peak and average business volume from the historical business data of each of the network points, and calculate the business volatility, wherein the business volatility is the ratio of the peak business volume to the average business volume; The business backlog risk value is calculated based on the business saturation and the business volatility, and the business backlog risk value increases as the business saturation and business volatility increase; The standardized congestion index is obtained by dividing the business backlog risk value by the standard processing capacity of the branch. Based on the standardized congestion index, a congestion level classification standard is set, and the probability value corresponding to the current congestion level is used as the business congestion probability.

6. The method according to claim 1, characterized in that, After the sub-business packages corresponding to the outlets are reallocated until the fund processing pressure of all the outlets is lower than the preset threshold, the method further includes: Calculate the capital utilization rate of each of the aforementioned outlets, wherein the capital utilization rate is the ratio of the average daily business capital of the outlet to the outlet's capital reserve. Based on the capital utilization rate, a capital early warning level is set. When the capital utilization rate is greater than a first preset value, it is a first early warning. When the capital utilization rate is greater than a second preset value and less than the first preset value, it is a second early warning. When the capital utilization rate is less than the second preset value, it is a third early warning. A mutual aid group is established based on the geographical distance between the locations, and each mutual aid group contains at least three geographically adjacent locations; When any branch is under the first warning, the branch with the lowest fund utilization rate is selected from the branches under the third warning in the fund mutual aid group to which the branch belongs for fund allocation.

7. The method according to claim 6, characterized in that, The step of selecting the branch with the lowest fund utilization rate from the mutual aid fund groups to which the branch belongs and allocating funds specifically includes: The fund utilization rate of each of the aforementioned outlets is calculated based on the average daily business fund volume and the outlet's fund reserve volume within the aforementioned mutual aid group. The fund utilization rate is compared with the first preset value and the second preset value to identify the outlets in the first warning state and the outlets in the third warning state. Calculate the distance between each of the aforementioned outlets based on their geographical locations, and determine at least three geographically adjacent outlets within the same mutual aid fund group; The branches in the third early warning state are selected from the mutual aid fund groups and sorted from low to high according to the fund utilization rate; Select the branch with the lowest capital utilization rate as the source branch for capital allocation, and allocate the funds to the target branch that is in the first warning state.

8. A smart transaction allocation system for banking operations based on big data analytics, characterized in that: The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-7.

Images