Analysis method for improving efficiency through hospital operation management

Abstract

The invention relates to the technical field of data processing, and particularly discloses an analysis method for improving efficiency through hospital operation management, and the method comprises the steps: obtaining a comprehensive doctor-seeing data set of a patient; taking the patient treatment comprehensive data set as input of a classification algorithm, outputting a fluctuation category label of the time range of the shift, and obtaining a high-load time period; extracting the occupancy rate of medical equipment and the number of patients in the high-load time period and taking the same as input, and predicting the number of medical personnel scheduling; calculating a difference value between the predicted scheduling number and the scheduling number value in the same high-load time period to obtain a deviation index value; and if the deviation index value exceeds a preset deviation threshold value, optimizing the scheduling number of the medical workers according to the deviation index value to obtain the adjustment amount of the scheduling number of the medical workers. According to the analysis method for improving the efficiency through hospital operation management, the problem that the overall efficiency of a hospital is difficult to improve due to supply and demand mismatching caused by the fact that existing hospital management is not fully mined in the aspect of data relevance is solved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to an analytical method for improving efficiency through hospital operation management. Background Technology

[0002] Hospital operations management, as a core component of the healthcare service system, directly impacts the rationality of medical resource allocation, patient experience, and the overall sustainability of medical institutions. Currently, many hospitals still rely on traditional management methods, primarily using manual experience and fixed rules for resource allocation and process arrangement. This approach often struggles to respond quickly to dynamic changes in patient needs. While some hospital operational data is collected, analysis often remains at a superficial statistical level, lacking in-depth correlation analysis of multi-departmental and multi-dimensional data, leading to a significant disconnect between management decisions and actual operational conditions.

[0003] There is a high degree of coupling between patient flow, equipment usage, and staffing: when patients flock to the hospital during a certain period, if staffing and equipment are arranged based solely on historical averages, it will lead to a significant increase in queuing time. This supply-demand mismatch caused by the lack of full data correlation results in repeated blockages in the medical process, becoming the most prominent contradiction restricting the overall efficiency improvement of the hospital. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the present invention provides an analytical method for improving efficiency through hospital operation management, aiming to solve the problems in the existing technology.

[0005] The technical solution adopted by this invention to solve its technical problem is: an analytical method for improving efficiency through hospital operation management, comprising the following steps: S1: Obtain historical medical records and historical medical equipment usage records from the hospital information system, and perform preliminary integration with the current medical staff scheduling records to obtain a comprehensive patient medical record dataset; S2: Use the patient visit comprehensive dataset as input to the classification algorithm, use the classification algorithm to analyze the data of each shift's time range, output the fluctuation category label of the shift's time range, and extract the time period labeled as high load from the fluctuation category label as the high load time period. S3: For high-load periods, extract the medical equipment occupancy rate and the number of patients from the comprehensive patient visit dataset for the corresponding time range. Using the medical equipment occupancy rate and the number of patients as input, a learning algorithm is used to predict the number of medical staff to be scheduled. S4: Calculate the difference between the predicted number of shifts and the number of shifts for medical staff in the same high-load period, and obtain the deviation index value based on all the differences within a shift cycle. S5: If the deviation index value exceeds the preset deviation threshold, based on the deviation index value, the number of medical staff shifts is optimized according to the medical equipment occupancy rate and the number of patients visiting the hospital during the same high-load period, to obtain the adjustment amount of the number of medical staff shifts during the high-load period.

[0006] It is worth noting that in step S1, historical medical records, historical medical equipment usage records, and medical staff shift records are obtained from the hospital information system. The time range of each shift is determined based on the medical staff's shift schedule, and the number of staff on each shift is counted based on the medical staff's shift schedule records. The number of patients visited was counted for each shift within a specific time frame based on historical medical records. The occupancy rate of medical equipment is calculated for each shift within a time range based on historical medical equipment usage records. The number of patients visiting each shift, the occupancy rate of medical equipment, and the number of staff on duty are arranged in chronological order according to the shift cycle to obtain the time series of patient visits, the time series of medical equipment occupancy rate, and the time series of medical staff on duty, forming a comprehensive dataset of patient visits with shift information.

[0007] Preferably, in step S2, the time series of patient visits and the time series of medical equipment occupancy rate are obtained from the comprehensive patient visit dataset; The time series data of patient visits and medical equipment occupancy rates are processed, and the SVC class of the Scikit-learn library is used to output fluctuation category labels. Extract time periods labeled as high volatility to form a set of high volatility time periods; Time range data is obtained from the set of high-fluctuation time periods. The number of staff on duty in the corresponding time range is extracted from the time series of the number of medical staff on duty. If the number of staff on duty is lower than a preset threshold, the time range is determined to be a high-load time period.

[0008] Optionally, in step S2, the NumPy library is used to calculate the arithmetic mean of the number of patients visiting patients in the time series of each shift's time range as the first mean, and the average of the squared differences between each number of patients visiting patients and the first mean is calculated as the first variance. The arithmetic mean of the medical equipment occupancy rate in the time series of medical equipment occupancy rate is calculated as the second mean, and the average of the squared differences between each medical equipment occupancy rate and the second mean is calculated as the second variance. The first mean, the first variance, the second mean, and the second variance are combined to obtain a fluctuation feature vector. Using the fluctuation feature vector as input, the SVC class of the Scikit-learn library is used to classify the time range of each shift and output the fluctuation category label.

[0009] Specifically, in step S3, for high-load time periods, the occupancy rate values ​​corresponding to each high-load time period are obtained from the medical equipment occupancy rate time series, and the visitation values ​​for each high-load time period are obtained from the patient visitation number time series. The occupancy rate values ​​and visitation values ​​are combined to obtain feature data pairs. Based on the feature data pairs, the number of shifts scheduled for the corresponding time period in the current cycle is calculated using the regression coefficients and intercept parameters determined in linear regression, thus forming a dataset of predicted shift numbers.

[0010] It is worth noting that in step S3, the steps of determining the regression coefficients and intercept parameters in linear regression include: extracting the number of scheduling personnel in the corresponding time period of the historical scheduling cycle that matches the feature data pair from the historical scheduling record database, and combining the feature data pair and the extracted number of scheduling personnel to construct a training sample set. The linear regression function of the sklearn library is used to fit the training sample set to determine the regression coefficients and intercept parameters.

[0011] Specifically, in step S4, the number of medical staff shifts during high-load periods in the current period is obtained from the time series of the number of medical staff shifts, and recorded as the actual shift dataset. For the actual shift scheduling dataset, compare it with the predicted shift scheduling dataset, calculate the difference between the two in each high-load period, and generate a preliminary deviation dataset; The deviation index value is calculated using the root mean square error for the difference in the initial deviation dataset.

[0012] Preferably, in step S5, the deviation index value and the preset deviation threshold are obtained; If the deviation index value exceeds the preset deviation threshold, then obtain the medical equipment occupancy rate and the number of patients visiting each high-load time period of the current cycle. For each high-load time period, the scheduling adjustment coefficient C is obtained by multiplying the number of patients visited by the occupancy rate of medical equipment. The absolute value D of the difference between the predicted number of medical staff on duty and the number of medical staff on duty in the time series corresponding to the high-load time period is obtained. The absolute value D of the difference is multiplied by the scheduling adjustment coefficient C, and then multiplied by the adjustment coefficient to obtain the optimization direction factor. For the high-load period, the adjustment amount of the number of medical staff on duty is obtained by multiplying the actual number of medical staff on duty by the optimization direction factor and rounding up.

[0013] The beneficial effects of this invention are as follows: In the analytical method for improving efficiency through hospital operation management, historical medical records, medical equipment usage records, and current scheduling data are integrated to form a comprehensive patient visit dataset. A classification algorithm is used to identify high-load time periods, and a learning algorithm is combined to predict the appropriate number of shifts, thus solving the problem of uneven resource allocation. This solution calculates the deviation index between the predicted and actual number of shifts, and dynamically optimizes the scheduling plan based on medical equipment occupancy and patient visit volume when the deviation exceeds the standard, ensuring accurate and efficient allocation of medical and nursing staff during high-load periods. This maximizes the utilization rate of hospital resources, improves the quality of medical services and patient satisfaction, reduces the workload of medical staff, and provides scientific decision support for hospital management. Attached Figure Description

[0014] Figure 1 A flowchart illustrating an analytical method for improving efficiency through hospital operations management; Figure 2 This is a flowchart of the steps in step S2; Figure 3 This is a flowchart of the steps in step S3. Detailed Implementation

[0015] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0016] Combination Figures 1 to 3 The analytical method shown includes the following steps to improve efficiency through hospital operations management: S1: Obtain historical medical records and historical medical equipment usage records from the hospital information system, and perform preliminary integration with the current medical staff scheduling records to obtain a comprehensive patient medical record dataset; S2: Use the patient visit comprehensive dataset as input to the classification algorithm, use the classification algorithm to analyze the data of each shift's time range, output the fluctuation category label of the shift's time range, and extract the time period labeled as high load from the fluctuation category label as the high load time period. S3: For high-load periods, extract the medical equipment occupancy rate and the number of patients from the comprehensive patient visit dataset for the corresponding time range. Using the medical equipment occupancy rate and the number of patients as input, a learning algorithm is used to predict the number of medical staff to be scheduled. S4: Calculate the difference between the predicted number of shifts and the number of shifts for medical staff in the same high-load period, and obtain the deviation index value based on all the differences within a shift cycle. S5: If the deviation index value exceeds the preset deviation threshold, based on the deviation index value, the number of medical staff shifts is optimized according to the medical equipment occupancy rate and the number of patients visiting the hospital during the same high-load period, to obtain the adjustment amount of the number of medical staff shifts during the high-load period.

[0017] In the aforementioned analytical method for improving efficiency through hospital operations management, a comprehensive patient visit dataset is formed by integrating historical visit records, medical equipment usage records, and current shift data. A classification algorithm is used to identify high-load time periods, and a learning algorithm is combined to predict the appropriate number of shifts, thus addressing the problem of uneven resource allocation. This solution calculates the deviation index between the predicted and actual shift numbers, and dynamically optimizes the shift plan based on medical equipment occupancy and patient visit volume when the deviation exceeds the target, ensuring accurate and efficient allocation of medical staff during high-load periods. This maximizes hospital resource utilization, improves the quality of medical services and patient satisfaction, while reducing the workload of medical staff and providing scientific decision support for hospital management.

[0018] It is worth noting that in step S1, historical medical records, historical medical equipment usage records, and medical staff shift records are obtained from the hospital information system. The time range of each shift is determined based on the medical staff's shift schedule, and the number of staff on each shift is counted based on the medical staff's shift schedule records. The number of patients visited was counted for each shift within a specific time frame based on historical medical records. The occupancy rate of medical equipment is calculated for each shift within a time range based on historical medical equipment usage records. The number of patients visiting each shift, the occupancy rate of medical equipment, and the number of staff on duty are arranged in chronological order according to the shift cycle to obtain the time series of patient visits, the time series of medical equipment occupancy rate, and the time series of medical staff on duty, forming a comprehensive dataset of patient visits with shift information.

[0019] When retrieving historical medical records, historical medical equipment usage records, and medical staff shift records from a hospital information system, this can be achieved through a database query interface. For example, historical medical records may involve querying fields such as patient ID, visit time, and department; historical medical equipment usage records may involve querying equipment ID, start time, end time, and occupancy status; and medical staff shift records may involve nurse and doctor shift arrangements, on-call dates, and staff names.

[0020] In one possible implementation, assuming the hospital's medical staff scheduling cycle is 7 days, with each day divided into morning shift (8:00-16:00), afternoon shift (16:00-24:00), and night shift (0:00-8:00), the cycle information in the scheduling records is first parsed and mapped to the specific time interval of each shift. For example, the time range of the morning shift on the first day of the cycle is day1 8:00 to 16:00, and the time range of the afternoon shift on the second day of the cycle is day2 16:00 to 24:00. Then, the number of scheduled personnel is counted for each shift's time range by counting the number of medical staff IDs assigned within that shift's time range. For example, if the record shows that there are 5 doctors and 10 nurses on duty in the morning shift on the first day of the cycle, then the number of scheduled personnel is 15.

[0021] Based on historical medical records, the number of patients seen within each shift's time range can be statistically analyzed. This process can continue with the aforementioned time range division. Specifically, patient visits are matched with shift time ranges. For example, if a patient's visit time is 10:00 AM on January 2, 2023, and January 2, 2023 falls within the first day of the cycle, with 10:00 AM falling within the early shift range, then the number of patients seen within the shift's time range of 8:00 AM to 4:00 PM is counted. By accumulating all matching records within each shift's time range, the number of patients seen in each shift is calculated.

[0022] Based on historical medical equipment usage records, the occupancy rate of medical equipment is calculated for each shift's time range, which involves the calculation of the proportion of usage time. For example, if the total usage time X minutes for all equipment in the hospital is calculated within a shift's time range, and the total number of devices Y in the hospital is obtained, then the medical equipment occupancy rate for that shift's time range is calculated as X / (Y). 8 60), of which 8 60 represents the total number of minutes within the time range of a single shift.

[0023] The number of patients visiting each shift, the occupancy rate of medical equipment, and the number of staff on duty are arranged chronologically according to the shift cycle to obtain the corresponding time series, forming a comprehensive patient visit dataset with shift information. This can be achieved using a time sorting algorithm. Specifically, the data is arranged chronologically, such as day1 8:00 to 16:00, day1 16:00 to 24:00, etc., to generate a time series of patient visits such as [200, 150, 50, ...]. The time series of medical equipment occupancy rate and the time series of medical staff on duty are generated similarly. This supports the application of subsequent prediction models.

[0024] Preferably, in step S2, the time series of patient visits and the time series of medical equipment occupancy rate are obtained from the comprehensive patient visit dataset; The time series data of patient visits and medical equipment occupancy rates are processed, and the SVC class of the Scikit-learn library is used to output fluctuation category labels. Extract time periods labeled as high volatility to form a set of high volatility time periods; Time range data is obtained from the set of high-fluctuation time periods. The number of staff on duty in the corresponding time range is extracted from the time series of the number of medical staff on duty. If the number of staff on duty is lower than a preset threshold, the time range is determined to be a high-load time period.

[0025] Optionally, in step S2, the NumPy library is used to calculate the arithmetic mean of the number of patients visiting patients in the time series of each shift's time range as the first mean, and the average of the squared differences between each number of patients visiting patients and the first mean is calculated as the first variance. The arithmetic mean of the medical equipment occupancy rate in the time series of medical equipment occupancy rate is calculated as the second mean, and the average of the squared differences between each medical equipment occupancy rate and the second mean is calculated as the second variance. The first mean, the first variance, the second mean, and the second variance are combined to obtain a fluctuation feature vector. Using the fluctuation feature vector as input, the SVC class of the Scikit-learn library is used to classify the time range of each shift and output the fluctuation category label.

[0026] For each shift's time series of patient visits and medical equipment occupancy rate, calculating the average and variance of patient visits can be understood as a way to measure the stability of the number of patients.

[0027] Specifically, after extracting the fluctuation feature vector, classification methods can be used to determine the fluctuation category label for each shift's time range. For example, a hospital might categorize fluctuations into three types: high fluctuation, medium fluctuation, and low fluctuation. For night shifts, if both the variance in the number of patients and the variance in equipment occupancy exceed the corresponding preset standards, it is classified as high fluctuation. This classification method helps hospital administrators quickly identify periods of resource strain.

[0028] In one embodiment, for sets of highly volatile time periods, further analysis can be conducted by combining the time series of medical staff shift schedules. For example, assuming the number of shifts scheduled from 8 PM to 10 PM on the fifth day of the period is 5, while the preset minimum threshold is 8, this time period is identified as a high-load period. This analysis process can help hospitals identify periods of insufficient staffing, providing a basis for subsequent shift optimization.

[0029] Specifically, in step S3, for high-load time periods, the occupancy rate values ​​corresponding to each high-load time period are obtained from the medical equipment occupancy rate time series, and the visitation values ​​for each high-load time period are obtained from the patient visitation number time series. The occupancy rate values ​​and visitation values ​​are combined to obtain feature data pairs. Based on the feature data pairs, the number of shifts scheduled for the corresponding time period in the current cycle is calculated using the regression coefficients and intercept parameters determined in linear regression, thus forming a dataset of predicted shift numbers.

[0030] In one embodiment, for identified high-load time periods, the medical equipment occupancy rate for the corresponding time period is extracted from the medical equipment occupancy rate time series, and the number of patients visiting the hospital within the same time range is extracted from the patient visit number time series. These two data pairs are combined to form a feature data pair. This feature data pair can intuitively reflect the degree of resource pressure. Combining these values ​​yields a feature data pair, such as a medical equipment occupancy rate of 85% and a number of patients visiting the hospital of 65, thus forming a multi-dimensional feature representation.

[0031] Furthermore, based on the regression coefficients and intercept parameters obtained from the fitting, the predicted number of medical staff shifts is calculated using real-time or recently extracted feature data for the same high-load time period in the current cycle. These prediction results can be compiled into a predicted shift quantity dataset, which can be directly referenced by the hospital's scheduling system.

[0032] It is worth noting that in step S3, the steps of determining the regression coefficients and intercept parameters in linear regression include: extracting the number of scheduling personnel in the corresponding time period of the historical scheduling cycle that matches the feature data pair from the historical scheduling record database, and combining the feature data pair and the extracted number of scheduling personnel to construct a training sample set. The linear regression function of the sklearn library is used to fit the training sample set to determine the regression coefficients and intercept parameters.

[0033] Specifically, for the medical equipment occupancy rate and patient visits in the feature data pairs used to construct the training sample set, these are extracted from the time series of medical equipment occupancy rates and patient visits generated in historical records. The high-load time periods obtained in step S2 are then used to find the corresponding scheduling time ranges within other scheduling cycles in the historical records. The medical equipment occupancy rate and patient visits within these time ranges are extracted. Additionally, the number of scheduled personnel within these time ranges is also extracted as the number of scheduled personnel for the corresponding time period in the historical scheduling cycle that matches the feature data pairs. Combining these numbers of scheduled personnel with the corresponding feature data pairs constitutes a complete training sample. It should be noted that by traversing multiple historical cycles, a large number of such samples can be accumulated to form a training sample set, thereby covering various situations with different load levels.

[0034] The process of fitting a model using linear regression from the sklearn library begins by using feature data pairs from the training sample set as independent variables, and inputting the number of employees in the corresponding time period of the historical scheduling cycle that matches the feature data pairs as the dependent variable. Linear regression is a linear regression method that uses the least squares method to solve for the regression coefficients and intercept parameters.

[0035] Specifically, in step S4, the number of medical staff shifts during high-load periods in the current period is obtained from the time series of the number of medical staff shifts, and recorded as the actual shift dataset. For the actual shift scheduling dataset, compare it with the predicted shift scheduling dataset, calculate the difference between the two in each high-load period, and generate a preliminary deviation dataset; The deviation index value is calculated using the root mean square error for the difference in the initial deviation dataset.

[0036] In one embodiment, the actual shift schedule dataset is compared with the predicted shift number dataset. The comparison process involves aligning the two datasets to the same high-load time period and then calculating the difference one by one. For example, if the actual number is 8 people but the predicted number is 9 people, the difference is -1 people. This calculation covers all personnel types and high-load time periods, and finally generates a preliminary deviation dataset, which records the difference for each high-load time period.

[0037] In one embodiment, the root mean square error (RMSE) is a statistical indicator used to quantify the average deviation between predicted and actual values, and its formula is: , Let represent the number of medical staff scheduled for the i-th high-load time period in the actual shift scheduling dataset. To predict the number of healthcare worker shifts in the i-th high-load time period of the predicted shift count dataset, where n is the total number of high-load time periods. This represents the deviation index value.

[0038] Preferably, in step S5, the deviation index value and the preset deviation threshold are obtained; If the deviation index value exceeds the preset deviation threshold, then obtain the medical equipment occupancy rate and the number of patients visiting each high-load time period of the current cycle. For each high-load time period, the scheduling adjustment coefficient C is obtained by multiplying the number of patients visited by the occupancy rate of medical equipment. The absolute value D of the difference between the predicted number of medical staff on duty and the number of medical staff on duty in the time series corresponding to the high-load time period is obtained. The absolute value D of the difference is multiplied by the scheduling adjustment coefficient C, and then multiplied by the adjustment coefficient to obtain the optimization direction factor. For the high-load period, the adjustment amount of the number of medical staff on duty is obtained by multiplying the actual number of medical staff on duty by the optimization direction factor and rounding up.

[0039] If the deviation index value exceeds the preset deviation threshold, the system will trigger a subsequent optimization mechanism to ensure that the scheduling is more in line with actual needs. For example, when the deviation index value of 2.5 exceeds the threshold of 1.0, the system automatically obtains the medical equipment occupancy rate and the number of patients visiting each high-load time period within the current cycle. This data comes from the hospital's real-time monitoring system. For instance, in one high-load time period, the medical equipment occupancy rate is recorded as 85%, and the number of patients visiting is 120. These indicators are obtained by collecting relevant records in real time through sensors and the registration system and then statistically analyzing them.

[0040] In one embodiment, the scheduling adjustment coefficient C reflects the overall load intensity during high-load periods. For example, during a high-load period, the number of patients visiting 120 multiplied by the occupancy rate of 0.85 yields a coefficient of 102. This calculation process emphasizes the interactive influence between patient flow and equipment utilization, serving as a quantitative basis for adjusting the scheduling. Through this multiplication operation, the system can dynamically capture load changes, avoiding deviations caused by relying solely on a single indicator, thereby providing more accurate guidance for subsequent steps.

[0041] After obtaining the scheduling adjustment coefficient, the system will extract the absolute value D of the difference between the predicted number of medical staff on duty and the actual number of medical staff on duty in the time series of the predicted number of medical staff on duty for the high-load period. This absolute value D is directly inherited from the previous comparison. For example, if the predicted number of doctors on duty is 10 and the actual number is 8, then D is 2. This process ensures that the adjustment is based on the actual deviation and connects the prediction and the reality.

[0042] The optimization direction factor determines the direction and magnitude of the adjustment. For example, D is 2 multiplied by C is 102 and then multiplied by the determination coefficient 0.001, resulting in an optimization direction factor of 0.204. This calculation not only considers the magnitude of the deviation but also incorporates the load intensity, thereby guiding the scheduling to tilt towards a better direction. In terms of operations, this helps hospitals to efficiently allocate manpower during periods of high load, reduce waiting time, and improve patient satisfaction.

[0043] For the high-load period, the adjustment amount of the medical staff shift number is obtained by multiplying the corresponding actual number of medical staff on duty by the optimization direction factor. For example, if the actual number of medical staff on duty is 8, multiplying by the factor 0.204 gives 1.632, and then rounding up gives 2, that is, the adjustment amount of the medical staff shift number is 2. Then, the adjustment amount of the medical staff shift number is added to the actual number of medical staff on duty to get the adjusted number of medical staff on duty as 10. Finally, the number is adjusted to a more suitable number. This step forms a complete optimization closed loop to ensure that the shift schedule adapts to dynamic changes.

[0044] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. An analytical method for improving efficiency through hospital operations management, characterized in that, Includes the following steps: S1: Obtain historical medical records and historical medical equipment usage records from the hospital information system, and perform preliminary integration with the current medical staff scheduling records to obtain a comprehensive patient medical record dataset; S2: Use the patient visit comprehensive dataset as input to the classification algorithm, use the classification algorithm to analyze the data of each shift's time range, output the fluctuation category label of the shift's time range, and extract the time period labeled as high load from the fluctuation category label as the high load time period. S3: For high-load periods, extract the medical equipment occupancy rate and the number of patients from the comprehensive patient visit dataset for the corresponding time range. Using the medical equipment occupancy rate and the number of patients as input, a learning algorithm is used to predict the number of medical staff to be scheduled. S4: Calculate the difference between the predicted number of shifts and the number of shifts for medical staff in the same high-load period, and obtain the deviation index value based on all the differences within a shift cycle. S5: If the deviation index value exceeds the preset deviation threshold, based on the deviation index value, the number of medical staff shifts is optimized according to the medical equipment occupancy rate and the number of patients visiting the hospital during the same high-load period, to obtain the adjustment amount of the number of medical staff shifts during the high-load period.

2. The analytical method for improving efficiency through hospital operation management according to claim 1, characterized in that: In step S1, historical medical records, historical medical equipment usage records, and medical staff shift records are obtained from the hospital information system. The time range of each shift is determined based on the medical staff's shift schedule, and the number of staff on each shift is counted based on the medical staff's shift schedule records. The number of patients visited was counted for each shift within a specific time frame based on historical medical records. The occupancy rate of medical equipment is calculated for each shift within a time range based on historical medical equipment usage records. The number of patients visiting each shift, the occupancy rate of medical equipment, and the number of staff on duty are arranged in chronological order according to the shift cycle to obtain the time series of patient visits, the time series of medical equipment occupancy rate, and the time series of medical staff on duty, forming a comprehensive dataset of patient visits with shift information.

3. The analytical method for improving efficiency through hospital operation management according to claim 1, characterized in that: In step S2, the time series of patient visits and the time series of medical equipment occupancy rate are obtained from the comprehensive patient visit dataset; The time series data of patient visits and medical equipment occupancy rates are processed, and the SVC class of the Scikit-learn library is used to output fluctuation category labels. Extract time periods labeled as high volatility to form a set of high volatility time periods; Time range data is obtained from the set of high-fluctuation time periods. The number of staff on duty in the corresponding time range is extracted from the time series of the number of medical staff on duty. If the number of staff on duty is lower than a preset threshold, the time range is determined to be a high-load time period.

4. The analytical method for improving efficiency through hospital operation management according to claim 3, characterized in that: In step S2, the NumPy library is used to calculate the arithmetic mean of the number of patients visiting patients in the time series of each shift's time range as the first mean, and the average of the squared differences between each number of patients visiting patients and the first mean is calculated as the first variance. The arithmetic mean of the medical equipment occupancy rate in the time series of medical equipment occupancy rate is calculated as the second mean, and the average of the squared differences between each medical equipment occupancy rate and the second mean is calculated as the second variance. The first mean, the first variance, the second mean, and the second variance are combined to obtain the fluctuation feature vector. Using the fluctuation feature vector as input, the SVC class of the Scikit-learn library is used to classify the time range of each shift and output the fluctuation category label.

5. The analytical method for improving efficiency through hospital operation management according to claim 1, characterized in that: In step S3, for high-load time periods, the occupancy rate values ​​corresponding to each high-load time period are obtained from the medical equipment occupancy rate time series, and the visitation values ​​for each high-load time period are obtained from the patient visitation number time series. The occupancy rate values ​​and visitation values ​​are combined to obtain feature data pairs. Based on the feature data pairs, the number of shifts scheduled for the corresponding time period in the current cycle is calculated using the regression coefficients and intercept parameters determined in linear regression, thus forming a dataset of predicted shift numbers.

6. The analytical method for improving efficiency through hospital operation management according to claim 5, characterized in that: In step S3, the steps of using the regression coefficients and intercept parameters determined in linear regression include: extracting the number of scheduling personnel in the corresponding time period of the historical scheduling cycle that matches the feature data pair from the historical scheduling record database, and combining the feature data pair and the extracted number of scheduling personnel to construct a training sample set. The linear regression function of the sklearn library is used to fit the training sample set to determine the regression coefficients and intercept parameters.

7. The analytical method for improving efficiency through hospital operation management according to claim 5, characterized in that: In step S4, the number of medical staff shifts during the high-load period in the current period is obtained from the time series of the number of medical staff shifts, and recorded as the actual shift dataset. For the actual shift scheduling dataset, compare it with the predicted shift scheduling dataset, calculate the difference between the two in each high-load period, and generate a preliminary deviation dataset; The deviation index value is calculated using the root mean square error for the difference in the initial deviation dataset.

8. The analytical method for improving efficiency through hospital operation management according to claim 1, characterized in that: In step S5, the deviation index value and the preset deviation threshold are obtained; If the deviation index value exceeds the preset deviation threshold, then obtain the medical equipment occupancy rate and the number of patients visiting each high-load time period of the current cycle. For each high-load time period, the scheduling adjustment coefficient C is obtained by multiplying the number of patients visited by the occupancy rate of medical equipment. The absolute value D of the difference between the predicted number of medical staff on duty and the number of medical staff on duty in the time series corresponding to the high-load time period is obtained. The absolute value D of the difference is multiplied by the scheduling adjustment coefficient C, and then multiplied by the adjustment coefficient to obtain the optimization direction factor. For the high-load period, the adjustment amount of the number of medical staff on duty is obtained by multiplying the actual number of medical staff on duty by the optimization direction factor and rounding up.

Images