A department-based automatic scheduling system and method for medical staff

Abstract

The present invention discloses a department-based automatic scheduling system and method for medical staff, relating to the technical field of intelligent scheduling. The system comprises a pre-scheduling module, a scheme evaluation module, an emergency response module and a temporary adjustment module which operate in sequence. The technical key points are as follows: the system can perform secondary priority sorting on nurses on leave according to the information and basic information of the responding nurses; by establishing a qualification quantification value library, the qualification level of the nurses is quantified, and combined with the proportion of the number of times marked in a fixed period, the priority index of each qualified nurse is generated through an exponential calculation function model, thereby realizing refined sorting of the nurses' priorities; the scheme not only takes into account the qualification level of the nurses, but also fully considers the response speed and enthusiasm of the nurses to emergency situations, thereby being able to arrange the nurses' schedules more scientifically and reasonably, and effectively solving the problems of unreasonable priority sorting and low scheduling efficiency in hospital scheduling management.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent scheduling, and in particular to a department-based automatic scheduling system and method for medical staff. Background Art

[0002] Intelligent scheduling is a process that uses modern information technology and algorithms to automatically or semi-automatically manage the scientific and reasonable scheduling of nurses and other medical personnel according to the actual needs of hospitals or medical institutions. It takes into account multiple factors and uses multiple models or algorithms to realize automatic scheduling of nurses in each department. The following contents may exist when scheduling, for example: 1. If there is a temporary vacation tomorrow, there will be a text message reminder if there is a change in the substitute; 2. Statistics of work, vacation and night shift time are kept every year; 3. There is a statistical time for various leaves such as maternity leave, marriage leave, work leave, personal leave, etc.; 4. If there is an unreasonable arrangement in the daily schedule, the system will give a prompt to avoid missing a shift; 5. The scheduling system will generate a schedule with one click. The working hours requirements of each department are different. By adjusting the mode to stipulate the time of each shift, enter the name and level of hierarchy, and realize one-click scheduling; 6. If you encounter a few hours of overtime, you can record it in the stored leave.

[0003] The existing authorization announcement number is CN118230923B, and the technical solution provided in the system and method for intelligent obstetric scheduling based on role collaboration includes: a personnel management module for recording basic information, qualifications, skills, experience and other data of obstetric personnel, and evaluating personnel capabilities; a role management module for defining and managing the permissions and access levels of different user roles in the system, and determining the number and types of roles required for daily scheduling; a scheduling rule management module for defining the rules and constraints for obstetric scheduling; a scheduling generation module for automatically generating reasonable obstetric scheduling, and allowing supervisors or administrators to manually adjust the schedule to meet special needs; the present invention can improve the work efficiency of medical staff, optimize resource utilization, improve patient care experience, and ensure the work-life balance of medical staff; although the solution also provides solutions to special situations, the special situations mostly cover scheduling conflicts and personnel preference issues, and do not involve emergency response solutions for staff shortages;

[0004] Another document, authorized by CN111883241B, states that while this plan provides a priority-based scheduling scheme, the priority is determined by radiation dose, and the priority cannot be adjusted twice or multiple times.

[0005] In combination with the above documents and existing technologies, when traditionally scheduling nurses in various departments within a hospital institution, although genetic algorithms can be used to generate multiple candidate scheduling plans, manual reference or selection is still required when finally determining the required scheduling plan. At the same time, the established scheduling plan cannot cope with sudden surgical arrangements, and it is usually necessary to actively notify personnel on leave or dispatch idle nurses from other departments. The priority of the notification order is often adjusted based on a single factor. For example, the higher the qualifications of the nurse, the higher the priority. However, there are some nurses with higher qualifications but slower response speeds. At the same time, some nurses with lower priority rankings can respond actively. This results in problems of unreasonable priority ranking and low scheduling efficiency in the existing hospital scheduling management system. Summary of the Invention

[0006] (1) Technical problems solved

[0007] In view of the deficiencies of the prior art, the present invention provides a department-based automatic scheduling system and method for medical staff, and solves the problems raised in the background technology by running the automatic scheduling system.

[0008] (2) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] A department-based automatic scheduling system and method for medical staff, the system comprising:

[0011] The pre-scheduling module summarizes the work requirements of the target department and the basic information of nurses, and uses a genetic algorithm to generate several candidate plans; the work requirements include at least the number of patients and the surgical schedule;

[0012] The plan evaluation module, after nurses receive candidate plans, builds an evaluation system engine based on their feedback and compares the human resource utilization under different candidate plans. It then combines the future work arrangements of the target department to come up with a preliminary plan.

[0013] The emergency response module prioritizes nurses on leave and sends temporary shift notifications when an emergency occurs. If a nurse with a higher priority fails to respond within the specified time, the shift is postponed to a nurse with a lower priority until a nurse responds. If no nurse responds, the call-up mechanism is triggered. The proportion of call-up mechanisms triggered within the predetermined period is simultaneously recorded. If the proportion exceeds the expected value, the plan adjustment instruction is executed.

[0014] The temporary adjustment module records the information of the responding nurses and conducts secondary priority sorting of the nurses on leave based on the basic information within a fixed period, and provides feedback on the sorting results.

[0015] Furthermore, the surgical schedule includes the date, time, type of surgery, and the specialty of the nurse required; the basic information of the nurse includes the nurse's name, qualifications, specialty, current scheduling status, and vacation plans.

[0016] Furthermore, the process of using genetic algorithm to generate several scheduling plans is as follows:

[0017] Coding: Integer coding is used. Each shift plan is represented as a chromosome. Each gene on the chromosome represents the working status of a nurse in a set time period.

[0018] Initial population: Randomly generate multiple scheduling plans as the initial population, each plan meets the set constraints;

[0019] Fitness function: Design a fitness function to score the quality of the scheduling plan and calculate the fitness of each scheduling plan. The quality of the scheduling plan includes at least the balance of nurse workload, the satisfaction of patient care needs, and the matching of professional skills.

[0020] Selection: Use the roulette wheel selection strategy to select individuals with satisfactory fitness from the current population as parents;

[0021] Crossover: Perform crossover operations on the selected parent generation to generate a new child generation schedule;

[0022] Mutation: Randomly mutate the offspring scheduling plan;

[0023] Iteration: Repeat the selection, crossover and mutation operations until the predetermined number of iterations is reached. After the iteration is completed, several scheduling schemes are selected from the final population as candidate schemes.

[0024] Furthermore, the nurses’ feedback data indicates the nurses’ satisfaction with each candidate solution. If the nurses are satisfied, the corresponding candidate solution is marked as 1; if they are dissatisfied, the corresponding candidate solution is marked as 0.

[0025] The utilization of human resources is quantitatively expressed through the utilization rate of human resources;

[0026] The calculation method of human resource utilization rate is:

[0027] Human resource utilization rate = actual human resources used / total human resources × 100%.

[0028] Furthermore, the process of running the evaluation system engine is as follows:

[0029] Based on the future work arrangement status of the target department and the total number of nurses in the target department, a mathematical calculation model is built to generate the required human resource standard value, and the candidate plans corresponding to the human resource utilization exceeding the human resource standard value are screened out. The comprehensive indicators obtained after weighted calculation are based on the difference between the human resource utilization rate and the human resource standard value, as well as the cumulative value of the feedback data corresponding to each candidate plan obtained after screening. The plans are sorted from large to small to generate a ranking table. The candidate plan corresponding to the first place in the ranking table is the preliminary plan.

[0030] Furthermore, the future work arrangement status under the target department represents: the total number of surgeries scheduled in the future preset period;

[0031] When building a mathematical calculation model, the formula is based on:

[0032] ;

[0033] Where R_min represents the required human resource standard value, T represents the total number of surgeries scheduled within the preset period, Nr represents the total number of nurses in the target department, and S i represents the number of nurses required for the i-th operation.

[0034] Furthermore, the weighted calculation formula corresponding to the obtained comprehensive index is as follows:

[0035] ;

[0036] Where Zb x The x in the formula represents the candidate number, Zb represents the comprehensive index, and R x represents the human resource utilization rate under the corresponding candidate solution x, Ac x It represents the cumulative value of feedback data under the corresponding candidate solution, α and β both represent weight coefficients, and α+β=1.

[0037] Furthermore, the process of prioritizing nurses on leave is as follows: obtaining the type corresponding to the current emergency status and matching it with the appropriate expertise, extracting nurses who both meet the appropriate expertise and are on leave from the basic information of each nurse, and sorting them from high to low according to their qualifications, with qualifications being positively correlated with priority;

[0038] The triggered recruitment mechanism is to recruit nurses from the entire hospital who have appropriate expertise and are on vacation, and the nurses' response speed is positively correlated with the call priority;

[0039] After recording the percentage of times the solicitation mechanism is triggered within the scheduled period, record the number of all relevant events within the scheduled period, including the number of events that triggered the solicitation mechanism and the number of events for which temporary scheduling notifications were sent, i.e., the total number of events. Calculate the percentage of times the solicitation mechanism is triggered, i.e., the number of events that triggered the solicitation mechanism divided by the total number of events.

[0040] The content of the executed scheme adjustment instruction is: extract the corresponding candidate scheme with the serial number of the first place + 1 in the original sorting table, and use it as the adjusted preliminary scheme, and then continue to record the proportion of triggering the solicitation mechanism within the predetermined period in the next cycle. If the scheme adjustment instruction is executed again, the corresponding candidate scheme with the serial number of the first place + 2 will continue to be used as the adjusted preliminary scheme, and the cycle will continue until there is no response.

[0041] Furthermore, the information of the responding nurse is recorded and represented based on the following: the responding nurse is marked;

[0042] The process for secondary prioritization of nurses on leave is as follows:

[0043] Establish a qualification quantitative value library, match the corresponding qualification level according to the qualifications of the nurses in the basic information, and the qualification degree is positively correlated with the qualification level; in response to the type corresponding to the current emergency status and matching the appropriate expertise, extract nurses who meet the appropriate expertise and are on vacation and shift status from the basic information of each nurse, determine the qualification level of each nurse through the qualification quantitative value library, and collect and calculate the proportion of the number of times each nurse is marked in a fixed period, build an index calculation function model to generate the priority index corresponding to each qualified nurse, and sort them from large to small according to the priority index, and the priority index is positively correlated with the priority level;

[0044] Among them, when running the exponential calculation function model, the formula is based on:

[0045] ;

[0046] ;

[0047] Among them, Rt, Mq and Td represent the proportion of the number of times marked, the number of times the corresponding nurse is marked in a fixed period and the fixed period respectively, Irt represents the priority index, Q represents the qualification level of the corresponding nurse, and k is the adjustment coefficient, with a value range of [0, 1].

[0048] A department-based automatic scheduling system and method for medical staff includes the following steps:

[0049] S1. Summarize the work requirements of the target department and the basic information of nurses, and use a genetic algorithm to generate several candidate solutions; the work requirements at least include the number of patients and surgical arrangements;

[0050] S2. After the nurses receive the candidate solutions, they will build an evaluation system engine based on their feedback and compare the human resource utilization under different candidate solutions. This engine will then combine the future work schedule of the target department to arrive at a preliminary solution.

[0051] S3. When an emergency occurs, prioritize the nurses on leave and send a temporary shift notification. If the nurses with the highest priority do not respond within the specified time, the shift will be postponed to the nurses with the lowest priority until a nurse responds. If no nurse responds, the recruitment mechanism is triggered. The proportion of nurses who trigger the recruitment mechanism within the predetermined period is simultaneously recorded. If the proportion exceeds the expected value, the plan adjustment instruction is executed.

[0052] S4. Within a fixed period, record and prioritize the nurses on leave based on the information of the responding nurses and the basic information, and provide feedback on the ranking results.

[0053] (3) Beneficial effects

[0054] The present invention provides a department-based automatic scheduling system and method for medical staff, which has the following beneficial effects:

[0055] 1. This plan comprehensively considers nurse feedback and the human resource utilization under different candidate plans. By building an evaluation system engine and combining it with the target department's future work schedule, it intelligently selects a preliminary plan. This not only improves the efficiency of scheduling, but also ensures the rationality of the plan in human resource allocation and the superiority of nurse satisfaction.

[0056] 2. This plan quickly prioritizes nurses on leave in emergencies and sends temporary scheduling notices, effectively responding to sudden surgical arrangements in departments or situations where nurses on duty are unexpectedly unable to continue working. Combined with the design of the solicitation mechanism, it further ensures timely response to emergencies. By recording the proportion of the solicitation mechanism triggered within the predetermined period and comparing it with the expected value, it can promptly identify problems in the preliminary plan. By executing the plan adjustment instruction, the next candidate plan with the highest ranking will be used as the adjusted preliminary plan, completing the optimization and adjustment of the scheduling plan. This not only improves the hospital's ability to respond to emergencies, but also ensures the flexibility and practicality of the scheduling plan, effectively solving the technical problems of untimely emergency response and inflexible plan adjustment in hospital scheduling management.

[0057] 3. This program conducts secondary prioritization of nurses on leave based on the responding nurses’ information and basic information;

[0058] By establishing a qualification quantitative value database, the qualification level of nurses is quantified, and combined with the proportion of marked times within a fixed period, the priority index of each qualified nurse is generated through an exponential calculation function model, thereby achieving a refined sorting of nurse priorities; not only the qualification level of nurses is taken into account, but also the response speed and enthusiasm of nurses to emergencies, so that the nurses' schedule can be arranged more scientifically and reasonably; to a certain extent, it can provide more opportunities for nurses with a positive response attitude, effectively solving the problems of unreasonable priority sorting and low scheduling efficiency in hospital scheduling management. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a modular schematic diagram of a department-based automatic scheduling system and method for medical staff in the present invention;

[0060] Figure 2 The figure is a schematic diagram of the overall process of an automatic scheduling system and method for medical staff based on departments in the present invention. DETAILED DESCRIPTION

[0061] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0062] Example 1:

[0063] See also Figure 1 ,This embodiment provides a department-based automatic scheduling system and method for medical staff;

[0064] The system includes a pre-scheduling module, a plan evaluation module, an emergency response module, and a temporary adjustment module, which operate in sequence. The pre-scheduling module and the plan evaluation module serve as the initial operation phase of the system; the emergency response module and the temporary adjustment module serve as the system's supplement or the phase for responding to emergencies.

[0065] System Overview

[0066] The application scenario of this system is designed for the arrangement of internal medicine and surgical operations, aiming to achieve efficient and reasonable scheduling of nursing resources through intelligent algorithms and flexible response mechanisms; the system can handle surgical arrangements known in advance, and at the same time, for emergency surgical needs, quickly and accurately notify and arrange suitable nurses for scheduling, and provide feedback based on the scheduling situation, thereby changing the order of suitable nurses, which can improve the response time to emergency surgeries to a certain extent, ensuring the rationality of scheduling and improving response efficiency, and can make adjustments to patient treatment more quickly to ensure that patients receive comprehensive care or treatment.

[0067] Functional module design

[0068] The pre-scheduling module summarizes the work requirements of the target department and the basic information of nurses (the data summarized here corresponds to Figure 1 The data source in the ,genetic algorithm is used to generate several candidate solutions;

[0069] Among them, work requirements include at least the number of patients and surgical arrangements;

[0070] In addition, the complexity of the disease can also be included;

[0071] Number of patients and complexity of their condition:

[0072] The number of patients in medical and surgical wards, as well as the condition complexity score of each patient, are obtained through the hospital information system (HIS). The condition complexity score can be determined based on the patient's diagnosis, treatment plan, and expected care needs. Although the specific calculation method may vary between hospitals and departments, it generally involves multiple factors. In this embodiment, the condition complexity score is calculated using the following example.

[0073] For example, the relative weight of DRG (Diagnosis Related Group) is used for calculation. Each DRG has a corresponding weight that reflects its resource consumption and the complexity of the disease. The more complex the disease, the more medical resources are required, and the higher the DRG weight. The hospital can count the number or weight of each patient classified into a certain DRG, and calculate the disease complexity score (such as CMI value) based on the total number of cases.

[0074] Surgery arrangements:

[0075] Obtain the date, time, type (medical / surgical) of the operation, and the specialty of the required nurse (i.e., the basic information of the corresponding nurse) from the operating room management system;

[0076] Basic information of the nurse:

[0077] Obtain the nurse's name, qualifications (e.g., licensed practical nurse, charge nurse, etc.), specialty (e.g., cardiovascular nursing, trauma nursing, etc.), current schedule status (e.g., on duty, on vacation, etc.), and vacation plan from the human resources management system;

[0078] Based on the aggregated data, a genetic algorithm is used to generate multiple scheduling solutions. The genetic algorithm is an optimization algorithm that simulates natural selection and heredity mechanisms and is suitable for solving complex scheduling problems.

[0079] The process of using genetic algorithm to generate several scheduling schemes is as follows:

[0080] Encoding: Integer encoding is used, with each scheduling plan represented as a chromosome. Each gene on the chromosome represents the work status of a nurse in a certain time period (e.g., 1 for working, 0 for resting). Initial population: Multiple scheduling plans are randomly generated as the initial population, and each plan meets the set constraints (e.g., nurse qualifications, vacation plans).

[0081] Fitness function: Design a fitness function to score the quality of the scheduling plan (such as nurse workload balance, patient care needs satisfaction, professional skills matching, etc.) and calculate the fitness of each scheduling plan; Selection: Use a roulette wheel selection strategy to select individuals with satisfactory fitness from the current population as parents; Crossover: Perform a crossover operation on the selected parents to generate a new offspring scheduling plan; Crossover operations include either single-point crossover or multi-point crossover; Mutation: Randomly mutate the offspring scheduling plan to increase the diversity of the population; Mutation operations include replacing gene values and exchanging gene positions; Iteration: Repeat the selection, crossover, and mutation operations until the predetermined number of iterations is reached or the fitness no longer significantly improves;

[0082] After the iteration, several scheduling schemes are selected from the final population as candidate schemes;

[0083] When designing the fitness function and calculating the fitness of each scheduling scheme, the fitness function is defined as:

[0084] Fit=a1*WB+a2*PC+a3*SD;

[0085] Where, Fit represents fitness, WB, PC, and SD represent the balance of nurse workload, patient care quality, and surgical requirement satisfaction, respectively; a1, a2, and a3 are weight coefficients, and their values range from [0 to 1].

[0086] Nurses' workload balance:

[0087] The calculation method is to calculate the total workload of each nurse: according to the scheduling plan, count the number of hours or shifts each nurse works within the scheduling cycle (such as a week); calculate the standard deviation of the workload: measure the degree of workload variation among nurses; the smaller the standard deviation, the more balanced the workload among nurses; normalize the workload standard deviation: normalize the workload standard deviation; the normalization here is to directly subtract the ratio of the standard deviation to the maximum possible standard deviation from 1, so as to ensure that the normalized value range is between [0, 1];

[0088] formula:

[0089] ;

[0090] The maximum possible standard deviation is calculated under extreme circumstances (e.g., all work is assigned to the same nurse). For example, suppose there are three nurses, A, B, and C, whose total weekly workload is 40 hours, 35 hours, and 45 hours, respectively. The calculated standard deviation of the workload is 4.08 hours (assuming this is a sample value, and the calculated standard deviation is the overall standard deviation). The maximum possible standard deviation is the standard deviation when all work is assigned to nurse A, which is 56.57 hours. Therefore, WB ≈ 0.93.

[0091] Quality of Patient Care:

[0092] Calculation method: Assess the nursing quality of each nurse: This can be done based on historical data, patient satisfaction surveys, nursing error rates, and other indicators. Calculate the nursing quality score: Based on the scheduling plan, weight the sum of the quality scores of the nurses responsible for nursing care in each time period. Normalize the nursing quality score to a value between [0, 1]. The normalization process here is to directly calculate the weighted sum of the quality scores of the nurses responsible for nursing care in each time period and divide it by the maximum quality score of the nurses responsible for nursing care in each time period.

[0093] formula:

[0094] ;

[0095] Among them, Q is the number of time periods in the scheduling cycle, and nurses k The weight factor corresponding to the nurse in the kth time period (reflecting the workload or importance of each time period), the value range is greater than 0, and the quality score kis the nursing quality score of the nurse in charge of nursing in the kth time period, and the maximum single time period quality score is the maximum possible quality score of each time period; for example: assuming there are 3 time periods, each time period has a corresponding nursing quality score (such as nurse A gets 8 points, nurse B gets 7 points, and nurse C gets 9 points), and the weight factors corresponding to the nurses in charge of each time period are 0.4, 0.3, and 0.4 respectively, then PC≈0.90 (assuming the maximum single time period quality score is 3 9 points);

[0096] Satisfaction of surgical needs:

[0097] Calculation method: Count surgical demand: Based on surgical schedules, count the number and type of surgeries in each time period; Assess nurses' surgical expertise matching: Match nurses' surgical expertise with surgical demand and calculate the matching score; Calculate surgical demand satisfaction: Weightedly calculate the sum of surgical demand satisfaction in each time period; Normalize: Normalize surgical demand satisfaction to a value between [0, 1];

[0098] formula:

[0099] ;

[0100] Where M is the number of surgeries within the scheduling cycle, surgical demand t is the weight factor of the tth surgery (reflecting the importance of the surgery in each time period), and matching score t is the matching score between the tth surgery and the responsible nurse (1 for a good match and 0 for an unmatched match). For example, suppose there are two surgeries, one requiring nurse A (specializing in internal medicine) and one requiring nurse B (specializing in surgery). Assume that the weight factors of both surgeries are 0.5. If the scheduling plan meets the needs of both surgeries, the matching scores are both 1. Therefore, SD = 1 (assuming that the matching score of the largest single time period is 1).

[0101] In summary, setting α = 0.4, β = 0.3, γ = 0.3, and using the previous example to calculate WB = 0.93, PC = 0.90, and SD = 1, then Fit = 0.4 * 0.93 + 0.3 * 0.90 + 0.3 * 1 = 0.942; this fitness value indicates that this scheduling plan performs well in comprehensively considering nurse workload balance, patient care quality, and surgical need satisfaction.

[0102] The corresponding standard of fitness is set to 0.9, so the fitness Fit in this example is 0.942, which exceeds 0.9 and therefore meets the standard;

[0103] For example:

[0104] Assume there is a small department with 3 nurses (A, B, C) who need to be scheduled for a week (7 days). Each nurse has a specific specialty and vacation schedule.

[0105] Nurse Information:

[0106] A: Specializes in pediatrics and is closed on Saturdays;

[0107] B: Specializes in first aid and is closed on Sundays;

[0108] C: No specific expertise, can work every day;

[0109] Scheduling requirements:

[0110] Monday: A large number of pediatric patients, requiring A work;

[0111] Tuesday: Emergency surgery required, B work required;

[0112] Wednesday to Friday: The number of patients is moderate, requiring rational allocation of nursing resources;

[0113] Weekends: Patient volume decreases, but there are still a small number of pediatric patients who require A or C work;

[0114] Genetic algorithm scheduling process (simplified):

[0115] Initial population: Randomly generate multiple shift plans to ensure that A does not work on Saturday and B does not work on Sunday;

[0116] Fitness function: Considers factors such as nurse expertise matching and workload balance;

[0117] Selection, crossover, and mutation: iteratively optimize the scheduling plan;

[0118] Final solution: Choose multiple excellent scheduling solutions, such as:

[0119] Plan 1: A (Monday, Wednesday, Friday, and Sunday), B (Tuesday and Thursday), C (Saturday);

[0120] Plan 2: A (Monday, Thursday, Saturday), B (Tuesday, Friday), C (Wednesday, Sunday);

[0121] The above scheme is for reference only. The actual scheduling process may involve more complex factors, such as emergencies, which need to be adjusted according to the actual situation. There will be corresponding solutions in the subsequent system operation.

[0122] The above technical solution effectively solved the complex and critical problem of hospital department scheduling;

[0123] This scheme aggregates the work needs of target departments and the basic information of nurses, and uses a genetic algorithm to generate multiple candidate scheduling schemes. This scheme fully considers multiple factors such as the number of patients, surgical schedules, disease complexity, and nurses' qualifications, expertise, current scheduling status, and vacation plans. The design of the fitness function ensures that the scheduling scheme is comprehensively optimized in terms of multiple dimensions, such as nurse workload balance, patient care quality, and surgical demand satisfaction. This not only improves scheduling efficiency and ensures that patients receive high-quality nursing services, but also rationally arranges nurses' work and rest time, avoiding workload overload or waste of resources.

[0124] In addition, the solution is flexible and scalable, and can be adjusted and optimized according to actual conditions to deal with emergencies or other complex factors. In short, this technical solution has successfully achieved intelligent and efficient scheduling of hospital departments, improved the utilization efficiency of medical resources, and enhanced the quality and level of medical services, bringing significant positive impacts to hospital management and patient care.

[0125] The plan evaluation module, after nurses receive candidate plans, builds an evaluation system engine based on their feedback and compares the human resource utilization under different candidate plans. It then combines the future work arrangements of the target department to come up with a preliminary plan.

[0126] Among them, the conditions under which the nurse receives the candidate plan are as follows:

[0127] Nurses usually need to return to the nurse station frequently during their free time or during their daily work. The nurse station is equipped with an LED display device. When instructions or information are received, they can be displayed on the screen of the device, thereby prompting the corresponding nurses. The corresponding schedule of candidate plans can also be displayed on the screen.

[0128] The nurses' feedback data represents the nurses' satisfaction with each candidate solution. If the nurse is satisfied, the corresponding candidate solution is marked as 1; if the nurse is dissatisfied, the corresponding candidate solution is marked as 0.

[0129] The utilization of human resources is quantitatively expressed through the utilization rate of human resources;

[0130] Human resource utilization rate refers to the ratio of actual human resources used to total human resources under the corresponding candidate plan;

[0131] Calculation method: Human resource utilization rate = actual human resources used / total human resources × 100%;

[0132] For example:

[0133] Under Alternative Plan A, the actual number of nurses employed during a certain time period is 20, and the total number of nurses is 30. Therefore, the human resource utilization rate is 20 / 30 × 100% ≈ 66.67%. Under Alternative Plan B, the actual number of nurses employed during the same time period is 18, and the total number of nurses is 30. Therefore, the human resource utilization rate is 18 / 30 × 100% = 60%. By comparison, it can be seen that Alternative Plan A has a higher human resource utilization rate during this time period than Alternative Plan B. That is, after comparing the human resource utilization under the different Alternative Plans, Alternative Plan A utilizes more human resources than Alternative Plan B.

[0134] The process of running the assessment system engine is as follows:

[0135] Based on the future work schedule of the target department and the total number of nurses in the target department, a mathematical calculation model is constructed to generate the required human resource standard value (i.e., the minimum human resource utilization rate). Candidate plans corresponding to human resource utilization exceeding the human resource standard value are screened out. Based on the difference between the human resource utilization rate and the human resource standard value, as well as the cumulative value of the feedback data corresponding to each candidate plan obtained after screening, the comprehensive indicators obtained after weighted calculation are sorted from large to small to generate a ranking table. The candidate plan corresponding to the first place in the ranking table is the preliminary plan.

[0136] The future work schedule of the target department represents the total number of surgeries scheduled within a preset period (usually one week). Therefore, the formula for building the mathematical calculation model is:

[0137] ;

[0138] Where R_min represents the required human resource standard value, T represents the total number of surgeries scheduled within the preset period, Nr represents the total number of nurses in the target department, and S i represents the number of nurses required for the i-th operation (this value may vary depending on factors such as the type and complexity of the operation and is a pre-set demand);

[0139] Brief Logic: The required human resource standard value, or the minimum human resource utilization rate R_min, can be defined as the ratio of the minimum number of nurses required to meet all surgical nursing needs to the total number of nurses. To simplify the calculation, the above formula is designed by assuming that all nurses have similar work abilities and ignoring issues related to nurses' rest.

[0140] For example: Assume that the target department will have 10 surgeries in the next week, with a total of 54 nurses; the number of nurses required for the surgeries is S i It varies depending on the type of surgery, such as:

[0141] Surgery 1-5: Each surgery requires 2 nurses;

[0142] Surgery 6-8: Each surgery requires 3 nurses;

[0143] Surgery 9-10: Each surgery requires 4 nurses;

[0144] but, ;

[0145] R = 27 / 54 ≈ 0.5 or 50%;

[0146] This means that in order to meet the surgical needs in the next week, at least 50% of human resources are needed; what needs to be ensured is that the total number of nurses needs to far exceed the minimum number of nurses actually required.

[0147] The cumulative value of the feedback data corresponding to each candidate solution obtained after screening represents the cumulative sum of different nurses' satisfaction with the same candidate solution. The following example helps to understand:

[0148]

[0149] Among them, 1 means satisfied (also known as satisfaction), and 0 means dissatisfied (also known as dissatisfaction);

[0150] Therefore, when the total number of nurses is three, the cumulative value of feedback data corresponding to plan A is: 1+1+1=3, the cumulative value of feedback data corresponding to plan B is: 0+1+0=1, and the cumulative value of feedback data corresponding to plan C is: 1+1+0=2;

[0151] Based on the difference between the human resource utilization rate and the human resource standard value, as well as the cumulative feedback data corresponding to each candidate solution obtained after screening, a weighted calculation is performed, and the weighted calculation formula corresponding to the obtained comprehensive index is as follows:

[0152] ;

[0153] Where Zb x The x in the formula represents the candidate number, Zb represents the comprehensive index, and R x Represents the human resource utilization rate under the corresponding candidate solution, Ac xIt represents the cumulative value of feedback data under the corresponding candidate plan, α and β both represent weight coefficients, and α+β=1; it should be noted that the greater the difference between the human resource utilization rate and the human resource standard value, the more the human resource utilization deviates from the human resource standard value, and the fewer nurses are idle or on vacation, which is not conducive to subsequent response to emergencies. The cumulative value of feedback data reflects the satisfaction of nurses with the corresponding subsequent plan, so the higher the value, the better; when the human resource utilization rate is equal to the human resource standard value, human resources are not considered when calculating the comprehensive indicator Zb, and the required comprehensive indicator Zb is obtained based on the cumulative value of feedback data under the corresponding candidate plan;

[0154] It should be noted that the weight coefficient is determined using the coefficient of variation method, which is a method of assigning weights to each indicator based on the degree of variation between the current value of each evaluation indicator and the target value. If the numerical difference of an indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich discrimination information, and thus the indicator should be given a larger weight. On the contrary, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish the evaluation objects is weak, and thus the indicator should be given a smaller weight. This method directly uses the information contained in each indicator to obtain the weight of the indicator through calculation, and therefore is objective.

[0155] Through the above technical solutions, an efficient nurse scheduling program evaluation module was built;

[0156] This module comprehensively considers nurse feedback data and the human resource utilization under different candidate plans. By building an evaluation system engine and combining it with the future work schedule of the target department, it intelligently selects the initial plan.

[0157] This technical solution not only improves the efficiency of scheduling, but also ensures the rationality of the plan in terms of human resource allocation and the superiority of nurse satisfaction. By quantifying human resource utilization and comparing it with the standard human resource value, it can intuitively evaluate the advantages and disadvantages of each candidate plan in terms of human resource utilization. At the same time, combined with nurses' satisfaction feedback on the candidate plans, a weighted calculation formula is used to comprehensively consider human resource utilization and nurse satisfaction, further improving the practicality and acceptability of the scheduling plan.

[0158] In short, this technical solution successfully solved the complex problem of evaluating hospital department scheduling plans, and brought a more intelligent, efficient and humane solution to hospital management and nurse scheduling.

[0159] Emergency response module: When an emergency occurs, the nurses on leave are prioritized based on basic information and a temporary shift notification is sent. If the nurses with the highest priority do not respond within the specified time, the shift will be postponed to the nurses with the lowest priority until a nurse responds. If no nurse responds, the recruitment mechanism is triggered.

[0160] Record the percentage of proposals that trigger the solicitation mechanism within a predetermined period. If the percentage exceeds the expected value, it indicates that the preliminary plan (i.e., the final candidate plan) needs to be adjusted, and the plan adjustment instruction is executed. If the percentage does not exceed the expected value, no response action is taken and the original preliminary plan is maintained.

[0161] The above-mentioned emergency situations include:

[0162] There is an unexpected surgery scheduled in a department or the nurse currently on duty is unable to continue to handle the work due to an accident;

[0163] The process for prioritizing nurses on leave is as follows:

[0164] Obtain the type corresponding to the current emergency status and match it with the appropriate expertise. Extract nurses who meet the appropriate expertise and are on vacation from each nurse's basic information and sort them from high to low based on their qualifications. Qualifications are positively correlated with priority.

[0165] Introducing real-time communication technology when sending temporary shift notifications;

[0166] Such as SMS, APP push, etc., to ensure that notifications can be delivered to nurses quickly and accurately; at the same time, the system supports an automatic retry mechanism. If the initial notification is not responded to, a reminder will be automatically sent at a preset time interval. If the nurse confirms the schedule (i.e. responds) within the specified time, the nurse will be assigned to deal with the emergency; if the nurse does not respond within the specified time, the next nurse in the queue will be postponed to perform the same operation until a nurse responds;

[0167] The triggered collection mechanism is:

[0168] Nurses with relevant expertise and on vacation are recruited from across the hospital. The nurse's response speed is positively correlated with the call priority, meaning the fastest-responding nurse will be called sooner. If only one vacancy is available, the fastest-responding nurse, Nurse A, will be called.

[0169] The relevant and subsequent processes for recording the percentage of triggering the collection mechanism within the predetermined period are as follows:

[0170] Within a predetermined period (i.e., a previously set period, such as monthly or quarterly), record the number of all relevant incidents, including the number of incidents triggering the call-up mechanism and the number of incidents for which temporary shift notifications were issued (i.e., the total number of incidents), and calculate the percentage of incidents triggering the call-up mechanism, i.e., the number of incidents triggering the call-up mechanism divided by the total number of incidents;

[0171] Compare the percentage with the expected value;

[0172] When the proportion exceeds the expected value, the plan adjustment instruction is executed;

[0173] If the proportion does not exceed the expected value, no response action will be taken;

[0174] The content of the executed plan adjustment instruction is:

[0175] Extract the candidate plan with the serial number of the first place + 1 from the original sorting table and use it as the adjusted preliminary plan. Then continue to record the proportion of triggering the solicitation mechanism within the predetermined period of the next cycle. If the plan adjustment instruction still needs to be executed, continue to use the candidate plan with the serial number of the first place + 2 as the adjusted preliminary plan. Repeat this cycle until there is no response.

[0176] The above technical solution enables the system to quickly prioritize nurses on leave and send temporary scheduling notifications in emergency situations, effectively responding to sudden surgical arrangements in departments or when nurses on duty are unable to continue working.

[0177] By introducing real-time communication technology and an automatic retry mechanism, we ensure that notifications are delivered quickly and accurately, improving the nurses' response speed. We also designed a recruitment mechanism so that when no nurses respond, qualified nurses can be quickly recruited from across the hospital, ensuring a timely response to emergencies.

[0178] In addition, by recording the percentage of times the solicitation mechanism is triggered within a predetermined period and comparing it with the expected value, problems in the initial plan can be discovered in a timely manner. By executing the plan adjustment instruction, the next candidate plan with the highest ranking will be used as the adjusted initial plan, thus optimizing and adjusting the scheduling plan according to the situation.

[0179] This technical solution not only improves the hospital's ability to respond to emergencies, but also ensures the flexibility and practicality of the scheduling plan, effectively solving the technical problems of untimely emergency response and inflexible plan adjustment in hospital scheduling management.

[0180] The temporary adjustment module records and prioritizes nurses on leave based on the information of responding nurses and basic information within a fixed period, and provides feedback on the ranking results (which are fed back to the initial priority ranking performed in the emergency response module, so that subsequent operations are operated or executed according to the secondary priority ranking);

[0181] The fixed period refers to a fixed period, which is set according to actual needs. In this embodiment, the fixed period is usually set to one quarter, and it is sufficient to ensure that the fixed period lasts longer than the aforementioned preset period.

[0182] The information was recorded and expressed based on the responding nurse: the responding nurse was marked;

[0183] The process for secondary prioritization of nurses on leave is as follows:

[0184] Establish a qualification quantitative value database, match the corresponding qualification level according to the qualifications of nurses in the basic information, and the qualification level is positively correlated with the qualification level. In this database, qualification level 1 corresponds to junior practicing nurses, qualification level 2 corresponds to intermediate practicing nurses, qualification level 3 corresponds to head nurses, qualification level 4 corresponds to deputy chief nurses, and qualification level 5 corresponds to chief nurses;

[0185] The specific qualification quantitative value library is as follows:

[0186]

[0187] In this database, level 1 represents the nurse with the lowest qualifications (such as junior nurse practitioners), while level 5 represents the nurse with the highest qualifications (such as head nurses). The numerical value of the qualification level is the corresponding quantitative value, which increases as the description (degree) of the qualification improves.

[0188] At this time, in response to obtaining the type corresponding to the current emergency status and matching the appropriate expertise, nurses who meet the appropriate expertise and are on vacation schedule are extracted from the basic information of each nurse. The qualification level of each nurse is determined through the qualification quantitative value library, and the proportion of each nurse's marked times within a fixed period is collected and calculated. An index calculation function model is built to generate the priority index corresponding to each qualified nurse (at this time, the nurses are not sorted from high to low by qualification), but are sorted from large to small by priority index, and the priority index is positively correlated with the priority level;

[0189] Among them, when running the exponential calculation function model, the formula is based on:

[0190] ;

[0191] ;

[0192] Where Rt, Mq, and Td represent the proportion of times marked, the number of times the corresponding nurse is marked within a fixed period, and the fixed period (usually expressed as days), respectively; Irt represents the priority index; Q represents the qualification level of the corresponding nurse; k is the adjustment coefficient used to control the influence of the proportion of marked cases on the priority index, and the value range of k is [0, 1];

[0193] Logical explanation: The higher the qualification level, the better the nurse's qualifications; Marked ratio: The more times the nurse is marked, the larger the ratio Rt, which means that the nurse reacts faster and more actively to the emergency; Adjustment coefficient k: It is used to adjust the impact of the marked ratio on the priority index. The larger the k value, the more significant the impact of the marked ratio on the priority index; Priority index calculation: Substitute the level value, the marked ratio, and the adjustment coefficient into the formula to calculate the priority index of each nurse; The higher the priority index, the higher the priority of the nurse in the current emergency state; In this embodiment, each formula needs to be dimensionless before calculation to remove the dimension to facilitate subsequent calculations;

[0194] For example:

[0195] Suppose there are two nurses:

[0196] Nurse A: Qualification level 3 (supervisory nurse), number of flags within a fixed period of 30 days is 5; Nurse B: Qualification level 2 (intermediate licensed nurse), number of flags within a fixed period of 30 days is 2; adjustment coefficient k = 0.1;

[0197] Calculate the proportion of marked times Rt:

[0198] Nurse A: Rt = 5 / 30 ≈ 0.1667;

[0199] Nurse B: Rt = 2 / 30 ≈ 0.0667;

[0200] Calculate the priority index:

[0201] Nurse A: Irt≈2.89;

[0202] Nurse B: Irt≈1.99;

[0203] According to the calculation results, Nurse A has a higher priority index and is therefore ranked first in the ranking. The value of the adjustment coefficient k can be adjusted according to actual conditions to adapt to different application scenarios and needs. It should be noted that due to the numerical range of qualification levels and marked proportions, as well as the selection of the adjustment coefficient k, the calculated priority index may need to be appropriately scaled or adjusted to ensure its effectiveness and rationality in actual application. For example, if the range of the priority index is too large or too small, it can be fine-tuned by adjusting the k value or.

[0204] After the initial prioritization based on nurses' qualifications, subsequent nurses with the same qualifications can be adjusted based on their response attitudes to emergencies (reflected by the proportion of times they are marked). This not only allows for a subdivision of the priorities of nurses with the same qualifications, but also provides opportunities for nurses with positive response attitudes. Due to their quick response, nurse scheduling can be completed more efficiently when dealing with emergencies.

[0205] Specifically, the system can perform secondary priority sorting of nurses on leave based on the responding nurses’ information and basic information;

[0206] By establishing a qualification quantification value database, nurses' qualification levels are quantified. Combined with the percentage of marked times within a fixed period, an exponential calculation function model is used to generate a priority index for each qualified nurse, enabling a refined ranking of nurse priorities. This solution not only considers the nurses' qualification levels, but also fully considers their response speed and enthusiasm for dealing with emergencies, enabling a more scientific and reasonable scheduling of nurses.

[0207] In summary, this technical solution can also provide more opportunities for nurses with a positive response attitude, motivate nurses to participate more actively in responding to emergencies, and effectively solve the problems of unreasonable priority sorting and low scheduling efficiency in hospital scheduling management.

[0208] Example 2:

[0209] See also Figure 2 Based on Example 1, this embodiment further provides a department-based automatic nurse scheduling method, which includes the following specific steps:

[0210] S1. Summarize the work requirements of the target department and the basic information of nurses, and use a genetic algorithm to generate several candidate solutions; the work requirements at least include the number of patients and surgical arrangements;

[0211] S2. After the nurses receive the candidate solutions, they will build an evaluation system engine based on their feedback and compare the human resource utilization under different candidate solutions. This engine will then combine the future work schedule of the target department to arrive at a preliminary solution.

[0212] S3. When an emergency occurs, prioritize the nurses on leave and send a temporary shift notification. If the nurses with the highest priority do not respond within the specified time, the shift will be postponed to the nurses with the lowest priority until a nurse responds. If no nurse responds, the recruitment mechanism is triggered. The proportion of nurses who trigger the recruitment mechanism within the predetermined period is simultaneously recorded. If the proportion exceeds the expected value, the plan adjustment instruction is executed.

[0213] S4. Within a fixed period, record and prioritize the nurses on leave based on the information of the responding nurses and the basic information, and provide feedback on the ranking results.

[0214] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0215] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0216] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. An automatic scheduling system for medical staff based on departments, characterized by: The system includes: The pre-scheduling module summarizes the work requirements of the target department and the basic information of nurses, and uses a genetic algorithm to generate several candidate plans; the work requirements include at least the number of patients and the surgical schedule; The plan evaluation module, after nurses receive candidate plans, builds an evaluation system engine based on their feedback and compares the human resource utilization under different candidate plans. It then combines the future work schedule of the target department to arrive at a preliminary plan. The process of running the evaluation system engine is as follows: Based on the future work schedule of the target department and the total number of nurses in the target department, a mathematical calculation model is built to generate the required human resource standard value, and candidate plans corresponding to human resource utilization exceeding the human resource standard value are screened out. Based on the difference between the human resource utilization rate and the human resource standard value, as well as the cumulative value of the feedback data corresponding to each candidate plan obtained after screening, a weighted calculation is performed to obtain a comprehensive indicator, which is then sorted from large to small to generate a ranking table. The candidate plan corresponding to the first place in the ranking table is the preliminary plan; The emergency response module prioritizes nurses on leave when an emergency occurs and sends a temporary shift notification. If a nurse with a higher priority fails to respond within the specified time, the call is postponed to a nurse with a lower priority, and so on until a nurse responds. If no nurse responds, the call-up mechanism is triggered. The proportion of the call-up mechanism triggered within the predetermined period is simultaneously recorded. If the proportion exceeds the expected value, the plan adjustment instruction is executed. The process of prioritizing nurses on leave is as follows: obtaining the type corresponding to the current emergency status and matching the corresponding expertise. Nurses who meet the corresponding expertise and are on leave are extracted from the basic information of each nurse and sorted from high to low according to qualifications, with qualifications being positively correlated with priority. The temporary adjustment module records and prioritizes the nurses on leave based on the information of the responding nurses and the basic information within a fixed period, and provides feedback on the ranking results. The process of secondary priority ranking of nurses on leave is as follows: establish a qualification quantitative value library, match the corresponding qualification level according to the qualifications of the nurses in the basic information, and the qualification level is positively correlated with the qualification level; obtain the type corresponding to the current emergency status and match it with the appropriate expertise, extract nurses who meet the appropriate expertise and are on vacation from the basic information of each nurse, determine the qualification level of each nurse through the qualification quantitative value library, collect and calculate the proportion of the number of times each nurse is marked within a fixed period, build an index calculation function model to generate the priority index corresponding to each qualified nurse, and sort them from large to small according to the priority index, and the priority index is positively correlated with the priority level; Among them, when running the exponential calculation function model, the formula is based on: ； ； Among them, Rt, Mq and Td represent the proportion of the number of times marked, the number of times the corresponding nurse is marked in a fixed period and the fixed period respectively, Irt represents the priority index, Q represents the qualification level of the corresponding nurse, and k is the adjustment coefficient, with a value range of [0, 1].

2. The automatic scheduling system for medical staff based on department according to claim 1 is characterized by: Surgical schedule: date, time, type of surgery, and the specialty of the nurse required; basic information of the nurse: the nurse's name, qualifications, specialty, current scheduling status, and vacation plans.

3. The automatic scheduling system for medical staff based on department according to claim 1 is characterized by: The process of using genetic algorithm to generate several scheduling plans is as follows: Coding: Integer coding is used. Each shift plan is represented as a chromosome. Each gene on the chromosome represents the working status of a nurse in a set time period. Initial population: Randomly generate multiple scheduling plans as the initial population, each plan meets the set constraints; Fitness function: Design a fitness function to score the quality of the scheduling plan and calculate the fitness of each scheduling plan. The quality of the scheduling plan includes at least the balance of nurse workload, the satisfaction of patient care needs, and the matching of professional skills. Selection: Use the roulette wheel selection strategy to select individuals with satisfactory fitness from the current population as parents; Crossover: Perform crossover operations on the selected parent generation to generate a new child generation schedule; Mutation: Randomly mutate the offspring scheduling plan; Iteration: Repeat the selection, crossover and mutation operations until the predetermined number of iterations is reached. After the iteration is completed, several scheduling schemes are selected from the final population as candidate schemes.

4. The automatic scheduling system for medical staff based on department according to claim 1 is characterized by: The nurses' feedback data represents the nurses' satisfaction with each candidate solution. If the nurse is satisfied, the corresponding candidate solution is marked as 1; if the nurse is dissatisfied, the corresponding candidate solution is marked as 0. The utilization of human resources is quantitatively expressed through the utilization rate of human resources; The calculation method of human resource utilization rate is: Human resource utilization rate = actual human resources used / total human resources × 100%.

5. The automatic scheduling system for medical staff based on department according to claim 1 is characterized by: The future work schedule status under the target department represents: the total number of surgeries scheduled in the future preset period; When building a mathematical calculation model, the formula is based on: ； Where R_min represents the standard value of human resources required, T represents the total number of surgeries scheduled within the preset period, Nr represents the total number of nurses in the target department, and S i represents the number of nurses required for the i-th operation.

6. The automatic scheduling system for medical staff based on department according to claim 5, characterized in that: The weighted calculation formula corresponding to the obtained comprehensive index is as follows: ； Where Zb x represents the comprehensive index under the candidate number x, R x represents the human resource utilization rate under candidate solution number x, Ac x It represents the cumulative value of feedback data under candidate solution number x, α and β are weight coefficients, and α+β=1.

7. The automatic scheduling system for medical staff based on department according to claim 1 is characterized by: The triggered recruitment mechanism is to recruit nurses from the entire hospital who have appropriate expertise and are on vacation, and the nurses' response speed is positively correlated with the call priority; After recording the percentage of times the solicitation mechanism is triggered within the scheduled period, record the number of all relevant events within the scheduled period, including the number of events that triggered the solicitation mechanism and the number of events for which temporary scheduling notifications were sent, i.e., the total number of events. Calculate the percentage of times the solicitation mechanism is triggered, i.e., the number of events that triggered the solicitation mechanism divided by the total number of events. The content of the executed scheme adjustment instruction is: extract the corresponding candidate scheme with the serial number of the first place + 1 in the original sorting table, and use it as the adjusted preliminary scheme, and then continue to record the proportion of triggering the solicitation mechanism within the predetermined period in the next cycle. If the scheme adjustment instruction is executed again, the corresponding candidate scheme with the serial number of the first place + 2 will continue to be used as the adjusted preliminary scheme, and the cycle will continue until there is no response.

8. The automatic scheduling system for medical staff based on department according to claim 7, characterized in that: Record and indicate according to the responding nurse's information: the responding nurse is marked.

9. A department-based automatic scheduling method for medical staff, applied to the system according to any one of claims 1 to 8, characterized in that: The steps include: S1. Summarize the work requirements of the target department and the basic information of nurses, and use a genetic algorithm to generate several candidate solutions; the work requirements at least include the number of patients and surgical arrangements; S2. After the nurses receive the candidate solutions, they will build an evaluation system engine based on their feedback and compare the human resource utilization under different candidate solutions. This engine will then combine the future work schedule of the target department to arrive at a preliminary solution. S3. When an emergency occurs, prioritize the nurses on leave and send a temporary shift notification. If the nurses with the highest priority do not respond within the specified time, the shift will be postponed to the nurses with the lowest priority until a nurse responds. If no nurse responds, the recruitment mechanism is triggered. The proportion of nurses who trigger the recruitment mechanism within the predetermined period is simultaneously recorded. If the proportion exceeds the expected value, the plan adjustment instruction is executed. S4. Within a fixed period, record and prioritize the nurses on leave based on the information of the responding nurses and the basic information, and provide feedback on the ranking results.

Images