Intelligent scheduling system for nurses in imaging department

Abstract

The invention discloses an imaging department nurse intelligent scheduling system, which comprises a nurse management module, a scheduling rule making module, a scheduling processing module and a real-time optimization feedback module, and is characterized in that the nurse management module is used for recording basic information, qualification, skill and experience data of imaging department personnel and evaluating work ability; the scheduling rule making module is used for defining scheduling rules and constraints of the imaging department and considering reasonable shift of nurses, personal work preference and hospital cost; the scheduling processing module is used for generating reasonable imaging department scheduling according to preset rules and constraint conditions; the real-time optimization feedback module is used for providing real-time scheduling state and information and evaluating the efficiency of the scheduling system; according to the invention, a better scheduling scheme can be obtained, the hospital cost is reasonably reduced, the work satisfaction of nurses is improved, the image experience of patients is improved, and the security of the scheduling system is improved through the block chain technology.

Description

Technical Field

[0001] The present invention relates to the field of intelligent scheduling systems, and more particularly, to an intelligent scheduling system for radiology nurses. Background Art

[0002] Many hospitals use manual methods for scheduling, especially for radiology nurses. Relying solely on the past experience of managers, it is not only inefficient but also unable to achieve fairness and justice. Current intelligent scheduling systems do not generate reasonable scheduling plans intelligently based on the skills, experience, personal preferences of medical staff, and the costs of hospitals. At the same time, the security performance of the scheduling system is relatively weak. The above problems lead to low job satisfaction among nurses and poor imaging experience for patients. The security performance of the scheduling system needs to be improved. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent scheduling system for radiology nurses to solve the above problems existing in the prior art.

[0004] Specifically, this application is as follows:

[0005] An intelligent scheduling system for radiology nurses, comprising:

[0006] A nurse management module: used to record the basic information, qualifications, skills, and experience data of radiology personnel, and evaluate work capabilities;

[0007] A scheduling rule formulation module: used to define the rules and constraints for radiology scheduling, considering the reasonable shifts of nurses, personal work preferences, and hospital costs;

[0008] A scheduling processing module: used to generate a reasonable radiology schedule according to the preset rules and constraint conditions;

[0009] A real-time optimization feedback module: used to provide real-time scheduling status and information, evaluate the efficiency of the scheduling system, set an efficiency threshold, and if the threshold is not met, optimize the scheduling rules and constraints.

[0010] Further, the nurse management module includes:

[0011] A nurse information unit: used to manage the basic information of radiology nurses;

[0012] A permission definition unit: used to create and define different user roles;

[0013] A permission traceability management unit: used to allocate and manage the access permissions of the system for nurses with different roles, and use blockchain technology to encrypt and trace the operations of each user.

[0014] Further, the scheduling rule formulation module includes:

[0015] Personnel preference analysis unit: used to collect and analyze the personal work preferences of nurses;

[0016] Hospital cost analysis unit: used to analyze the costs generated by scheduling and formulate cost control rules.

[0017] Furthermore, the scheduling processing module includes:

[0018] Automatic scheduling generation unit: automatically generates a reasonable scheduling plan for radiology nurses according to preset rules and constraints;

[0019] Temporary adjustment unit: used to identify the temporary needs of the changed personnel through AI technology, realize the temporary dynamic adjustment of the scheduling plan, and output a new scheduling plan.

[0020] Furthermore, the specific implementation process of the temporary adjustment unit is as follows:

[0021] S1. Collect the facial images of nurses who make scheduling-related sounds and identify and process the nurse information;

[0022] S2. Identify the voices of nurses who make scheduling-related sounds and double-verify with the facial information of the nurses to confirm the identity information of the nurses;

[0023] S3. If the nurse is a radiology department staff, translate and identify the nurse's voice, and extract the semantics of the translated nurse's voice;

[0024] S4. Make temporary scheduling adjustments according to the semantic recognition results of the nurse's voice, output a temporary scheduling plan, and at the same time encrypt and trace the change through blockchain technology.

[0025] Furthermore, the specific implementation process of the automatic scheduling generation unit is as follows:

[0026] S5. Construct the preference distribution and coverage requirement distribution of nurse scheduling;

[0027] S6. Construct a nurse preference matrix according to the preference distribution and coverage requirement distribution, and generate a nurse preference weight coefficient, which is obtained by the weighted average sum of the nurse preference matrix;

[0028] S7. Construct a number of objective functions and a number of constraint conditions, which are based on the nurse preference matrix and the nurse preference weight coefficient;

[0029] S8. Solve the number of objective functions and the number of constraint conditions through the variable neighborhood search algorithm to obtain an initial scheduling solution;

[0030] S9. Iteratively optimize the initial scheduling solution through the tabu search algorithm to obtain the iterative solution;

[0031] S10. Output the nurse scheduling plan according to the iterative solution.

[0032] Furthermore, the specific process of constructing the preference distribution and coverage requirement distribution for nurse scheduling in S5 is as follows:

[0033] S51. Nurse preference distribution:

[0034]

[0035] Among them, N represents the total number of nurses; S represents the total number of shift types; D represents the scheduling cycle / day; y it = 1 indicates that the preference degree of nurse i for a certain shift on a certain day is l; y it = 0 indicates that the preference degree of nurse i for a certain shift on a certain day is not l;

[0036] S52. Shift preference distribution:

[0037]

[0038] Among them, represents the number of different preference values of nurse i for all shifts on a certain day;

[0039] S53. Working day preference distribution:

[0040]

[0041] Among them, represents the number of different preference values of nurse i for shift z during the entire scheduling cycle;

[0042] S54. Obtain the following system of equations from formula (1):

[0043]

[0044] Among them, x represents the number of nurses with a preference for shift S1, y represents the number of nurses with a preference for shift S2, z represents the number of nurses with a preference for shift S3. Considering the actual situation of scheduling, nurses are required to work on each shift, and in practice, it is necessary to exclude the situation where the same nurse works more than two shifts in a day. Therefore, assuming that for the preferences of the three shifts, the sum of the numbers is the sum of the number of nurses, m represents N / S, M represents the product of NPD and , N, M, and m are constants, and the solution of formula (4) is obtained through programming language. The solution is the nurse preference matrix;

[0045] S55. Overall coverage constraint:

[0046]

[0047] Among them, r jz represents the demand for nurses in shift z on the j-th day, represents the average daily demand for nurses throughout the scheduling period, represents the average demand for nurses in each shift on the j-th day, P ijz represents the preference of nurse i for shift z on the j-th day;

[0048] S56. Daily demand distribution:

[0049]

[0050] S57. Shift demand distribution:

[0051]

[0052] S58. Obtain the nurse coverage demand equation system from formulas (4), (5), and (6):

[0053]

[0054] Among them, x 1 , x 2 ,... x D respectively represent the number of nurses going to work each day, y j1 , y j2 , y j3 respectively represent the number of nurses going to work in each shift on the j-th day. Both G and H are constants, and the solution of the equation system (8) is obtained through programming language, and the said solution is the coverage demand distribution.

[0055] Furthermore, the specific process of constructing several objective functions and several constraint conditions in S7 is as follows:

[0056] The construction of the objective function is as follows:

[0057]

[0058] Among them, p ijz and d ijz respectively represent the salary grade coefficient and preference grade coefficient of the nurse. The said preference grade coefficient is obtained according to the nurse preference matrix. α represents the weight of the hospital cost; β represents the weight of the nurse preference;

[0059] The construction of the constraint conditions is to construct 8 constraint conditions in total, including the first constraint condition, the second constraint condition, the third constraint condition, the fourth constraint condition, the fifth constraint condition, the sixth constraint condition, the seventh constraint condition, and the eighth constraint condition:

[0060] The said first constraint condition:

[0061]

[0062] Among them

[0063]

[0064] The first constraint condition indicates that the demand quantity of the shift of the z-th type on the j-th day for nurses at level g cannot be lower than D gjz ;

[0065] The second constraint condition:

[0066]

[0067] The second constraint condition indicates the lower limit of the number of working days of each nurse within the scheduling period;

[0068] The third constraint condition:

[0069]

[0070] The third constraint condition indicates the upper limit of the number of working days of each nurse within the scheduling period;

[0071] The fourth constraint condition:

[0072]

[0073] The fourth constraint condition indicates that each nurse can only work one type of shift per day;

[0074] The fifth constraint condition:

[0075]

[0076] The fifth constraint condition indicates that the day immediately following a nurse's night shift requires a day of rest;

[0077] The sixth constraint condition:

[0078] x ijz +x i(j-1)z +x i(j-2)z ≤2 i = 1, 2, ……, n, j = 3, 4, ……, m z = 1, 2, 3,

[0079] The sixth constraint condition indicates that a nurse cannot work for three consecutive days, that is, among three consecutive days, at least one day is for rest;

[0081] The seventh constraint condition:

[0082] x ijz ≤w ijzi = 1, 2, ……, n, j = 1, 2, ……, m, z = 1, 2, 3

[0083] Among them,

[0084] The seventh constraint condition indicates that the i-th nurse takes leave on the j-th day for the z-th type of shift;

[0085] The eighth constraint condition:

[0086] x ijz ≤a i3 i = 1, 2, ……, n, j = 1, 2, ……, m z = 1, 2, 3

[0087] The eighth constraint condition indicates that the i-th nurse does not work the night shift;

[0088] The expressions in the first constraint condition to the eighth constraint condition and the objective function are explained as follows:

[0089] z represents encoding using the APN scheduling model. There are three types of work shifts: morning shift from 07:30 to 15:30, represented by 1; middle shift from 14:00 to 22:30, represented by 2; night shift from 22:00 to 8:00, represented by 3; n represents the number of nurses; g represents the nurse level; m represents the scheduling cycle; c ijz represents the salary obtained by the i-th nurse for the z-th type of shift on the j-th day; x ijz represents whether the i-th nurse works the z-th type of shift on the j-th day. If so, the value is 1; otherwise, it is 0; D gjz represents the demand number of nurses at level g for the z-th shift on the j-th day; q ig represents that if the level of the i-th nurse is not lower than g, the value is 1; otherwise, it is 0; low represents the lower limit of the number of working shifts of a nurse in a scheduling cycle; up represents the upper limit of the number of working shifts of a nurse in a scheduling cycle; w ijz represents whether the i-th nurse takes leave for the z-th type of shift on the j-th day. If so, the value is 1; otherwise, it is 0; a i3 represents whether the i-th nurse works the night shift.

[0090] Furthermore, the specific implementation process of evaluating the efficiency of the scheduling system in the real-time optimization feedback module is as follows:

[0091] L1. Through a preset feedback mechanism, collect the feedback information on the scheduling of nurses within a preset time period, including work quality and satisfaction ratings;

[0092] L2. Use statistical analysis methods to process the collected data and extract key indicators, which include the average satisfaction score X and the average work quality score Y. The specific calculation formula is as follows:

[0093] and where x i and y i are the satisfaction and work quality of a single nurse respectively, and n is the total number of nurses;

[0094] L3. Calculate the efficiency Q of the scheduling system based on the average satisfaction score X and the average work quality score Y. The specific calculation formula is as follows:

[0095] Q = X * a + Y * b, where a and b represent the average satisfaction score coefficient and the average work quality score coefficient respectively.

[0096] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0097] The embodiments of the present invention provide a nurse management module for recording the basic information, qualifications, skills and experience data of radiology department personnel and evaluating work capabilities; a scheduling rule formulation module for defining the rules and constraints of radiology department scheduling, considering the reasonable shift rotation of nurses, personal work preferences and hospital costs; a scheduling processing module for generating a reasonable radiology department scheduling according to the preset rules and constraint conditions; a real-time optimization feedback module for providing real-time scheduling status and information and evaluating the efficiency of the scheduling system. The present invention can obtain a better scheduling plan, reasonably reduce hospital costs, improve nurses' job satisfaction, improve the imaging experience of patients, and enhance the security of the scheduling system through blockchain technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 is a system diagram of an intelligent scheduling system for radiology nurses provided by an embodiment of the present invention;

[0099] Figure 2 is a diagram of the nurse level and salary table of an intelligent scheduling system for radiology nurses provided by an embodiment of the present invention;

[0100] Figure 3 is a diagram of the required number of people for each shift of an intelligent scheduling system for radiology nurses provided by an embodiment of the present invention;

[0101] Figure 4 is a diagram of the nurse scheduling plan of an intelligent scheduling system for radiology nurses provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0102] The present invention will be described in detail below with reference to the accompanying drawings.

[0103] Embodiment 1

[0104] An embodiment of the present invention provides an intelligent scheduling system for radiology nurses, as Figure 1 , including:

[0105] Nurse management module: used to record the basic information, qualifications, skills and experience data of radiology department personnel, and evaluate work ability;

[0106] Scheduling rule formulation module: used to define the rules and constraints for radiology department scheduling, considering the reasonable shift rotation of nurses, personal work preferences and hospital costs;

[0107] Scheduling processing module: used to generate a reasonable radiology department scheduling according to the preset rules and constraints;

[0108] Real-time optimization feedback module: used to provide real-time scheduling status and information, evaluate the efficiency of the scheduling system, set an efficiency threshold, and if the threshold is not met, optimize the scheduling rules and constraints.

[0109] Specifically, the nurse management module is used to record the basic information, qualifications, skills and experience data of radiology department personnel, and evaluate work ability; the scheduling rule formulation module is used to define the rules and constraints for radiology department scheduling, considering the reasonable shift rotation of nurses, personal work preferences and hospital costs; the scheduling processing module is used to generate a reasonable radiology department scheduling according to the preset rules and constraints; the real-time optimization feedback module is used to provide real-time scheduling status and information, and evaluate the efficiency of the scheduling system; the present invention can obtain a better scheduling plan, reasonably reduce hospital costs, improve nurses' job satisfaction, and improve the imaging experience of patients

[0110] In the above embodiment, specifically, the nurse management module includes:

[0111] Nurse information unit: used to manage the basic information of radiology nurses, including personal qualifications, skills, work experience, hobbies and contract data;

[0112] Permission definition unit: used to create and define different user roles, such as head nurse, main shift nurse, responsible nurse, night shift nurse, general nurse, to reflect the responsibilities and authorities of each user in the system;

[0113] Permission traceability management unit: used to assign and manage the access permissions of the system for nurses with different roles, and use blockchain technology to encrypt and trace the operations of each user.

[0114] Specifically, the specific implementation process of encrypting and tracing the operations of each user using blockchain technology is:

[0115] N1. Generate a log according to the operation request of the user, and record information such as operation time, operator, operation content, etc.;

[0116] N2. Deploy a log collection module in the scheduling system to collect operation log data in real time and perform preprocessing. The preprocessing includes at least filtering and keyword extraction. The keyword extraction uses the unsupervised keyword extraction RAKE algorithm to identify keywords through natural language stop words and part-of-speech tagging;

[0117] N3. Package the operation log data into block data and generate a hash value;

[0118] N4. Write the log data into the blockchain ledger, ensuring that each block contains the hash value of the previous block, thus forming an immutable chain;

[0119] N5. Store the log data in the blockchain through a smart contract, and at the same time record the detailed information of the operation;

[0120] N6. Query through the query interface of the blockchain (such as the query method in the smart contract) according to the hash value or keyword of the operation log.

[0121] In the above embodiment, specifically, the scheduling rule formulation module includes:

[0122] Personnel preference analysis unit: used to collect and analyze the personal work preferences of nurses, including specific shift times, rest days, and work locations;

[0123] Hospital cost analysis unit: used to analyze the costs generated by scheduling and formulate cost control rules.

[0124] In the above embodiment, specifically, the scheduling processing module includes:

[0125] Automatic scheduling generation unit: automatically generate a reasonable scheduling plan for radiology nurses according to pre-set rules and constraints;

[0126] Temporary adjustment unit: used to identify the temporary needs of the changed personnel through AI technology, realize the temporary dynamic adjustment of the scheduling plan, and output a new scheduling plan.

[0127] In the above embodiment, specifically, the specific implementation process of the temporary adjustment unit is as follows:

[0128] S1. Collect the facial images of nurses who make scheduling-related sounds and identify and process the nurse information;

[0129] S2. Identify the voices of nurses who make scheduling-related sounds, and perform double verification with the facial information of the nurses to confirm the identity information of the nurses;

[0130] S3. If the nurse is from the imaging department, translate the nurse's voice and perform recognition to obtain it, and extract and process the semantics of the translated nurse's voice;

[0131] S4. Adjust the temporary schedule according to the semantic recognition result of the nurse's voice, output the temporary schedule plan, and at the same time encrypt and trace the change through blockchain technology;

[0132] It should be noted that the security of the scheduling system is improved through blockchain technology to ensure that the operations of users in the scheduling system are traceable.

[0133] Furthermore, the specific implementation process of the automatic scheduling generation unit is as follows:

[0134] S5. Construct the preference distribution and coverage requirement distribution of nurse scheduling, as Figure 3 shown;

[0135] S6. Construct a nurse preference matrix according to the preference distribution and coverage requirement distribution, and generate a nurse preference weight coefficient, which is obtained by the weighted average sum of the nurse preference matrix;

[0136] S7. Construct a number of objective functions and a number of constraint conditions, which are based on the nurse preference matrix and the nurse preference weight coefficient;

[0137] S8. Solve the number of objective functions and the number of constraint conditions through the variable neighborhood search algorithm to obtain an initial scheduling solution;

[0138] S9. Iteratively optimize the initial scheduling solution through the tabu search algorithm to obtain an iterative solution;

[0139] S10. Output a nurse scheduling plan according to the iterative solution, as Figure 4 shown.

[0140] In the above embodiment, specifically, the specific process of constructing the preference distribution and coverage requirement distribution of nurse scheduling in S5 is as follows:

[0141] S51. Nurse preference distribution:

[0142]

[0143] Among them, N represents the total number of nurses; S represents the total number of shift types; D represents the scheduling cycle / day; y it = 1 indicates that the preference degree of nurse i for a certain shift on a certain day is l; y it = 0 indicates that the preference degree of nurse i for a certain shift on a certain day is not l;

[0144] S52. Shift preference distribution:

[0145]

[0146] Among them, represents the number of different preference values of nurse i for all shifts on a certain day;

[0147] S53. Working day preference distribution:

[0148]

[0149] Among them, represents the number of different preference values of nurse i for shift z during the entire scheduling period;

[0150] S54. The following system of equations is obtained from formula (1):

[0151]

[0152] Among them, x represents the number of nurses who prefer shift S1, y represents the number of nurses who prefer shift S2, z represents the number of nurses who prefer shift S3. Considering the actual situation of scheduling, nurses are required to work on each shift, and in actual situations, it is necessary to exclude the situation where the same nurse works more than two shifts in a day. Therefore, assuming that for the preference of three shifts, the sum of the numbers is the sum of the number of nurses, m represents N / S, M represents the product of NPD and , N, M, and m are constants, and the solution of formula (4) is obtained through programming language. The solution is the nurse preference matrix;

[0153] S55. Overall coverage constraint:

[0154]

[0155] Among them, r jz represents the demand for nurses for shift z on the jth day, represents the average demand for nurses per day during the entire scheduling period, represents the average demand for nurses per shift on the jth day, P ijz represents the preference of nurse i for shift z on the jth day;

[0156] S56. Daily demand distribution:

[0157]

[0158] S57. Shift demand distribution:

[0159]

[0160] S58. The nurse coverage demand system of equations is obtained from formulas (4), (5), and (6):

[0161]

[0162] Among them, x 1 、x 2 、...x D respectively represent the number of nurses going to work every day, and y j1 、y j2 、y j3 respectively represent the number of nurses going to work in each shift on the j-th day. Both G and H are constants. The solution of the system of equations (8) is obtained through programming language, and the solution is the coverage demand distribution.

[0163] In the above embodiment, specifically, the specific process of constructing several objective functions and several constraint conditions in S7 is as follows:

[0164] The construction of the objective function is:

[0165]

[0166] Among them, p ijz and d ijz respectively represent the salary grade coefficient and preference grade coefficient of the nurse. The preference grade coefficient is obtained according to the nurse preference matrix. α represents the weight of the hospital cost; β represents the weight of the nurse preference, as Figure 2 shown;

[0167] The construction of the constraint conditions is to construct a total of 8 constraint conditions, including the first constraint condition, the second constraint condition, the third constraint condition, the fourth constraint condition, the fifth constraint condition, the sixth constraint condition, the seventh constraint condition and the eighth constraint condition:

[0168] The first constraint condition:

[0169]

[0170] Among them

[0171]

[0172] The first constraint condition means that the demand number of the z-th type of shift on the j-th day for nurses at level g cannot be lower than D gjz ;

[0173] The second constraint condition:

[0174]

[0175] The second constraint condition means the lower limit of the number of working days of each nurse within the scheduling period;

[0176] The third constraint condition:

[0177]

[0178] The third constraint represents the upper limit of the number of working days for each nurse within the scheduling period;

[0179] The fourth constraint:

[0180]

[0181] The fourth constraint represents that each nurse can only work one type of shift per day;

[0182] The fifth constraint:

[0183]

[0184] The fifth constraint represents that a nurse needs to rest for one day on the day immediately following a night shift;

[0185] The sixth constraint:

[0186] x ijz +x i(j-1)z +x i(j-2)z ≤2 i = 1, 2, ……, n, j = 3, 4, ……, m, z = 1, 2, 3. The sixth constraint represents that a nurse cannot work for three consecutive days, that is, among three consecutive days, there is at least one day of rest;

[0187] The seventh constraint:

[0188] x ijz ≤w ijz i = 1, 2, ……, n, j = 1, 2, ……, m, z = 1, 2, 3, where The seventh constraint represents that the i-th nurse takes a leave from working the z-th type of shift on the j-th day;

[0189] The eighth constraint:

[0190] x ijz ≤a i3 i = 1, 2, ……, n, j = 1, 2, ……, m, z = 1, 2, 3. The eighth constraint represents that the i-th nurse does not work the night shift;

[0191] The expressions in the first constraint to the eighth constraint and the objective function are explained as follows:

[0192] Let \(z\) represent the encoding using the APN scheduling model. There are three types of work shifts: morning shift A from 07:30 to 15:30, represented by 1; mid-shift P from 14:00 to 22:30, represented by 2; night shift N from 22:00 to 8:00, represented by 3. Let \(n\) represent the number of nurses, \(g\) represent the nurse level, \(m\) represent the scheduling period, and \(c\). ijz represents the salary obtained by the \(i\)-th nurse for working the \(z\)-th type of shift on the \(j\)-th day; \(x\). ijz represents whether the \(i\)-th nurse works the \(z\)-th type of shift on the \(j\)-th day. If so, the value is 1; otherwise, it is 0; \(D\). gjz represents the demand number of nurses with level \(g\) for the \(z\)-th shift on the \(j\)-th day; \(q\). ig represents that if the level of the \(i\)-th nurse is not lower than \(g\), the value is 1; otherwise, it is 0. Let low represent the lower limit of the number of working shifts of a nurse within a scheduling period, and up represent the upper limit of the number of working shifts of a nurse within a scheduling period; \(w\). ijz represents whether the \(i\)-th nurse takes leave on the \(z\)-th type of shift on the \(j\)-th day. If so, the value is 1; otherwise, it is 0; \(a\). i3 represents whether the \(i\)-th nurse works the night shift.

[0193] In the above embodiment, specifically, the specific implementation process of constructing the nurse preference matrix according to the preference distribution and the coverage demand distribution in step S6 is as follows:

[0194] S61. Construct a preference distribution matrix according to the preference distribution;

[0195] S62. Construct a coverage demand distribution matrix according to the coverage demand distribution;

[0196] S63. Construct a nurse preference matrix according to the preference distribution matrix and the coverage demand distribution matrix;

[0197] The specific implementation of step S61 is as follows:

[0198] S611. Determine the preference attributes and levels. Attribute selection: According to actual needs, select the key attributes that affect nurse preferences, such as shift types (morning shift, evening shift, night shift), working days (Monday to Sunday), consecutive working days, etc.; Level setting: Set specific levels for each attribute. For example, shift types can be divided into morning shift, evening shift, and night shift; working days can be divided into Monday to Sunday.

[0199] S612. Generate a choice set. Use fractional factorial design FFD to generate a choice set. The number of choice sets should meet orthogonality (attributes are independent of each other), balance (the number of occurrences of each level of each attribute is the same), and minimum overlap.

[0200] S613. Data collection: Through questionnaires or interviews, nurses are asked to make preference selections from the generated selection sets. For example, two different shift arrangements are provided, and nurses are asked to choose which one they prefer more;

[0201] S614. Construction of the preference distribution matrix: Based on the selection results of nurses, a preference matrix P is constructed, where P ij represents the preference value of nurse i for shift j, and the preference value is calculated through a utility function:

[0202] U ij = β 0 + β 1 Z 1j + β 2 Z 2j +... + β n Z nj + ∈ ij ,

[0203] where Z kj represents the k-th attribute level of shift j, β k represents the weight of this attribute, and ∈ ij represents the random error; the final preference distribution matrix P is an m×n matrix, where m is the number of nurses and n is the number of shifts;

[0204] The specific implementation of step S62 is as follows:

[0205] S621. Determine the coverage requirements: According to the actual needs of the imaging department, determine the number of nurses required for each shift, which can be completed through historical data, patient flow analysis, or expert consultation. Assume that D j represents the required number of people for the j-th shift, and the coverage requirement distribution can be expressed as a vector D = [D 1 , D 2 ,..., D n , where n represents the total number of shifts;

[0206] S622. Demand prediction model: A demand prediction model (such as a random forest model) can be used to predict the nurse demand for each shift;

[0207] S623. Construction of the coverage requirement distribution matrix: The coverage requirement matrix D is a 1×n vector, representing the required number of people for each shift.

[0208] Specifically, a nurse preference matrix is constructed based on the preference distribution matrix and the coverage requirement distribution matrix. For example, the nurse preference matrix can be used as a constraint condition, and the optimal scheduling plan can be solved through integer programming or other optimization algorithms (such as genetic algorithms, tabu search algorithms).

[0209] In the above embodiments, specifically, in step S9, the initial scheduling solution is iteratively optimized by the tabu search algorithm, and the specific implementation process of obtaining the iterative solution is as follows:

[0210] S91. Initialize the algorithm and set parameters, including the tabu list length, candidate set length, and maximum number of iterations.

[0211] S92. Encode the scheduling data obtained by solving the variable neighborhood search algorithm to obtain the encoding result of a binary string, and generate an initial feasible solution based on the encoding result as the starting point of the iterative search.

[0212] S93. Apply the intensification strategy and diversification strategy respectively to generate a candidate solution set.

[0213] S94. According to the evaluation function, determine whether there is a solution in the current candidate solution set that satisfies the aspiration criterion. If so, select this solution as the initial solution for the next iteration; otherwise, select the current local optimal solution that is not tabu in the candidate solution set as the initial solution for the next iteration.

[0214] S95. Update the tabu list.

[0215] S96. Determine whether the algorithm meets the termination condition. If it does, output the optimal solution that appears during the iteration and terminate the algorithm; if not, use the currently selected solution as the starting point for the next iteration, and go to step S93.

[0216] In the above embodiments, specifically, the specific implementation process of evaluating the efficiency of the scheduling system in the real-time optimization feedback module is as follows:

[0217] L1. Through a preset feedback mechanism, collect the feedback information of nurses' scheduling within a preset time period, including work quality and satisfaction ratings.

[0218] L2. Apply statistical analysis methods to process the collected data and extract key indicators. The indicators include the average satisfaction rating X and the average work quality rating Y. The specific calculation formulas are:

[0219] and where x i and y i are the satisfaction and work quality of a single nurse respectively, and n is the total number of nurses.

[0220] L3. Calculate the efficiency Q of the scheduling system according to the average satisfaction rating X and the average work quality rating Y. The specific calculation formula is:

[0221] Q = X * a + Y * b, where a and b represent the average satisfaction rating coefficient and the average work quality rating coefficient respectively.

[0222] Example 2

[0223] Suppose there are 18 nurses in the imaging department of a hospital, and the scheduling cycle is 14 days. The grades and daily wage coefficients of the nurses are as follows Figure 2 ; The number of personnel required for each shift every day is as follows Figure 3 , where N1 is the head nurse and does not work the night shift; N6, N7, and N12 are over 50 years old and also do not work the night shift. N8 takes a leave on the 7th scheduling day. Each nurse works at least 5 times and at most 10 times within the scheduling cycle. The values of α and β are 0.7 and 0.3 respectively. Perform scheduling, and the scheduling result is as follows Figure 4 shown

[0224] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings based herein. Based on the above description, the structure required to construct such systems is obvious. In addition, the present invention is not directed to any particular programming language. It should be understood that the content of the present invention described herein can be implemented using various programming languages, and the description of the specific language above is for disclosing the best mode of the present invention

[0225] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification

[0226] Similarly, it should be understood that, in order to streamline this disclosure and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting the intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the present invention

[0227] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be adopted to combine all the features disclosed in this specification (including the accompanying claims, abstract and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise explicitly stated, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) can be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0228] In addition, those skilled in the art can understand that although some of the embodiments herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present invention and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0229] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the device according to the embodiments of the present invention. The present invention can also be implemented as a device or device program (such as a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Claims

1. An intelligent scheduling system for nurses in the radiology department, characterized in that: include: Nurse management module: used to record the basic information, qualifications, skills and experience data of radiology department personnel and evaluate their work ability; Scheduling rule formulation module: used to define the rules and constraints for scheduling in the radiology department, taking into account the reasonable shifts of nurses, personal work preferences, and hospital costs; Scheduling processing module: used to generate reasonable radiology department schedules based on pre-set rules and constraints; Real-time optimization feedback module: used to provide real-time scheduling status and information, evaluate the efficiency of the scheduling system, set efficiency thresholds, and optimize scheduling rules and constraints if the thresholds are not met.

2. The intelligent scheduling system for radiology nurses according to claim 1 is characterized in that: The nurse management module includes: Nurse information unit: used to manage the basic information of nurses in the imaging department; Permission definition unit: used to create and define different user roles; Permission traceability management unit: used to allocate and manage access rights to the system for nurses with different roles, and use blockchain technology to encrypt and trace the operations of each user.

3. The intelligent scheduling system for radiology nurses according to claim 1 is characterized in that: The scheduling rule formulation module includes: Personnel Preference Analysis Unit: used to collect and analyze nurses’ personal work preferences; Hospital cost analysis unit: used to analyze the costs incurred by scheduling and formulate cost control rules.

4. The intelligent scheduling system for radiology nurses according to claim 1 is characterized in that: The shift scheduling processing module includes: Automatic scheduling generation unit: automatically generates a reasonable scheduling plan for imaging department nurses based on pre-set rules and constraints; Temporary adjustment unit: used to identify temporary needs of changed personnel through AI technology, realize temporary dynamic adjustment of shift scheduling, and output new shift scheduling plan.

5. The intelligent scheduling system for radiology nurses according to claim 4 is characterized in that: The specific implementation process of the temporary adjustment unit is as follows: S1. Collect facial images of nurses who make voices related to shift scheduling and identify and process nurse information; S2. Recognize the voice of the nurse who makes the voice related to the shift scheduling, and double-verify the nurse's facial information to confirm the nurse's identity information; S3. If the nurse is a radiology staff, the nurse's voice is translated and recognized, and the semantics of the translated nurse's voice is extracted and processed; S4. Make temporary shift adjustments based on the semantic recognition results of the nurse’s voice, output a temporary shift plan, and encrypt and trace the change through blockchain technology.

6. The intelligent scheduling system for nurses in the radiology department according to claim 4 is characterized in that: The specific implementation process of the automatic scheduling generation unit is as follows: S5, construct the preference distribution and coverage demand distribution of nurse scheduling; S6. Constructing a nurse preference matrix according to the preference distribution and the coverage demand distribution, and generating a nurse preference weight coefficient, wherein the nurse preference weight coefficient is obtained by taking a weighted average of the nurse preference matrix; S7, constructing a plurality of objective functions and a plurality of constraints, wherein the objective functions and the constraints are based on the nurse preference matrix and the nurse preference weight coefficient; S8. Solving the objective functions and constraints by using a variable neighborhood search algorithm to obtain an initial solution for scheduling; S9, iteratively optimize the initial solution of the scheduling by using the tabu search algorithm to obtain an iterative solution; S10. Output the nurse scheduling schedule based on the iterative solution.

7. The intelligent scheduling system for radiology nurses according to claim 6 is characterized in that: The specific process of constructing the preference distribution and coverage demand distribution of nurse scheduling in S5 is as follows: S51. Distribution of nurses’ preferences: Where N is the total number of nurses; S is the total number of shift types; D is the shift cycle / day; y it =1 means that nurse i has a preference for a certain shift on a certain day; y it =0 means that nurse i’s preference for a certain shift on a certain day is not l; S52, shift preference distribution: in, represents the number of different preference values ​​of nurse i for all shifts on a certain day; S53. Weekday preference distribution: in, represents the number of different preference values ​​of nurse i for shift z in the entire scheduling cycle; S54. The following equations are obtained from formula (1): Among them, x represents the number of nurses who prefer shift S1, y represents the number of nurses who prefer shift S2, and z represents the number of nurses who prefer shift S3. Considering the actual situation of shift scheduling, each shift needs to have a nurse on duty, and in actual situations, the same nurse who works more than two shifts a day needs to be excluded. Therefore, it is assumed that the sum of the number of nurses who prefer the three shifts is the sum of the number of nurses, m represents N / S, and M represents the difference between NPD and The product of, N, M, m are constants, and the solution of formula (4) is obtained by programming language, and the solution is the nurse preference matrix; S55. Overall coverage constraints: Among them, r jz represents the demand for nurses on shift z on day j, represents the average demand for nurses every day in the entire scheduling cycle, represents the average demand for nurses per shift on day j, P ijz represents nurse i’s preference for shift z on day j; S56, daily demand distribution: S57, shift demand distribution: S58. The nurse coverage requirement equations are obtained from formulas (4), (5) and (6): Among them, x1, x2, ...x D Respectively represent the number of nurses working every day, y j1 ,y j2 ,y j3 They respectively represent the number of nurses working in each shift on the jth day. G and H are both constants. The solution of equation group (8) is obtained by programming language, and the solution is the coverage demand distribution.

8. The intelligent scheduling system for radiology nurses according to claim 6 is characterized in that: The specific process of constructing several objective functions and several constraints in S7 is as follows: The objective function is constructed as: Among them, p ijz and d ijz They respectively represent the salary grade coefficient and preference grade coefficient of the nurses, the preference grade coefficient is obtained according to the nurse preference matrix, α represents the weight of hospital cost; β represents the weight of nurse preference; The construction of the constraints consists of a total of 8 constraints, including the first constraint, the second constraint, the third constraint, the fourth constraint, the fifth constraint, the sixth constraint, the seventh constraint and the eighth constraint: The first constraint condition is: in The first constraint condition indicates that the number of nurses with level g required for the z-th type of shift on the j-th day cannot be less than D gjz ; The second constraint is: The second constraint condition represents the lower limit of the number of working days for each nurse in the scheduling cycle; The third constraint condition is: The third constraint condition represents the upper limit of the number of working days for each nurse in the scheduling cycle; The fourth constraint condition is: The fourth constraint condition indicates that each nurse can only work one type of shift per day; The fifth constraint condition: The fifth constraint condition indicates that the nurse needs to take a day off the next day after finishing the night shift; The sixth constraint condition: x ijz +x i(j-1)z +x i(j-2)z ≤2i=1,2,……,n,j=3,4,……,m z=1,2,3, The sixth constraint condition indicates that the nurse cannot work for three consecutive days, that is, there must be at least one day off in three consecutive days; The seventh constraint condition: x ijz ≤w ijz i=1,2,……,n,j=1,2,……,m,z=1,2,3, in, The seventh constraint condition indicates that the i-th nurse takes leave for the z-th type of shift on the j-th day; The eighth constraint condition: x ijz ≤a i3 i=1,2,……,n,j=1,2,……,m z=1,2,3, The eighth constraint condition indicates that the i-th nurse does not work the night shift; The formulas in the first constraint condition to the eighth constraint condition and the objective function are expressed as follows: z indicates that the APN scheduling model is used for coding. There are three types of shifts: morning, midday, and evening. The morning shift is 07:30-15:30, represented by 1, the midday shift is 14:00-22:30, represented by 2, and the evening shift is 22:00-8:00, represented by 3; n indicates the number of nurses; g indicates the nurse level; m indicates the scheduling cycle; c indicates the number of nurses; ijz represents the salary of the i-th nurse working the z-th type of shift on the j-th day; x ijz Indicates whether the i-th nurse works the z-th type of shift on the j-th day. If yes, the value is 1, otherwise it is 0; D gjz represents the number of nurses with grade g required on the jth day and the zth shift; q ig If the grade of the i-th nurse is not lower than g, the value is 1, otherwise it is 0; low represents the lower limit of the number of shifts a nurse can work in a scheduling cycle; up represents the upper limit of the number of shifts a nurse can work in a scheduling cycle; w ijz Indicates whether the i-th nurse takes leave for the z-type shift on the j-th day. If so, the value is 1, otherwise 0; a i3 Indicates whether the i-th nurse is on night shift.

9. The intelligent scheduling system for radiology nurses according to claim 1 is characterized in that: The specific implementation process of evaluating the efficiency of the scheduling system in the real-time optimization feedback module is as follows: L1. Collect feedback information about nurses’ shifts within a preset time period through a preset feedback mechanism, including work quality and satisfaction scores; L2. Apply statistical analysis methods to process the collected data and extract key indicators, which include the average satisfaction score X and the average work quality score Y. The specific calculation formula is: and Among them, x i and i are the satisfaction and work quality of a single nurse, respectively, and n is the total number of nurses; L3. Calculate the efficiency Q of the scheduling system based on the average satisfaction score X and the average work quality score Y. The specific calculation formula is: Q=X*a+Y*b, where a and b represent the average satisfaction rating coefficient and the average work quality rating coefficient respectively.

Images