Hospital check-in screen maintenance system based on cloud computing

Abstract

This invention relates to the field of screen maintenance technology, specifically a cloud-based hospital check-in screen maintenance system. The system includes a communication behavior analysis module, an interruption time identification module, a check-in screen instruction filtering module, a maintenance resource matching module, and a rescheduling task window module. In this invention, by extracting request behavior timing and filtering status codes, the system identifies the clustering characteristics of retry behaviors, forming an abnormal communication pattern profile, enhancing the accuracy of device anomaly identification. Combined with time overlap comparison between logs and instruction segment intervals, the system accurately deduces the fault window, avoiding the impact of time drift on diagnosis. Abnormal events are generated based on the cross-generation of task status and response time, possessing dynamic adaptability and supporting multi-scenario adaptation. Resource scheduling is based on time slot matching, forming a highly compatible combination between tasks and personnel, improving scheduling efficiency and execution success rate, providing structured support for maintenance strategy optimization, and enhancing overall maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of screen maintenance technology, and in particular to a cloud computing-based hospital check-in screen maintenance system. Background Technology

[0002] The field of screen maintenance technology encompasses remote monitoring, maintenance, fault diagnosis, and information management of the operational status of various display screen devices. Its core lies in using information technology to manage the lifecycle of display terminals deployed in various scenarios, including but not limited to display fault early warning, component operation monitoring, maintenance record tracking, and remote control management. Overall, it covers the construction of maintenance systems for various types of terminal display devices in industrial, commercial, and public service sectors, with a focus on ensuring the stability, security, and maintainability of display terminal equipment during long-term operation.

[0003] The cloud-based hospital check-in screen maintenance system refers to a system that uses a cloud computing architecture to centrally and remotely manage the check-in screens used in hospitals. It primarily targets self-service check-in terminals deployed within hospitals, encompassing real-time collection of their operational status information, automatic fault identification, maintenance task scheduling and management, and maintenance record archiving and analysis. By setting up data processing nodes in the cloud to receive data streams from the front-end check-in screens, it utilizes a rule-driven monitoring mechanism to determine abnormal states and performs hierarchical management based on defined equipment classification standards. Simultaneously, it integrates with the back-end management system to dispatch maintenance work orders, collect maintenance records, and integrate equipment lifecycle information, thus achieving unified maintenance management of the hospital check-in screens.

[0004] In the current maintenance process of hospital check-in screens, a pattern recognition system targeting communication behavior characteristics has not been established. Processing methods largely rely on independent data points, lacking methods to mine the temporal sequence and density between behavior groups. This makes it difficult to identify repetitive and aggregated retry behaviors during anomaly detection, hindering the effective identification of device-level communication anomalies. Regarding time window positioning, it relies heavily on one-way recording of fault logs and manual comparison, making it difficult to establish logical mappings across multiple time sources. This often leads to window offsets and biased judgments, reducing the accuracy of fault handling. Task anomaly judgment criteria are singular, typically relying on preset thresholds or rule bases, which are insufficient to cope with changing equipment operating states, and results are easily distorted by environmental interference. Resource scheduling focuses on the formal matching of personnel shifts, lacking a collaborative calculation mechanism based on task characteristics, personnel capabilities, and available time periods. This results in passively delaying the response cycle of some tasks and low efficiency in maintenance resource allocation. The anomaly report generation process is disconnected from task management data, failing to utilize statistical correlation and structured analysis to extract task anomaly patterns and lacking basic data support for strategy iteration. For example, in hospital scenarios where scheduling conflicts frequently occur, multiple overlapping tasks are repeatedly assigned to unreachable time slots because the overlap of time slots is not assessed, resulting in task rollback and response failures, exposing the shortcomings of existing solutions in resource coordination and anomaly assessment. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a cloud computing-based hospital check-in screen maintenance system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A cloud-based hospital check-in screen maintenance system includes:

[0007] The communication behavior analysis module obtains the cloud request ID, UTC timestamp and retry count of each hospital number, reads the record group with the same session ID and continuous time interval, filters it using HTTP status codes, extracts the instruction segment interval that meets the conditions by combining time sequence, establishes the frequency summary of adjacent repeated segments, and generates the request retry distribution density.

[0008] The interruption time identification module obtains the fault time window based on the hospital number corresponding to the request retry distribution density, compares the instruction segment time interval with the time in the fault log, and calculates and outputs the time overlap rate.

[0009] The check-in screen instruction filtering module, based on the time overlap rate, statistically analyzes the hospital check-in screen SLA service records, extracts task execution status and response time, calculates the anomaly occurrence rate, and generates an anomaly event report.

[0010] The maintenance resource matching module extracts the current maintenance personnel schedule based on the hospital number corresponding to the abnormal event report, filters the set of maintenance personnel in an idle state, calculates the matching degree between the task time window and the available time slot, and generates a resource scheduling plan.

[0011] As a further aspect of the present invention, the request retry distribution density includes retry frequency distribution records, abnormal request concentration, and time series consistency index; the time overlap rate includes overlap duration, overlap ratio, and time matching level; the abnormal event report includes abnormal category statistics, response delay distribution data, and task status labels; and the resource scheduling plan includes personnel availability mapping, task time window matching score, and preliminary scheduling priority ranking results.

[0012] As a further aspect of the present invention, the communication behavior parsing module includes:

[0013] The request identifier extraction submodule obtains the cloud request ID, UTC timestamp, and retry count for each hospital number, reads record groups with consistent session IDs and continuous time intervals, performs group matching based on the UTC timestamp and request ID in each record group, establishes a mapping index between request time and request ID, and obtains the request sequence time index.

[0014] The abnormal segment filtering submodule reads the request records under the same session ID according to the request sequence time index, filters the HTTP status code and arranges each request segment in the request order, determines whether the time difference between adjacent segments is less than the maximum allowed interval value under the same session, and extracts the abnormal request interval segment set.

[0015] The retry density generation submodule counts the number of repeated request identifiers within the abnormal request interval segment set, calculates the ratio of the repetition rate of each request identifier in each segment to the length of the interval, determines whether the ratio is greater than the fixed request density benchmark value, and counts the request retry distribution density.

[0016] As a further aspect of the present invention, the interruption time identification module includes:

[0017] The retry density screening submodule extracts the abnormal request intervals corresponding to each hospital number based on the hospital number corresponding to the request retry distribution density, counts and judges whether the request repetition rate in each interval exceeds the set request repetition rate threshold, and screens out high-frequency abnormal interval segments.

[0018] The log window extraction submodule obtains the fault log corresponding to the matching hospital number based on the time interval in the high-frequency abnormal interval segment, extracts the start time and end time marked as fault status in the log record, and combines them into a continuous time range list to obtain the fault time window set.

[0019] The overlap calculation output submodule calculates and outputs the time overlap rate by summarizing the intersection duration of the two time range sets based on the corresponding time ranges of the high-frequency abnormal intervals and the fault time window set, and combining it with the total duration of the corresponding abnormal intervals.

[0020] As a further aspect of the present invention, the time overlap rate is calculated using the following formula:

[0021] ;

[0022] Calculations are performed, in which, Indicates the time overlap rate. Indicates the first The start time of each overlapping time period Indicates the first The end time of each overlapping time period Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault time window segment Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault period This represents the total number of overlapping time periods. This indicates the number of time period combinations involved in the crossover estimation. Indicates the sequence number of the high-frequency anomaly interval. This represents the total number of all high-frequency abnormal intervals. This represents the absolute value operation. To represent the square root operation, This indicates a summation operation.

[0023] As a further aspect of the present invention, the check-in screen instruction filtering module includes:

[0024] The service record summary submodule assigns hospital numbers based on the time overlap rate, extracts SLA service records for the check-in screen under the matching hospital number, identifies maintenance record items that conform to IT service management standards, counts the number of service tasks and task time distribution for each hospital number, and obtains standard task records.

[0025] The status response extraction submodule extracts the execution status and response time value of each task based on the service record content in the standard task record, determines whether the task status is a failure or timeout mark, summarizes the number of tasks in the corresponding status, and obtains abnormal status task data.

[0026] The anomaly rate output submodule calculates the ratio of the number of abnormal tasks to the total number of tasks based on the number of tasks for each hospital number in the standard task record and the abnormal status task data, and establishes an abnormal event report.

[0027] As a further aspect of the present invention, the maintenance resource matching module includes:

[0028] The idle personnel extraction submodule extracts the current maintenance personnel schedule of the hospital, determines whether the schedule status of each person is marked as idle, filters the set of personnel with no assigned tasks, and obtains the number of available personnel.

[0029] The time matching calculation submodule extracts the task time window from the abnormal event report based on the idle time period of the corresponding personnel in the available personnel quantity data, calculates the overlap length between the time period of each idle personnel and the task window, sorts them, and generates a matching degree ranking result.

[0030] The scheduling result output submodule selects the set of personnel with the highest matching value based on the matching degree sorting result, binds the corresponding personnel to the task window, creates a scheduling pre-allocation table under the corresponding hospital number, and establishes a resource scheduling plan.

[0031] As a further aspect of the present invention, the overlap length between the idle time period of each person and the task window is determined by the following formula:

[0032] ;

[0033] Calculations are performed, in which, Indicates the first The available personnel and the first The length of overlap of all intersecting time periods between task windows. Indicates the number of idle personnel and the task window. The start and end times of the intersecting time intervals. Indicates the number of available personnel The duration of each idle period Indicates the task window number The duration of each time period This represents the absolute value operation. To represent the square root operation, This represents the summation over all intersecting time intervals. This indicates the number of time periods where there is overlap between idle personnel and the task window.

[0034] As a further aspect of the present invention, the system further includes:

[0035] Based on the resource scheduling plan, the task rescheduling window module combines the current maintenance task backlog status and asset number on the check-in screen to filter maintenance personnel binding combinations that meet the SLA response time limit, establishes a task-maintenance personnel mapping, and generates a maintenance task scheduling plan.

[0036] The maintenance task scheduling scheme includes a task number and maintenance personnel matching list, a priority scheduling list, and response time limit compliance assessment results.

[0037] As a further aspect of the present invention, the task rescheduling window module includes:

[0038] The delayed task screening submodule extracts the list of assigned tasks based on the resource scheduling plan, identifies tasks that are in the unstarted state and have an allocation delay mark according to the current check-in screen maintenance task status data, and filters them in combination with the asset number corresponding to the task to obtain a set of delayed task numbers.

[0039] The response combination judgment submodule extracts the priority of the maintenance task according to each task number in the task delay number set, and filters the personnel and task window combinations in the current scheduling plan according to the response time threshold in the SLA response time limit to obtain the compliant task combination result.

[0040] The task mapping output submodule extracts the corresponding task number and personnel number set based on the compliant task combination result, establishes a two-way mapping from task number to personnel number, and generates a maintenance task scheduling plan.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0042] In this invention, by extracting the timing of request behavior and filtering status codes, the clustering characteristics of retry behavior are identified, forming an abnormal communication pattern profile, which enhances the accuracy of device anomaly identification. By combining the time overlap comparison of logs and instruction segment intervals, the fault window can be accurately deduced, avoiding the impact of time drift on diagnosis. Abnormal events are generated based on the cross-generation of task status and response time, which has dynamic adaptability and supports multi-scenario adaptation. Resource scheduling is based on time slot matching degree, forming a highly compatible combination between tasks and personnel, improving scheduling efficiency and execution success rate, providing structured support for maintenance strategy optimization, and improving overall maintenance efficiency. Attached Figure Description

[0043] Figure 1 This is a system flowchart of the present invention;

[0044] Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0046] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0047] Please see Figure 1 and Figure 2 The cloud-based hospital check-in screen maintenance system includes:

[0048] The communication behavior analysis module obtains the cloud request ID, UTC timestamp and retry count of each hospital number, reads the record group with the same session ID and continuous time interval, filters the segments with HTTP status codes of 5xx and 408, extracts the instruction segment interval that meets the conditions by combining the time sequence, establishes the frequency summary of adjacent repeated segments, and generates the request retry distribution density.

[0049] The interruption time identification module obtains the fault time window based on the hospital number corresponding to the request retry distribution density, compares the instruction segment time interval with the time in the fault log, and calculates and outputs the time overlap rate.

[0050] The check-in screen instruction filtering module assigns the hospital number based on the time overlap rate, compiles the hospital check-in screen SLA service records (operational data records that conform to IT service management standards), extracts the task execution status and response time, calculates the anomaly occurrence rate, and generates anomaly event reports.

[0051] The maintenance resource matching module extracts the current maintenance personnel schedule based on the hospital number corresponding to the abnormal event report, filters the set of maintenance personnel in an idle state, calculates the matching degree between the task time window and the available time slot, and generates a resource scheduling plan.

[0052] The task rescheduling window module, based on the resource scheduling plan and combined with the current status of maintenance tasks on the reporting screen and asset numbers, selects maintenance personnel binding combinations that meet the SLA response time limit (the maximum response time threshold allowed for different priority fault maintenance services as specified in the Service Level Agreement (SLA) (e.g., P1 level ≤ 15 minutes, P2 level ≤ 2 hours), and ensures the timeliness of medical equipment maintenance services through a graded timing mechanism), establishes a task-maintenance personnel mapping, and generates a maintenance task scheduling plan.

[0053] The request retry distribution density includes retry frequency distribution records, abnormal request concentration, and time series consistency indicators. The time overlap rate includes overlap duration, overlap ratio, and time matching level. The abnormal event report includes abnormal category statistics, response delay distribution data, and task status labels. The resource scheduling plan includes personnel availability mapping, task time window matching score, and preliminary scheduling priority ranking results. The maintenance task scheduling plan includes a task number and maintenance personnel matching list, a priority scheduling list, and response time limit compliance assessment results.

[0054] Please see Figure 2 The communication behavior parsing module includes:

[0055] The request identifier extraction submodule obtains the cloud request ID, UTC timestamp, and retry count for each hospital number, reads record groups with consistent session IDs and continuous time intervals, performs group matching based on the UTC timestamp and request ID in each record group, establishes a mapping index between request time and request ID, and obtains the request sequence time index.

[0056] The request identifier extraction submodule obtains the cloud request ID, UTC timestamp, and retry count for each hospital. It first reads the hospital ID field from the log records and then identifies the tag field. , , , And extract the corresponding parameter values. For example, in a certain log entry, a field is... , obtain parameter values ​​in sequence , , , Then follow the same session number Group all log records, let the record set be . With the maximum interval Perform a time continuity check every second; if it exists... If they are recorded, they are grouped together. , If the difference is 180 seconds or less than 300 seconds, the two records are grouped together. Within each group, all records are arranged in ascending order of time to form a time series. , corresponding request ID Establish mapping index ,For example This completes the index structure from timestamp to request identifier, forming a request sequence time index for subsequent analysis.

[0057] The abnormal segment filtering submodule reads request records under the same session ID based on the request sequence time index, filters request items with HTTP status codes 5xx and 408, arranges each request segment in the request order, determines whether the time difference between adjacent segments is less than the maximum allowed interval value under the same session, and extracts the abnormal request interval segment set.

[0058] The exception segment filtering submodule reads the same session number based on the request time series index. All request records below, filtered by HTTP status code. satisfy or For example, if the status code set is Then, records with status codes 502, 408, and 503 are retained to form a candidate set. According to its corresponding timestamp sequence Sort in ascending order and calculate the time difference for any adjacent records. Let the maximum allowable interval be seconds, for example Seconds, because ,but Merge them into a single abnormal interval segment, and repeat this process recursively, recording all segment combinations that meet the merging criteria. If an abnormal interval occurs... If the merging process terminates and a new paragraph begins, the set of abnormal request intervals will ultimately be extracted. Each paragraph corresponds to a set of records and a time range, which are then output.

[0059] The retry density generation submodule counts the number of repeated request identifiers within the interval based on the abnormal request interval segment set, calculates the ratio of the repetition rate of each request identifier in each segment to the length of the interval, determines whether the ratio is greater than the fixed request density benchmark value, and counts the request retry distribution density.

[0060] The retry density generation submodule generates a set of intervals for abnormal requests. To conduct the analysis, let each paragraph... Includes request ID sequence For any request ID Count the number of occurrences Calculate its repetition rate If a paragraph contains 1 record, ID is Appear Next, then Then take the time range of this paragraph as ,like , ,but seconds, density value Set the requested density baseline value The average density in non-abnormal paragraphs is increased by two standard deviations. ,like , ,but The judgment condition is Requests that meet the conditions are considered to have abnormal density. Finally, the request identifiers that meet the density conditions in each paragraph and their corresponding density values ​​are counted and output.

[0061] Please see Figure 2 The interruption time identification module includes:

[0062] The retry density screening submodule extracts the abnormal request intervals corresponding to each hospital number based on the hospital number corresponding to the request retry distribution density, counts and judges whether the request repetition rate in each interval exceeds the set request repetition rate threshold, and screens out high-frequency abnormal interval segments.

[0063] The retry density screening submodule, based on the hospital number corresponding to the request retry distribution density, first sorts by hospital number. The request density data is grouped, and a set of all identified abnormal request intervals is extracted for each hospital number. Each paragraph Each segment contains a request ID sequence and retry statistics, and then the repetition rate of requests within each segment is calculated segment by segment. ,in Indicates the first The number of times a request ID is repeated. This represents the total number of requests within a paragraph. For example, if 3 out of 10 requests in a paragraph have the same request ID, then... Determine whether the repetition rate exceeds a set threshold. This threshold is based on the hospital's historical average retry rate. Its standard deviation Settings, for example, set , Then we can set To satisfy all paragraph Filter into set This is the set of high-frequency abnormal intervals, and the final output is the high-frequency abnormal interval corresponding to each hospital number.

[0064] The log window extraction submodule obtains the fault logs corresponding to the matching hospital number based on the time intervals in the high-frequency abnormal intervals, extracts the start time and end time marked as fault status in the log records, and combines them into a continuous time range list to obtain the fault time window set.

[0065] The log window extraction submodule first extracts each segment based on the time interval within the high-frequency anomaly interval. Time boundary The time range set for all paragraphs Based on the hospital numbers within the paragraphs Retrieve the corresponding hospital's fault log file, and filter out entries marked as fault states from each log entry. These entries include a status value field. The corresponding time field is and For example, extracting the log line "HospitalID=1023, Status=Fault, StartTime=1716888600, EndTime=1716888950" will give you the time range. Extract all record intervals that meet the condition of a fault, and merge the time ranges extracted from all logs into a set. After removing overlapping intervals, a set of continuous non-overlapping fault time windows is formed, and the output is the hospital number. The corresponding set of fault time periods.

[0066] The overlap calculation output submodule summarizes the duration of the intersection of the two time range sets based on the corresponding time ranges of the high-frequency abnormal intervals and the fault time window sets, and calculates and outputs the time overlap rate by combining the total duration of the corresponding abnormal intervals.

[0067] The time overlap rate is calculated using the following formula:

[0068] ;

[0069] Calculations are performed, in which, Indicates the time overlap rate. Indicates the first The start time of each overlapping time period Indicates the first The end time of each overlapping time period Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault time window segment Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault period This represents the total number of overlapping time periods. This indicates the number of time period combinations involved in the crossover estimation. Indicates the sequence number of the high-frequency anomaly interval. This represents the total number of all high-frequency abnormal intervals. This represents the absolute value operation. To represent the square root operation, This indicates a summation operation.

[0070] The collected data is as follows:

[0071] Number of high-frequency abnormal intervals The paragraph time is:

[0072] Paragraph 1: 10:00:00–10:05:00 → Second;

[0073] Paragraph 2: 10:10:00–10:15:00 → Second;

[0074] Paragraph 3: 10:20:00–10:25:00 → Second;

[0075] Fault Time Window :

[0076] Paragraph 1: 10:02:00–10:06:00 → Second;

[0077] Paragraph 2: 10:12:00–10:16:00 → Second;

[0078] Crossover :

[0079] Segment 1: The overlapping time period is 10:02:00–10:05:00 → continuing Second;

[0080] Paragraph 2: The overlapping time period is 10:12:00–10:15:00 → continuous Second;

[0081] The calculation process is as follows:

[0082] First intersection segment:

[0083] ;

[0084] ;

[0085] ;

[0086] ;

[0087] Sum of values ​​within a segment:

[0088] ;

[0089] The second intersection segment is exactly the same, with a value of 223.92.

[0090] Sum of the numerators:

[0091] ;

[0092] Denominator (total duration of outlier intervals):

[0093] ;

[0094] Substitute into the general formula:

[0095] ;

[0096] The results indicate that approximately 49.87% of the identified high-frequency anomaly intervals overlap with the fault time windows recorded in the system, and can be used in the overall analysis to evaluate the time coverage capability and matching accuracy of fault early warning.

[0097] The time overlap rate measures the degree of overlap between high-frequency anomaly intervals and fault time windows on the time axis. It is a ratio-based indicator representing the strength of the correlation between two types of events in the time dimension. This indicator is calculated by dividing the combined effective duration of all overlapping time periods by the total duration of all high-frequency anomaly intervals. It reflects whether abnormal behavior and fault events occur in a concentrated manner in time and whether there is a causal relationship trend. A higher value indicates that high-frequency anomalies occur more concentratedly near the fault time, exhibiting strong leading signal characteristics; a lower value indicates that anomalies and faults are dispersed in time, with weak event correlation. This indicator plays a crucial role in system anomaly correlation analysis, early warning mechanism verification, and causal modeling evaluation.

[0098] The formula combines the crossover duration term, difference adjustment term, and mean compensation term to comprehensively measure the degree of matching between the abnormal interval and the fault window in the time dimension. Crossover duration As the most basic measure of overlap, it represents the duration of the actual intersection between two time periods; adjustment term The difference in duration between high-frequency anomalies and fault windows is introduced. This product term, after taking the square root, forms a nonlinear adjustment for the effect of the duration difference, used to express the dilution effect of duration inconsistency on crossover; mean difference compensation term. This is used to eliminate systematic biases and remove the tendency for overlapping ratios to be raised or lowered due to the overall length or length of abnormal intervals. The aforementioned comprehensive value is adjusted back through subtraction to prevent short-term intersections from being exaggerated globally. The final numerator sums the local intersection adjustment values ​​of each segment, reflecting the total effective overlap intensity over the entire period. The denominator is the total duration of all abnormal intervals. A standardized time overlap rate is obtained by constructing a fractional form for consistency comparison across segments and tasks.

[0099] Please see Figure 2 The check-in screen instruction filtering module includes:

[0100] The service record summary submodule assigns hospital numbers based on time overlap rate, extracts SLA service records for check-in screens under matching hospital numbers, identifies maintenance record items that meet IT service management standards, counts the number of service tasks and task time distribution for each hospital number, and obtains standard task records.

[0101] The service record summary submodule assigns hospital numbers based on time overlap rate, first retrieving the time overlap rate index calculated in the previous processing stage. Corresponding hospital number For each number Access the check-in screen service record database as an index item, and retrieve the field with a matching value. All service record log entries, let the service record set be... Extract the field values ​​related to task execution status and task time information from each record, such as... For each record, determine whether it conforms to IT service management standards by comparing its SLA flag field value. The records with a value of True are selected and used to form the standard task set. Based on this, the hospital numbers were statistically analyzed. Total number of tasks and sorted by task start and end time fields. Calculate task time This forms a task time distribution set. For example, if a hospital has three tasks that take 300 seconds, 450 seconds, and 600 seconds respectively, then its time distribution is {300, 450, 600}. This set is used for subsequent abnormal response judgment and statistical analysis, and finally generates a standard task record set corresponding to the hospital number.

[0102] The status response extraction submodule extracts the execution status and response time value of each task based on the service record content in the standard task record, determines whether the task status is a failure or timeout mark, summarizes the number of tasks in the corresponding status, and obtains abnormal status task data.

[0103] The status response extraction submodule first iterates through the set of standard tasks generated in the previous stage based on the service record content in the standard task record. Read the task execution status field one by one With response time field Perform a judgment operation on the field value, and mark the task item with the abnormal status. or If established, then number the task. Add to abnormal state task set Simultaneously, it calculates whether the response time for each task exceeds a threshold, and sets the response threshold as follows: This value is set according to the maximum response time agreed upon in the SLA, for example, it is set to 10 ... seconds, if If the condition is met, it is recorded as a response timeout item. For example, if the response time of a task is 350 seconds, then because... The task was deemed abnormal and further categorized into a "failure + timeout" combination based on its status field. The cumulative number of tasks with failures, timeouts, and combined abnormalities were as follows: The sum of these three is the number of tasks in abnormal states. Finally, the system outputs a set of statistical data for abnormal status tasks under each hospital number.

[0104] The anomaly rate output submodule calculates the ratio of the number of abnormal tasks to the total number of tasks based on the number of tasks for each hospital number in the standard task records and abnormal status task data, and establishes an abnormal event report.

[0105] The anomaly rate output submodule calculates the number of tasks for each hospital number based on the standard task records and the anomaly status task data, setting the standard task count as follows: The number of abnormal tasks is The hospital numbers are taken from the output results of the first two sub-modules, respectively. Perform ratio calculation on the corresponding values. The result This is the hospital number. The corresponding anomaly rate, for example if , ,but Based on this, an abnormal event reporting data structure was established, the structure of which includes the hospital number. Total number of tasks Number of abnormal tasks Abnormality rate The data from each hospital will be compiled into a final output report list.

[0106] Please see Figure 2 The maintenance resource matching module includes:

[0107] The idle personnel extraction submodule extracts the current maintenance personnel schedule of the hospital, determines whether the schedule status of each person is marked as idle, filters the set of personnel with no assigned tasks, and obtains the number of available personnel.

[0108] The idle personnel extraction submodule extracts the current maintenance personnel schedule of the hospital, first reading the hospital number. The following is the shift schedule data table for the current date. ,in Number the personnel. For the scheduling status field, retrieve the field value one by one. For each record, perform a conditional operation: if... If established, then the personnel will be... Add to the current available personnel set At the same time, extract the person's free time period. Record to If a hospital has three shift records, respectively , , Then personnel It is identified as being in an idle state, and its idle time period is extracted. , This forms a set of available personnel time periods. At the same time, statistical sets Total number of people For example, in the case above, the number of available personnel is 2, and the number of available personnel corresponding to the hospital number is finally obtained.

[0109] The time matching calculation submodule extracts the task time window from the abnormal event report based on the idle time period of the corresponding personnel in the available personnel quantity data, calculates the overlap length between the time period of each idle personnel and the task window, sorts them, and generates matching degree ranking results.

[0110] The overlap length between the idle time period of each person and the task window is calculated using the following formula:

[0111] ;

[0112] Calculations are performed, in which, Indicates the first The available personnel and the first The length of overlap of all intersecting time periods between task windows. Indicates the number of idle personnel and the task window. The start and end times of the intersecting time intervals. Indicates the number of available personnel The duration of each idle period Indicates the task window number The duration of each time period This represents the absolute value operation. To represent the square root operation, This represents the summation over all intersecting time intervals. This indicates the number of time periods where there is overlap between idle personnel and the task window.

[0113] The available time of idle personnel is automatically extracted from the duty system's scheduling data by personnel number and can be quantified into a set of segments composed of start and end timestamps. The task window time is obtained from the task reception and processing intervals registered in the scheduling task record. The time intersection is obtained by set operation to obtain the intersection interval of two time periods.

[0114] For example, the idle time for maintenance personnel g is from 10:00 to 12:00, confirmed synchronously by terminal check-in and shift records, lasting 120 minutes. The task window h's time period is from 11:00 to 13:00, calculated using the task assignment time and task deadline feedback time, also lasting 120 minutes. The intersection of these two is from 11:00 to 12:00, with an intersection duration of 60 minutes. The specific parameters obtained are substituted as follows:

[0115] ;

[0116] ;

[0117] ;

[0118] Substituting into the formula, we get:

[0119] ;

[0120] The results show that the actual intersection length of the time period between the idle personnel numbered g and the task window numbered h within the scheduling cycle is 52.25 minutes. This value is used as the calculation benchmark for task matching degree ranking and is input into the subsequent ranking process.

[0121] Overlap length represents the total actual intersection of available time slots for idle personnel and task scheduling windows in the time dimension, and is a key numerical indicator for evaluating the time matching degree between personnel and tasks. This indicator not only reflects whether two types of time slots overlap on the time axis, but also quantifies the effective utilization intensity of idle resources within the task window by integrating the difference in time slot length and the degree of redundancy. The larger the value, the higher the match between personnel idle time and task window, the more concentrated the available time slots, and the higher the scheduling efficiency; the smaller the value, the more it indicates a mismatch between idle time and task arrangement or insufficient utilization, which has a decisive impact on the subsequent task allocation priority ranking.

[0122] The formula's operational logic is based on multi-level time dimension difference fusion. It uses the intersection duration as the basic overlap metric, directly reflecting the amount of overlap between idle time and task windows. A half-value of the duration difference is introduced as an adjustment term to correct the impact of the overall imbalance in the configuration of task and idle time periods on the intersection evaluation. This term is added in an additive form because the difference itself reflects potential scheduling flexibility. The square root term serves as a penalty factor to suppress situations where idle time is much longer than the intersection time. In this case, although there is an intersection, the personnel idle redundancy is high, indicating that time utilization is not concentrated. Therefore, the square root form of the difference between idle duration and intersection time is constructed to penalize the overall overlap. The introduction of this term through subtraction gives the formula a non-linear compression effect in numerical calculation, making the results more discernible. The three parts combined constitute the final overlap length value, which covers both the intersection intensity and takes into account the differences in time structure and the risk of redundant matching.

[0123] The scheduling result output submodule selects the set of personnel with the highest matching value according to the matching degree sorting result, binds the corresponding personnel to the task window, creates a scheduling pre-allocation table under the corresponding hospital number, and establishes a resource scheduling plan.

[0124] The scheduling result output submodule selects the set of personnel with the highest matching value based on the matching degree sorting results, and iterates through each task window. From the corresponding sorted set Take the first person with the highest matching value. Combine it with the task window Binding, recorded as a scheduling binding pair For already bound users, the globally available set The process involves marking tasks and removing them from subsequent matches to avoid duplicate assignments. Then, it continues processing the next task window. If multiple tasks share the same available time slot, the task with the longest overlap is assigned first. If there are identical matches, the task with the earliest available time slot is selected first, meaning the task with the shortest overlap is prioritized. The personnel responsible for this task should repeat the binding process until all task windows are assigned, ultimately completing the process for all tasks. To constitute the scheduling pre-allocation record set Each record contains the following field: Hospital Number Task Number Personnel Number Matching duration Output a scheduling pre-allocation table to establish a resource scheduling plan.

[0125] Please see Figure 2 The task reset window module includes:

[0126] The delayed task screening submodule extracts the list of assigned tasks based on the resource scheduling plan, identifies tasks that are in the unstarted state and have an assignment delay mark based on the current check-in screen maintenance task status data, and filters them in combination with the asset number corresponding to the task to obtain the set of delayed task numbers.

[0127] The delayed task screening submodule, based on the resource scheduling plan, first extracts the set of currently assigned maintenance task numbers from the scheduling pre-allocation table. Read the current status field of each task one by one. and timestamp field Determine if the task is in an unstarted state, i.e., if the condition is met. Then, read the current system time. Determine if there is an allocation delay, the delay condition is: Among them, the delay threshold Set to 30 minutes, if a certain task , , If the task delay time is 40 minutes, which is greater than the 30-minute threshold, it meets the filtering criteria and is assigned a number. Add to the list of delayed tasks And read the asset number field corresponding to the task. Combine the results into tuples Finally, the extraction operation of the set of stranded task numbers is completed.

[0128] The response combination judgment submodule extracts the priority of maintenance tasks based on each task number in the task delay number set, and filters the personnel and task window combinations in the current scheduling plan according to the response time threshold in the SLA response time limit (e.g., 15 minutes for P1 level and 2 hours for P2 level) to obtain the compliant task combination result.

[0129] The response combination judgment submodule extracts the task priority field for each task number in the task delay number set. The maximum permissible response time is determined based on the level response threshold table in the maintenance SLA requirements. ,set up Minutes , Minutes For each delayed task Extract the candidate list from the dispatch plan. and the start time of the window bound to the task. Perform a response time limit check on all personnel combinations; if the conditions are met... Then determine the combination Compliance, for example, if the task Priority is P1. , Minutes, then if personnel P3 are ,but Within minutes, tasks that meet time constraints are recorded and combined into compliant task combinations, ultimately forming a set of all task and personnel combinations that meet response time limits. .

[0130] The task mapping output submodule extracts the corresponding task number and personnel number set based on the compliant task combination result, establishes a two-way mapping from task number to personnel number, and generates a maintenance task scheduling plan.

[0131] The task mapping output submodule extracts the corresponding task number and personnel number set based on the compliance task combination result, and constructs the task number set. People gathering Establish two mapping structures in different directions, with the task-to-person mapping as follows: Personnel to task mapping If a certain task Matching personnel Then in Add key-value pairs ,exist Add To avoid duplicate mappings, the assigned records are checked to ensure that each person is assigned only once. If a person already appears in another task binding item, they are removed from the current combination. After mapping is completed, a maintenance task scheduling scheme table is constructed, with fields including task number, priority, task response time window, personnel number and its available response time start. Finally, the scheduling scheme is output as the basis for resource reallocation.

[0132] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A cloud-based hospital check-in screen maintenance system, characterized by: The system includes: The communication behavior analysis module obtains the cloud request ID, UTC timestamp and retry count of each hospital number, reads the record group with the same session ID and continuous time interval, filters it using HTTP status codes, extracts the instruction segment interval that meets the conditions by combining time sequence, establishes the frequency summary of adjacent repeated segments, and generates the request retry distribution density. The interruption time identification module obtains the fault time window based on the hospital number corresponding to the request retry distribution density, compares the instruction segment time interval with the time in the fault log, and calculates and outputs the time overlap rate. The check-in screen instruction filtering module, based on the time overlap rate, statistically analyzes the hospital check-in screen SLA service records, extracts task execution status and response time, calculates the anomaly occurrence rate, and generates an anomaly event report. The maintenance resource matching module extracts the current maintenance personnel schedule based on the hospital number corresponding to the abnormal event report, filters the set of maintenance personnel in an idle state, calculates the matching degree between the task time window and the available time slot, and generates a resource scheduling plan.

2. The cloud computing based hospital check-in kiosk maintenance system as claimed in claim 1, wherein: The request retry distribution density includes retry frequency distribution records, abnormal request concentration, and time series consistency index. The time overlap rate includes overlap duration, overlap ratio, and time matching level. The abnormal event report includes abnormal category statistics, response delay distribution data, and task status labels. The resource scheduling plan includes personnel availability mapping, task time window matching score, and preliminary scheduling priority ranking results.

3. The cloud computing based hospital check-in kiosk maintenance system as claimed in claim 1, wherein: The communication behavior parsing module includes: The request identifier extraction submodule obtains the cloud request ID, UTC timestamp, and retry count for each hospital number, reads record groups with consistent session IDs and continuous time intervals, performs group matching based on the UTC timestamp and request ID in each record group, establishes a mapping index between request time and request ID, and obtains the request sequence time index. The abnormal segment filtering submodule reads the request records under the same session ID according to the request sequence time index, filters the HTTP status code and arranges each request segment in the request order, determines whether the time difference between adjacent segments is less than the maximum allowed interval value under the same session, and extracts the abnormal request interval segment set. The retry density generation submodule counts the number of repeated request identifiers within the abnormal request interval segment set, calculates the ratio of the repetition rate of each request identifier in each segment to the length of the interval, determines whether the ratio is greater than the fixed request density benchmark value, and counts the request retry distribution density.

4. The cloud computing based hospital check-in kiosk maintenance system of claim 1, wherein: The interruption time identification module includes: The retry density screening submodule extracts the abnormal request intervals corresponding to each hospital number based on the hospital number corresponding to the request retry distribution density, counts and judges whether the request repetition rate in each interval exceeds the set request repetition rate threshold, and screens out high-frequency abnormal interval segments. The log window extraction submodule obtains the fault logs corresponding to the matching hospital number based on the time intervals in the high-frequency abnormal intervals, extracts the start and end times marked as fault states in the log records, and combines them into a continuous time range list to obtain the fault time window set. The overlap calculation output submodule calculates and outputs the time overlap rate by summarizing the intersection duration of the two time range sets based on the corresponding time ranges of the high-frequency abnormal intervals and the fault time window set, and combining it with the total duration of the corresponding abnormal intervals.

5. The cloud computing based hospital check-in kiosk maintenance system as claimed in claim 4, wherein: The time overlap rate is calculated using the following formula: ； Calculations are performed, in which, Indicates the time overlap rate. Indicates the first The start time of each overlapping time period Indicates the first The end time of each overlapping time period Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault time window segment Indicates the first The duration of each high-frequency abnormal interval. Indicates the first The duration of each fault period This represents the total number of overlapping time periods. This indicates the number of time period combinations involved in the crossover estimation. Indicates the sequence number of the high-frequency anomaly interval. This represents the total number of all high-frequency abnormal intervals.

6. The cloud computing based hospital check-in kiosk maintenance system as claimed in claim 1, wherein: The check-in screen instruction filtering module includes: The service record summary submodule assigns hospital numbers based on the time overlap rate, extracts SLA service records for the check-in screen under the matching hospital number, identifies maintenance record items that conform to IT service management standards, counts the number of service tasks and task time distribution for each hospital number, and obtains standard task records. The status response extraction submodule extracts the execution status and response time value of each task based on the service record content in the standard task record, determines whether the task status is a failure or timeout mark, summarizes the number of tasks in the corresponding status, and obtains abnormal status task data. The anomaly rate output submodule calculates the ratio of the number of abnormal tasks to the total number of tasks based on the number of tasks for each hospital number in the standard task record and the abnormal status task data, and establishes an abnormal event report.

7. The cloud computing based hospital check-in kiosk maintenance system of claim 1, wherein: The maintenance resource matching module includes: The idle personnel extraction submodule extracts the current maintenance personnel schedule of the hospital, determines whether the schedule status of each person is marked as idle, filters the set of personnel with no assigned tasks, and obtains the number of available personnel. The time matching calculation submodule extracts the task time window from the abnormal event report based on the idle time period of the corresponding personnel in the available personnel quantity data, calculates the overlap length between the time period of each idle personnel and the task window, sorts them, and generates a matching degree ranking result. The scheduling result output submodule selects the set of personnel with the highest matching value based on the matching degree sorting result, binds the corresponding personnel to the task window, creates a scheduling pre-allocation table under the corresponding hospital number, and establishes a resource scheduling plan.

8. The cloud computing based hospital check-in kiosk maintenance system of claim 7, wherein: The overlap length between the idle time period of each person and the task window is determined by the formula: ； Calculations are performed, in which, Indicates the first The available personnel and the first The length of overlap of all intersecting time periods between task windows. Indicates the number of idle personnel and the task window. The start and end times of the intersecting time intervals. Indicates the number of available personnel The duration of each idle period Indicates the task window number The duration of each time period.

9. The cloud-based hospital check-in screen maintenance system according to claim 1, characterized in that, The system also includes: Based on the resource scheduling plan, the task rescheduling window module combines the current maintenance task backlog status and asset number on the check-in screen to filter maintenance personnel binding combinations that meet the SLA response time limit, establishes a task-maintenance personnel mapping, and generates a maintenance task scheduling plan. The maintenance task scheduling scheme includes a task number and maintenance personnel matching list, a priority scheduling list, and response time limit compliance assessment results.

10. The cloud computing based hospital check-in kiosk maintenance system as claimed in claim 9, wherein: The task rescheduling window module includes: The delayed task screening submodule extracts the list of assigned tasks based on the resource scheduling plan, identifies tasks that are in the unstarted state and have an allocation delay mark according to the current check-in screen maintenance task status data, and filters them in combination with the asset number corresponding to the task to obtain a set of delayed task numbers. The response combination judgment submodule extracts the priority of the maintenance task according to each task number in the task delay number set, and filters the personnel and task window combinations in the current scheduling plan according to the response time threshold in the SLA response time limit to obtain the compliant task combination result. The task mapping output submodule extracts a corresponding task number and a personnel number set according to the compliance task combination result, establishes a bidirectional mapping of the task number to the personnel number, and generates a maintenance task scheduling scheme.

Images