Financial equipment maintenance method and equipment management system

Abstract

The invention discloses a financial equipment maintenance method and an equipment management system. According to the method, each piece of financial equipment analyzes local operation data to generate equipment state index data; the edge gateway node analyzes the equipment state index data from the financial equipment in the management range to generate maintenance request information; the dispatching center receives maintenance request information from each edge gateway node in the jurisdiction area; evaluating each piece of maintenance request information based on each preset evaluation dimension of the dynamic adjustment weight, and determining a global priority score of each piece of maintenance request information; screening out candidate maintenance request information according to the global priority score, performing correlation analysis on the candidate maintenance request information, and identifying a correlation type between the candidate maintenance request information; and according to the global priority score, the association type and the available resource state in the jurisdiction area of the candidate maintenance request information, generating a scheduling scheme to realize equipment maintenance. According to the technical scheme, the decision response efficiency and the resource scheduling rationality are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology, and in particular to a method for maintaining financial equipment and an equipment management system. Background Technology

[0002] As commercial banks become increasingly digital and intelligent, key financial equipment such as ATMs, VTMs, and bill processing terminals have become core pillars of branch operations. Ensuring the stable operation of these devices is crucial for maintaining customer service continuity, preserving the bank's reputation, and mitigating operational risks.

[0003] The current banking equipment maintenance system relies excessively on centralized cloud processing and lacks intelligent collaborative capabilities at the edge, resulting in low decision-making and response efficiency. Furthermore, the simple priority decision-making rules of the current banking equipment maintenance system lead to unscientific resource scheduling. Summary of the Invention

[0004] This invention provides a method for maintaining financial equipment and an equipment management system to improve decision-making response efficiency and the rationality of resource scheduling.

[0005] According to one aspect of the present invention, a financial equipment maintenance method is provided, applied to an equipment management system, the equipment management system including a dispatch center, edge gateway nodes, and financial equipment managed by each edge gateway node, the financial equipment maintenance method comprising:

[0006] Each financial device analyzes its own operating data to generate device status indicator data.

[0007] The edge gateway node analyzes the device status indicator data from various financial devices within its management scope to generate maintenance request information.

[0008] The dispatch center receives maintenance request information from each edge gateway node within its jurisdiction; evaluates each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filters candidate maintenance request information based on the global priority score and performs correlation analysis on the candidate maintenance request information to identify the correlation types between the candidate maintenance request information; and generates a dispatch plan to achieve equipment maintenance based on the global priority score of the candidate maintenance request information, the correlation type, and the available resource status within the jurisdiction.

[0009] According to another aspect of the present invention, a device management system is provided, the device management system comprising a scheduling center, edge gateway nodes, and financial devices managed by each edge gateway node, the device management system comprising:

[0010] Each of the aforementioned financial devices is used to analyze its own operating data to generate device status indicator data;

[0011] The edge gateway node is used to analyze the device status indicator data from various financial devices within the management scope to generate maintenance request information;

[0012] The scheduling center is used to receive maintenance request information from each edge gateway node within its jurisdiction; evaluate each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filter candidate maintenance request information based on the global priority score and perform correlation analysis on the candidate maintenance request information to identify the correlation types between the candidate maintenance request information; and generate a scheduling scheme to achieve equipment maintenance based on the global priority score of the candidate maintenance request information, the correlation type, and the available resource status within the jurisdiction.

[0013] The technical solution of this invention involves each financial device analyzing its local operating data to generate device status index data; edge gateway nodes analyzing device status index data from each financial device within their management scope to generate maintenance request information; a dispatch center receiving maintenance request information from each edge gateway node within its jurisdiction; evaluating each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filtering candidate maintenance request information based on the global priority score and performing correlation analysis on the candidate maintenance request information to identify the correlation types between them; and generating a scheduling scheme to achieve device maintenance based on the global priority score, correlation type, and available resource status within the jurisdiction of the candidate maintenance request information. This solution employs a combination of edge-layer collaborative processing, a multi-dimensional evaluation mechanism, and analysis of the correlations between maintenance tasks to generate a scheduling scheme. This addresses the current banking device maintenance system's over-reliance on centralized cloud processing and lack of intelligent edge collaboration capabilities, resulting in low decision-making efficiency. It also addresses the problem of unscientific resource scheduling caused by the simple priority decision-making rules in the current banking device maintenance system, thereby improving decision-making efficiency and the rationality of resource scheduling.

[0014] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart of a financial equipment maintenance method provided in Embodiment 1 of the present invention;

[0017] Figure 2 This is a schematic diagram of the structure of an equipment management system provided in Embodiment 2 of the present invention;

[0018] Figure 3 This is a schematic diagram of the structure of an electronic device that implements the financial equipment maintenance method of this invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] Example 1

[0022] Figure 1 This is a flowchart illustrating a financial equipment maintenance method according to Embodiment 1 of the present invention. This embodiment is applicable to situations involving the maintenance and management of financial equipment. The method can be executed by an equipment management system, which may include a dispatch center, various edge gateway nodes, and the financial equipment managed by each edge gateway node. Figure 1 As shown, the method includes:

[0023] S110. Each financial device analyzes its own operating data to generate device status indicator data.

[0024] The local operating data refers to data generated during the operation of the financial equipment itself. This local operating data may include at least one of the following: equipment operating temperature, equipment vibration data, equipment operating current data, and equipment transaction monitoring data. Edge intelligent terminals can be deployed on each financial device. These edge intelligent terminals can integrate multiple types of sensors (such as temperature sensors, vibration sensors, current monitors, and transaction anomaly monitoring modules) to collect the financial equipment's operating data in real time.

[0025] Equipment status index data refers to various status indicators generated by financial equipment during operation. These indicators can include the predicted failure probability, abnormal change rate (e.g., temperature rise per minute, vibration intensity fluctuation), environmental correlation indicators (e.g., current harmonic distortion rate, network latency spikes), and transaction anomaly rate for a specified future period. Equipment status index data can be generated by analyzing and collecting local operating data from edge intelligent terminals deployed within the financial equipment itself.

[0026] In one optional implementation, each financial device analyzes its own operating data to generate device status index data, which may include: the edge intelligent terminal of the current financial device collecting its own operating data in real time; the current financial device using a predictive model to analyze its own operating data and predict the probability of failure of the current financial device in a future specified period; and the current financial device generating device status index data based on its own operating data and the probability of failure.

[0027] In this embodiment, the edge intelligent terminal of the financial device can run a predictive model to analyze its own operating data and predict the probability of failure of the current financial device within a specified future time period, such as the probability of failure occurring within the next 30 minutes. The edge intelligent terminal can generate device status index data of the current financial device based on its own operating data and the failure probability.

[0028] By enabling local intelligent analysis of device status indicators on financial devices, this setup can distribute the computational burden of centralized cloud processing.

[0029] S120: The edge gateway node analyzes the device status indicator data from various financial devices within its management scope to generate maintenance request information.

[0030] The maintenance request information can be information determined by the aggregated analysis data of the edge gateway node, and can include equipment information and fault description information of the device to be maintained.

[0031] In one optional implementation, the edge gateway node analyzes device status indicator data from various financial devices within its management scope to generate maintenance request information, which may include:

[0032] The current edge gateway node receives device status indicator data from various financial devices within its management scope. Based on a specified time window, the current edge gateway node aggregates and analyzes the device status indicator data, obtaining a first type of dataset and a second type of dataset. The first type of dataset is used to statistically analyze the distribution of various device status indicators for each financial device managed by the current edge gateway node within the specified time window. The second type of dataset records the fluctuation of each device status indicator for each financial device within a specified historical period. Based on the first and second types of datasets, the current edge gateway node analyzes the regional cascading failure risk within its management scope using a three-layer progressive method. This three-layer progressive method includes abnormal device group detection, abnormal device group dependency detection, and abnormal persistence detection. If the regional cascading failure risk analysis results meet the maintenance conditions, the current edge gateway node generates corresponding maintenance request information.

[0033] In this embodiment, multiple edge intelligent terminals (financial devices) deployed in adjacent physical locations (e.g., the same bank branch) can exchange key operational statuses in real time through a low-power wide-area network (LPWAN) ad hoc network. Each financial device can broadcast lightweight data packets containing device status indicator data. Furthermore, an edge gateway node deployed in the same branch can continuously receive broadcast data (i.e., lightweight data packets containing device status indicator data) from each financial device within the network, and perform aggregation analysis on the received broadcast data based on a specified time window (e.g., 30 seconds) to obtain a first type of dataset and a second type of dataset. The first type of dataset can be a horizontal comparison dataset, used to statistically analyze the distribution of various device status indicators for each financial device managed by the current edge gateway node within a specified time window, such as the temperature rise rate distribution of all ATMs (Automatic Teller Machines). The second type of dataset can be a historical trend dataset, used to record the fluctuation of each device status indicator for each financial device within a specified historical period, such as recording the fluctuation of the failure probability of the same device over the past 24 hours.

[0034] In this embodiment, the risk of regional cascading failures refers to the risk of cascading failures caused by the failure of one or more devices within the management scope of the edge gateway node. Maintenance conditions may include, for example, a single device's predicted failure probability exceeding a preset threshold or a regional cascading failure risk exceeding a safety threshold.

[0035] In this embodiment, the regional cascading failure risk is analyzed using a three-layer progressive method, which can be: (1) Abnormal device group detection: When more than a threshold number of devices (e.g., ≥3 devices in a network point) are found to have the same type of abnormal characteristics at the same time, they can be detected as an abnormal device group to trigger a primary warning. (2) Dependency detection between abnormal device groups: This can be understood as risk transmission path verification. It can automatically check the physical dependency between abnormal device groups, confirm whether the abnormal devices belong to the same circuit branch, and check whether there is data interaction dependency between devices (e.g., shared network switches). (3) Abnormal persistence detection: In order to avoid false alarms due to instantaneous interference, it can detect whether the abnormal state meets the time persistence requirement, for example, the abnormal indicator appears continuously in 3 detection cycles (3×30 seconds). Based on the above regional cascading failure risk analysis results, if the maintenance conditions are met, the current edge gateway node can generate corresponding maintenance request information.

[0036] The advantage of this setup is that it enables regional risk early warning through collaborative analysis of device status within the region.

[0037] This embodiment, based on localized fault prediction and regional risk collaborative perception of edge layer devices, avoids communication latency associated with cloud-based decision-making to a certain extent. Combined with the efficient inference capabilities of a lightweight prediction model, it achieves a real-time closed loop throughout the entire process from data acquisition to maintenance request generation, significantly shortening the emergency response window for critical faults.

[0038] S130. The dispatch center receives maintenance request information from each edge gateway node within its jurisdiction; it evaluates each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine the global priority score for each maintenance request information; it filters out candidate maintenance request information based on the global priority score and performs correlation analysis on the candidate maintenance request information to identify the correlation types between them; based on the global priority score, correlation type, and available resource status within the jurisdiction of the candidate maintenance request information, it generates a dispatch plan to achieve equipment maintenance.

[0039] In this embodiment, the scheduling center can evaluate the global priority score of maintenance request information for each edge gateway node based on various preset evaluation dimensions, thereby filtering candidate maintenance request information according to the order of global priority scores for subsequent scheduling processing. However, the global priority score of a single maintenance request is insufficient to guarantee optimal scheduling because there may be complex dependencies between maintenance tasks. Therefore, the scheduling center in this embodiment can perform task correlation analysis on the maintenance tasks corresponding to the candidate maintenance request information. The core purpose is to identify potential hidden dependencies between tasks, prevent scheduling conflicts, and explore opportunities for collaborative optimization. Through global priority scoring and in-depth correlation and coupling analysis, the scheduling center can generate an optimal scheduling scheme to achieve equipment maintenance based on the global priority score of the candidate maintenance request information, the correlation between maintenance tasks, and the status of available resources within its jurisdiction.

[0040] In one optional implementation, the dispatch center evaluates each maintenance request based on preset evaluation dimensions with dynamically adjusted weights to determine the global priority score for each maintenance request. This may include: the dispatch center quantitatively evaluating the current preset evaluation dimensions of the current maintenance request to obtain a score for the current preset evaluation dimension; the dispatch center detecting the current preset evaluation dimension according to preset detection indicators, and adjusting the weight of the current preset evaluation dimension if the detection results meet the weight adjustment rules; and the dispatch center using the adjusted weights to perform a weighted summation of the scores for each preset evaluation dimension to determine the global priority score for the current maintenance request.

[0041] In this embodiment, the preset evaluation dimensions may include business urgency, repair feasibility, risk severity, and resource consumption ratio. The business urgency dimension assesses the potential loss of customer waiting time, potential transaction volume, and the importance of the equipment's location (e.g., a main branch). The repair feasibility dimension assesses whether the current availability of spare parts, the skills of technical maintenance personnel, and the distance between the technician's current location and the faulty equipment match the fault type in the maintenance request. The risk severity dimension assesses the security level of the faulty equipment in the maintenance request (e.g., involving sensitive information such as cash), the probability of the fault spreading to other equipment or systems, and the impact on the overall operation of the branch. The resource consumption ratio dimension assesses the required man-hours, spare parts costs, and the expected benefits or avoided losses after the repair.

[0042] Furthermore, the dispatch center can quantitatively evaluate each maintenance request based on the aforementioned preset evaluation dimensions. Each preset evaluation dimension corresponds to a weight factor, and each weight factor has a corresponding detection index and dynamic weight adjustment rules. Specifically:

[0043] The weighting factor for the business urgency dimension can be expressed as a time sensitivity coefficient. The detection metric can be the ratio of current period traffic to historical peak traffic; the weighting adjustment rule can be set when a certain value is detected. When business traffic is close to its peak, the weight of the business urgency dimension is automatically increased (e.g., +15%), indicating that the failure of high-transaction-volume devices can be prioritized.

[0044] The weighting factor for the feasibility dimension of repair can be represented as the resource supply coefficient. The testing indicators can be spare parts inventory or the readiness status of technical personnel; the weighting adjustment rules can be as follows: When there is a shortage of spare parts or manpower, the weight of the repair feasibility dimension increases by 20%, indicating that resources can be prioritized for fault tasks that can be resolved quickly at present.

[0045] The weighting factors for the severity of risk can be expressed as risk evolution coefficients. The detection indicator can be the rate of change in the intensity of the risk transmission chain; the weighting adjustment rule can be... (Rapid spread of risk) The severity weight of the risk is increased by 25%, indicating that the risk of cascading failures can be blocked first.

[0046] The weighting factor for the resource consumption ratio dimension can be expressed as the feedback calibration coefficient. The detection index can be the error of comparing service loss from historical scheduling results; the weight adjustment rule can be a periodic pass... Correcting historical weighting errors, for example: if Then amplify the weight of key dimensions. Then reduce the weight of redundant dimensions.

[0047] The scheduling center can calculate a global priority score for each maintenance request based on each preset evaluation dimension and the corresponding weight factor.

[0048] In this embodiment, after the scheduling center completes the independent quantitative scoring of each maintenance request across multiple dimensions, it can calculate a global priority score (range 0-100) for each request. This score integrates the contributions of all dimensions under dynamic weights, thus dynamically adjusting the preset evaluation dimension weights to improve the accuracy of maintenance request evaluation. Subsequently, the scheduling center can select tasks with scores higher than a set threshold or those ranking high (e.g., top 10%) as candidate maintenance requests and send them to the maintenance decision pool for further scheduling processing.

[0049] In one optional implementation, the scheduling center performs correlation analysis on the candidate maintenance request information to identify the correlation types between the candidate maintenance request information. This may include: the scheduling center performs correlation analysis on the candidate maintenance request information according to preset detection rules to identify the correlation types between the candidate maintenance request information.

[0050] In one optional implementation, the scheduling center generates a scheduling scheme to achieve equipment maintenance based on the global priority score, association type, and available resource status within its jurisdiction of the candidate maintenance request information. This may include: the scheduling center performing association processing on the candidate maintenance request information according to processing rules matching the association type; and the scheduling center generating a scheduling scheme based on a specified dimension based on the global priority score of the associated maintenance request information and the available resource status within its jurisdiction to achieve equipment maintenance.

[0051] In this embodiment, the association type can include at least one of fault propagation association, resource competition association, geographic collaboration association, and temporal dependency association. Each association type can have corresponding detection methods and processing rules. Specifically, the detection method for fault propagation association can be based on the device topology map (e.g., circuit dependency or network dependency) to calculate the fault propagation probability, and the processing rule can be to bundle and schedule upstream and downstream maintenance tasks, prioritizing the source device. The detection method for resource competition association can be to detect the demand of multiple maintenance tasks for the same scarce resource (e.g., a single highly skilled engineer or a specific spare part), and the processing rule can be to allocate resources according to the maintenance task's rating or resource consumption order. The detection method for geographic collaboration association can be to calculate the Euclidean distance between maintenance task points through spatial clustering analysis (e.g., within the same network point or within 1km), and the processing rule can be to group them into geographic groups and have the same engineer process them according to the optimal path. The detection method for temporal dependency association can be to check maintenance sequence constraints (e.g., repairing the core switch before repairing the terminal), and the processing rule can be to add hard temporal constraints, freezing subsequent maintenance tasks if the preceding maintenance task is not completed.

[0052] Based on the above analysis, the scheduling center can obtain the correlation between maintenance request information, laying a solid foundation for generating efficient and feasible scheduling strategies in the next step.

[0053] This embodiment can deploy a decision-making agent in the scheduling center to generate scheduling schemes. The decision-making agent can be trained based on a multi-agent near-end policy optimization reinforcement learning framework to process maintenance request information allocated to its region in parallel. The core input parameters of the decision-making agent can include the global priority score of the maintenance request information after association and the status of available resources within its jurisdiction. The status of available resources can include the status of available technicians (e.g., the location, skills, and availability of available technicians), the status of available maintenance vehicles, and the status of available spare parts (real-time inventory and location information of available spare parts), thereby outputting a scheduling scheme based on specified dimensions. The specified dimensions can include personnel allocation, spare parts allocation, vehicle allocation, and execution sequence. Personnel allocation can refer to identifying and assigning the nearest technician with the highest skill matching degree; spare parts allocation and vehicle allocation can refer to planning the optimal spare parts transportation route (e.g., retrieving from a warehouse or nearby network point) and assigning maintenance vehicles; execution sequence can refer to setting a strict handling time window for high-priority tasks (especially tasks with TOP priority scores) (e.g., requiring technicians to arrive on-site within 15 minutes to respond).

[0054] This embodiment utilizes a four-dimensional spatiotemporal weight adjustment mechanism based on a dynamic priority engine to accurately quantify the cascading risk impact and economic value loss of equipment failures. Under resource constraints, it intelligently guides core resources such as spare parts and manpower towards high-value equipment, effectively avoiding resource misallocation and idle waste.

[0055] Optionally, after generating a scheduling plan to achieve equipment maintenance, the scheduling center in this embodiment can also obtain maintenance feedback data, which may include actual fault conditions, actual repair time, actual resource usage, and actual maintenance results. The scheduling center optimizes the generation strategy of the scheduling plan based on the maintenance feedback data.

[0056] The actual fault conditions can be used to verify the accuracy of fault prediction. Actual resource usage can include the types and quantities of spare parts actually consumed, the matching of actual technical personnel, and the number of vehicles actually used. Actual maintenance results can include types such as successful maintenance, partial maintenance, and requiring further processing. This embodiment can automatically trigger the policy network of the decision-making agent to perform iterative updates when the fault misjudgment rate (e.g., exceeds 5%) or the scheduling efficiency is not up to standard.

[0057] Optionally, this embodiment can also periodically update the predictive models of each financial device based on maintenance feedback data. The dispatch center in this embodiment can also periodically analyze historical data and calibrate and adjust the weights of each preset evaluation dimension to adapt to changes in business needs, device distribution, or failure modes, thereby adapting to changes in business models.

[0058] The layered and decoupled architecture of the device management system in this embodiment can ensure that the system has elastic scalability. The device layer only requires lightweight sensors and communication modules, and the edge layer can deploy general computing nodes to carry local models. The cloud center realizes cross-regional strategy coordination, which significantly reduces the cost of intelligent transformation for small and medium-sized financial institutions.

[0059] To enable those skilled in the art to better understand the financial equipment maintenance method of this embodiment, the following provides a workflow of the financial equipment maintenance method of this embodiment in a concrete scenario.

[0060] On a Friday afternoon at 3:30 PM (peak business hours), a sudden voltage fluctuation occurred in the power grid in the city center.

[0061] 1. Affected equipment

[0062] Core Branch A (Main Branch, daily transaction volume exceeding 3000 transactions): ATM1's power module has a hidden defect (failure probability increased to 85%); VTM1's card reader is faulty (currently down); network latency in the security access control system has surged (failure probability 72%). Neighboring Branch B (Medium-sized Branch): ATM2 / ATM3's power modules are malfunctioning (failure probability > 60%). Suburban Branch C: The ticket processing terminal is mechanically jammed (currently down).

[0063] 2. Resource constraints

[0064] Available technical personnel: 2 (Technician X is skilled in electronic fault handling, Technician Y is skilled in mechanical repair); Scarce spare parts: Only 1 high-power power supply module remains.

[0065] The full-process response of the device management system in this embodiment is as follows:

[0066] Phase 1: Real-time prediction and collaborative early warning for edge financial devices (15:30-15:32)

[0067] 1. Single-machine prediction: The edge terminal of ATM1 at site A detected a sudden increase in the current harmonic distortion rate. The local prediction model calculated that its failure probability within the next 30 minutes reached 85% (exceeding the threshold of 50%), and an early warning was immediately triggered; ATM2 / ATM3 at site B simultaneously detected an abnormal temperature rise rate, and the failure probability exceeded 60%.

[0068] 2. Regional Collaborative Analysis: The edge gateway of site A receives abnormal broadcasts (including failure probability, temperature rise rate, and network latency increment) from ATM1, VTM1, and the access control system through the self-organizing network.

[0069] 3. Three-tier risk verification initiated:

[0070] (1) Abnormal equipment group detection: 3 devices in point A simultaneously report power / network abnormalities (≥3 devices).

[0071] (2) Detection of inter-device dependency relationship (transmission path verification): The topology diagram shows that ATM1, VTM1 and access control system share the same circuit branch.

[0072] (3) Abnormal persistence detection: Abnormal indicators do not subside for three consecutive detection cycles (90 seconds).

[0073] Conclusion: A regional cascading failure warning is generated and marked as a high-risk event.

[0074] Phase Two: Maintenance Task Assessment and Correlation Analysis (15:32-15:35)

[0075] 1. Evaluation Dimensions:

[0076] (1) Business urgency: Branch A is currently experiencing peak business hours ( The weight is automatically increased by 15%. The VTM1 outage caused a large amount of business to stop, with an urgency score of 92. (2) Feasibility of repair: There is only 1 power module in stock. ), the feasibility score for repairing ATM1 dropped sharply. (3) Severity of risk: ATM1 failure may spread to the access control system ( ), Risk score 95. (4) Resource consumption ratio: Suburban branch C has low maintenance costs for the ticket terminal but low revenue, score 45.

[0077] Global priority rating:

[0078] ATM1 (Power Supply): Priority 96 points, high risk transmission + high business impact; VTM1 (Card Reader): Priority 88 points, high business urgency; Ticket Terminal: Priority 40 points, low resource consumption.

[0079] 2. Association Analysis:

[0080] Fault propagation correlation: ATM1 and the access control system are bundled for scheduling (same circuit branch).

[0081] Resource competition: ATM1 competes with ATM2 / ATM3 for the only power module, triggering an auction mechanism: ATM1's score / resource consumption = 96 / 1 = 96, ATM2's score / resource consumption = 75 / 1 = 75, result: the power module is preferentially allocated to ATM1.

[0082] Geographical coordination: VTM1 and access control system at site A are packaged into the same task group (distance <50 meters).

[0083] Phase 3: Scheduling Plan and Execution (15:35-15:50)

[0084] 1. Personnel Assignment: Technician X (Electronics Specialist) → Prioritizes handling task group A at branch A (ATM1 power supply replacement + VTM1 card reader repair). Technician Y (Mechanical Specialist) → Repairs ticket terminal at suburban branch C (low priority task).

[0085] 2. Resource allocation: Power modules are delivered directly from the regional warehouse to point A (route optimization saves 20 minutes).

[0086] 3. Execution sequence: Technician X must arrive at site A within 15 minutes (high priority task, hard window).

[0087] 4. Execution Feedback: Technician X repaired ATM1 at 15:48, preventing the risk of a cascading failure in the access control system. VTM1 was restored at 16:05, reducing peak service interruption time to 35 minutes (compared to the traditional method which required...). (90 minutes).

[0088] Phase Four: Closed-Loop Optimization (Next Day)

[0089] Predictive Model Update: The predictive model has been trained with voltage fluctuation features to improve the accuracy of predicting latent faults in power modules. Strategy Optimization: The decision-making agent has learned a 23% improvement in the benefit of the strategy of "prioritizing backbone network points during peak business periods." Weight Calibration: The weight of business urgency has been permanently increased by 10%. The matrix shows historically underestimated peak impact.

[0090] In this embodiment, the technical solution involves each financial device analyzing its local operating data to generate device status index data; edge gateway nodes analyzing device status index data from various financial devices within their management scope to generate maintenance request information; the dispatch center receiving maintenance request information from each edge gateway node within its jurisdiction; evaluating each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filtering candidate maintenance request information based on the global priority score and performing correlation analysis on the candidate maintenance request information to identify the correlation types between them; and generating a scheduling scheme based on the global priority score, correlation type, and available resource status within the jurisdiction of the candidate maintenance request information to achieve device maintenance. This approach, combining edge-layer collaborative processing, constructing a multi-dimensional evaluation mechanism, and analyzing the correlations between maintenance tasks to generate a scheduling scheme, addresses the current banking device maintenance system's over-reliance on centralized cloud processing and lack of intelligent edge collaboration capabilities, leading to low decision-making efficiency; and the problem of unscientific resource scheduling caused by the simple priority decision-making rules of the current banking device maintenance system, thereby improving decision-making efficiency and the rationality of resource scheduling.

[0091] Example 2

[0092] Figure 2This is a schematic diagram of a device management system provided in Embodiment 2 of the present invention. Figure 2 As shown, the system includes: a dispatch center 210, various edge gateway nodes 220, and various financial devices 230 managed by each edge gateway node. Among them:

[0093] Each of the financial devices 230 is used to analyze its own operating data to generate device status index data;

[0094] The edge gateway node 220 is used to analyze the device status indicator data from various financial devices within the management scope to generate maintenance request information;

[0095] The scheduling center 210 is used to receive maintenance request information from each edge gateway node within its jurisdiction; evaluate each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filter candidate maintenance request information based on the global priority score and perform correlation analysis on the candidate maintenance request information to identify the correlation type between the candidate maintenance request information; and generate a scheduling scheme to achieve equipment maintenance based on the global priority score of the candidate maintenance request information, the correlation type, and the available resource status within the jurisdiction.

[0096] In this embodiment, the technical solution involves each financial device analyzing its local operating data to generate device status index data; edge gateway nodes analyzing device status index data from various financial devices within their management scope to generate maintenance request information; the dispatch center receiving maintenance request information from each edge gateway node within its jurisdiction; evaluating each maintenance request information based on preset evaluation dimensions with dynamically adjusted weights to determine a global priority score for each maintenance request information; filtering candidate maintenance request information based on the global priority score and performing correlation analysis on the candidate maintenance request information to identify the correlation types between them; and generating a scheduling scheme based on the global priority score, correlation type, and available resource status within the jurisdiction of the candidate maintenance request information to achieve device maintenance. This approach, combining edge-layer collaborative processing, constructing a multi-dimensional evaluation mechanism, and analyzing the correlations between maintenance tasks to generate a scheduling scheme, addresses the current banking device maintenance system's over-reliance on centralized cloud processing and lack of intelligent edge collaboration capabilities, leading to low decision-making efficiency; and the problem of unscientific resource scheduling caused by the simple priority decision-making rules of the current banking device maintenance system, thereby improving decision-making efficiency and the rationality of resource scheduling.

[0097] Optionally, each of the aforementioned financial devices can be specifically used for:

[0098] The edge intelligent terminal of the current financial equipment collects the local operating data in real time, and the local operating data includes at least one of the following: equipment operating temperature, equipment vibration data, equipment operating current data, and equipment transaction monitoring data;

[0099] The current financial equipment uses a predictive model to analyze its own operating data and predict the probability of failure of the current financial equipment in a future specified period of time;

[0100] The current financial device generates the device status index data based on the local operating data and the failure probability.

[0101] Optionally, the edge gateway node can be specifically used for:

[0102] The current edge gateway node receives device status indicator data from various financial devices within its management scope;

[0103] The current edge gateway node performs aggregate analysis on the device status index data based on a specified time window to obtain a first type of dataset and a second type of dataset. The first type of dataset is used to statistically analyze the distribution of various device status indicators of each financial device managed by the current edge gateway node within the specified time window, and the second type of dataset is used to record the fluctuation of each device status indicator of each financial device within a specified historical period.

[0104] The current edge gateway node analyzes the regional cascading failure risk within its management scope using a three-layer progressive method based on the first type of dataset and the second type of dataset; the three-layer progressive method includes, in sequence, abnormal device group detection, abnormal device group dependency detection, and abnormal persistence detection;

[0105] If the regional cascading failure risk analysis results meet the maintenance conditions, the current edge gateway node generates corresponding maintenance request information, which includes equipment information of the device to be maintained and fault description information.

[0106] Optionally, the scheduling node can be specifically used for:

[0107] The current preset evaluation dimensions of the current maintenance request information are quantitatively evaluated to obtain a score for the current preset evaluation dimensions;

[0108] The current preset evaluation dimension is tested according to the preset detection index. If the detection result meets the weight adjustment rule, the weight of the current preset evaluation dimension is adjusted.

[0109] The scores of each preset evaluation dimension are weighted and summed using the adjusted weights to determine the global priority score of the current maintenance request information.

[0110] Optionally, the scheduling node can also be used for:

[0111] The candidate maintenance request information is subjected to correlation analysis according to preset detection rules to identify the correlation types between the candidate maintenance request information; the correlation types include at least one of fault propagation correlation, resource competition correlation, geographical collaboration correlation and temporal dependency correlation.

[0112] Optionally, the scheduling node can also be used for:

[0113] The candidate maintenance request information is associated with the processing rules that match the association type.

[0114] Based on the global priority score of the maintenance request information after association processing and the status of available resources within the jurisdiction, a scheduling scheme is generated based on a specified dimension to achieve equipment maintenance.

[0115] The available resource status includes the status of available technicians, available maintenance vehicles, and available spare parts. The specified dimensions include personnel allocation, spare parts allocation, vehicle allocation, and execution sequence.

[0116] Optionally, the dispatch center can also be used for:

[0117] Obtain maintenance feedback data, which includes actual fault conditions, actual repair time, actual resource usage, and actual maintenance results;

[0118] The generation strategy of the scheduling scheme is optimized based on the maintenance feedback data.

[0119] Optionally, the aforementioned financial devices may also be used for:

[0120] The prediction model is updated periodically.

[0121] Optionally, the dispatch center can also be used for:

[0122] Regularly analyze historical data and calibrate and adjust the weights of each preset evaluation dimension.

[0123] Example 3

[0124] Figure 3 A schematic diagram of an electronic device 300 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers or various forms of mobile devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0125] like Figure 3 As shown, the electronic device 300 includes at least one processor 301 and a memory, such as a read-only memory (ROM) 302 or a random access memory (RAM) 303, communicatively connected to the at least one processor 301. The memory stores computer programs executable by the at least one processor. The processor 301 can perform various appropriate actions and processes based on the computer program stored in the ROM 302 or loaded into the RAM 303 from storage unit 308. The RAM 303 can also store various programs and data required for the operation of the electronic device 300. The processor 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.

[0126] Multiple components in electronic device 300 are connected to I / O interface 305, including: input unit 306, such as keyboard, mouse, etc.; output unit 307, such as various types of displays, speakers, etc.; storage unit 308, such as disk, optical disk, etc.; and communication unit 309, such as network card, modem, wireless transceiver, etc. Communication unit 309 allows electronic device 300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0127] Processor 301 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 301 performs the various methods and processes described above, such as financial equipment maintenance methods.

[0128] In some embodiments, the financial device maintenance method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 308. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 300 via ROM 302 and / or communication unit 309. When the computer program is loaded into RAM 303 and executed by processor 301, one or more steps of the financial device maintenance method described above may be performed. Alternatively, in other embodiments, processor 301 may be configured to perform the financial device maintenance method by any other suitable means (e.g., by means of firmware).

[0129] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0130] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0131] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0132] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0133] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0134] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0135] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0136] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for maintaining financial equipment, applied to an equipment management system, the equipment management system comprising a dispatch center, edge gateway nodes, and financial equipment managed by each edge gateway node, characterized in that, include: Each financial device analyzes its own operating data to generate device status indicator data. The edge gateway node analyzes the device status indicator data from various financial devices within its management scope to generate maintenance request information. The dispatch center receives maintenance request information from each edge gateway node within its jurisdiction; Each maintenance request is evaluated based on preset evaluation dimensions with dynamically adjusted weights to determine the global priority score for each maintenance request. Candidate maintenance request information is selected based on the global priority score, and correlation analysis is performed on the candidate maintenance request information to identify the correlation type between the candidate maintenance request information; based on the global priority score of the candidate maintenance request information, the correlation type, and the available resource status within the jurisdiction, a scheduling scheme is generated to realize equipment maintenance.

2. The method according to claim 1, characterized in that, Each financial device analyzes its own operating data to generate device status indicator data, including: The edge intelligent terminal of the current financial equipment collects the local operating data in real time, and the local operating data includes at least one of the following: equipment operating temperature, equipment vibration data, equipment operating current data, and equipment transaction monitoring data; The current financial equipment uses a predictive model to analyze its own operating data and predict the probability of failure of the current financial equipment in a future specified period of time; The current financial device generates the device status index data based on the local operating data and the failure probability.

3. The method according to claim 1, characterized in that, The edge gateway node analyzes the device status indicator data from various financial devices within its management scope to generate maintenance request information, including: The current edge gateway node receives device status indicator data from various financial devices within its management scope; The current edge gateway node performs aggregate analysis on the device status index data based on a specified time window to obtain a first type of dataset and a second type of dataset. The first type of dataset is used to statistically analyze the distribution of various device status indicators of each financial device managed by the current edge gateway node within the specified time window, and the second type of dataset is used to record the fluctuation of each device status indicator of each financial device within a specified historical period. The current edge gateway node analyzes the regional cascading failure risk within its management scope using a three-layer progressive method based on the first type of dataset and the second type of dataset; the three-layer progressive method includes, in sequence, abnormal device group detection, abnormal device group dependency detection, and abnormal persistence detection; If the regional cascading failure risk analysis results meet the maintenance conditions, the current edge gateway node generates corresponding maintenance request information, which includes equipment information of the device to be maintained and fault description information.

4. The method according to claim 1, characterized in that, The scheduling center evaluates each maintenance request based on preset evaluation dimensions with dynamically adjusted weights, and determines a global priority score for each maintenance request, including: The scheduling center performs a quantitative evaluation of the current preset evaluation dimension of the current maintenance request information to obtain a score for the current preset evaluation dimension; The scheduling center performs detection on the current preset evaluation dimension according to preset detection indicators, and adjusts the weight of the current preset evaluation dimension if the detection result meets the weight adjustment rules. The scheduling center uses adjusted weights to perform a weighted summation of the scores for each preset evaluation dimension to determine the global priority score for the current maintenance request information.

5. The method according to claim 1, characterized in that, The scheduling center performs correlation analysis on the candidate maintenance request information to identify the correlation types between the candidate maintenance request information, including: The scheduling center performs correlation analysis on the candidate maintenance request information according to preset detection rules to identify the correlation types between the candidate maintenance request information; the correlation types include at least one of fault propagation correlation, resource competition correlation, geographical collaboration correlation and temporal dependency correlation.

6. The method according to claim 1, characterized in that, The scheduling center generates a scheduling plan to implement equipment maintenance based on the global priority score of the candidate maintenance request information, the association type, and the available resource status within the jurisdiction, including: The scheduling center performs association processing on the candidate maintenance request information according to the processing rules that match the association type; The scheduling center generates a scheduling scheme based on a specified dimension to achieve equipment maintenance, according to the global priority score of the maintenance request information after association processing and the status of available resources within the jurisdiction. The available resource status includes the status of available technicians, available maintenance vehicles, and available spare parts. The specified dimensions include personnel allocation, spare parts allocation, vehicle allocation, and execution sequence.

7. The method according to claim 1, characterized in that, After generating a scheduling plan to implement equipment maintenance, the scheduling center also includes: The dispatch center obtains maintenance feedback data, which includes actual fault conditions, actual repair time, actual resource usage, and actual maintenance results. The scheduling center optimizes the generation strategy of the scheduling scheme based on the maintenance feedback data.

8. The method according to claim 2, characterized in that, Also includes: Each of the financial devices updates the prediction model periodically.

9. The method according to claim 1, characterized in that, Also includes: The scheduling center periodically analyzes historical data and calibrates and adjusts the weights of each preset evaluation dimension.

10. A device management system, comprising a dispatch center, edge gateway nodes, and financial devices managed by each edge gateway node, characterized in that, include: Each of the aforementioned financial devices is used to analyze its own operating data to generate device status indicator data; The edge gateway node is used to analyze the device status indicator data from various financial devices within the management scope to generate maintenance request information; The dispatch center is used to receive maintenance request information from each edge gateway node within its jurisdiction; Each maintenance request is evaluated based on preset evaluation dimensions with dynamically adjusted weights to determine the global priority score for each maintenance request. Candidate maintenance request information is selected based on the global priority score, and correlation analysis is performed on the candidate maintenance request information to identify the correlation type between the candidate maintenance request information; based on the global priority score of the candidate maintenance request information, the correlation type, and the available resource status within the jurisdiction, a scheduling scheme is generated to realize equipment maintenance.

Images