System, method and equipment for scheduling medical technicians in imaging department and storage medium
By converting scheduling rules into a computable form, setting priorities, and performing conflict detection and resolution, combined with reinforcement learning optimization, the problem of rule conflicts in scheduling software was solved, improving the rationality of scheduling and employee satisfaction.
Patent Information
- Application Number
- CN202511097469.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-06
- Publication Date
- 2025-11-18
AI Technical Summary
The high complexity of scheduling rules in existing automatic scheduling software leads to frequent rule conflicts, affecting the reliability and rationality of the scheduling software, increasing the workload of administrators, and reducing the work quality and satisfaction of employees.
The rule modeling module converts scheduling rules into a computable form, sets rule priorities, uses a conflict detection module to identify and resolve conflicts, and combines a reinforcement learning optimization module to comprehensively consider employee preferences and work needs to optimize the scheduling scheme.
It enables automatic detection and resolution of scheduling rule conflicts, improving the rationality and efficiency of scheduling plans, and enhancing employee work quality and satisfaction.
Smart Images
Figure CN120977523A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of personnel management conflict detection technology, specifically relating to a scheduling system, method, equipment, and storage medium for medical, technical, and nursing staff in the radiology department. Background Technology
[0002] Scheduling medical, technical, and nursing staff in the radiology department is a routine task, performed weekly. However, due to the high degree of arbitrariness and complexity of scheduling, it is primarily done manually by the administrators of the department's medical, technical, and nursing teams. This method results in a high workload for the staff.
[0003] The emergence of automated scheduling software has alleviated the labor intensity problem of scheduling work to some extent. However, scheduling managers often add more and more scheduling rules or achieve specific management objectives. Scheduling rules include, but are not limited to: setting employee workload, setting available work shifts, setting daily available work shifts, setting consecutive shifts, setting fixed shifts, and setting desired shifts. Some rules apply to all employees, while others apply to individual employees. As the number of scheduling rules increases, potential conflicts between rules also increase, affecting the reliability and rationality of the scheduling software. Summary of the Invention
[0004] To address the rule conflict issues in existing automatic scheduling software and improve its reliability and rationality, this application proposes a scheduling system, method, equipment, and storage medium for medical, technical, and nursing staff in the radiology department. This application makes the scheduling rules more reasonable by detecting and resolving conflicts. At the same time, it also comprehensively considers employee preferences and work needs, learns and optimizes the scheduling scheme, thereby quickly obtaining a more reasonable scheduling scheme, improving work quality and employee satisfaction.
[0005] This application is achieved through the following technical solution: A scheduling conflict detection system for medical, technical, and nursing staff in the radiology department, the scheduling system comprising: The rule modeling module converts scheduling rules from textual form into a computable form. The rule priority module sets the priority for scheduling rules; A conflict detection module performs conflict detection on the computable scheduling rules generated by the rule modeling module and identifies conflicting rules. A conflict resolution module provides solution suggestions for conflict rules identified by the conflict detection module based on the priority set by the rule priority module. The user interaction module is used to visually output solution suggestions and prompt the scheduling administrator to modify the scheduling rules to resolve conflicts. In addition, there is a reinforcement learning optimization module, which comprehensively considers scheduling rules, employee preferences and work requirements, and uses an iterative learning algorithm to continuously optimize the scheduling scheme to obtain the optimal scheduling scheme.
[0006] In some implementations, the rule modeling module is configured to execute the following algorithm: For each scheduling rule, it is first broken down into several sub-rules, each of which represents a specific constraint or objective function; Each sub-rule is converted into a corresponding mathematical expression or logical rule; For each sub-rule, it is categorized and stored in different data structures according to its type; Integrate all sub-rules into a complete scheduling rule model.
[0007] In some implementations, the rule priority module is configured to execute the following algorithm: Each scheduling rule is assigned a priority value, where the priority value is a positive integer, and the larger the value, the higher the priority. For each scheduling rule, it is assigned to three priority levels based on its priority value: high, medium, and low. High priority rules must be followed, low priority rules can be adjusted, and medium priority rules are between high and low priority. All scheduling rules are sorted according to their priority level, and rules with the same priority level are sorted according to their priority value, forming an ordered list of scheduling rules.
[0008] In some implementations, the collision detection module is configured to execute the following algorithm: For each scheduling rule, compare it with other related rules according to its type to determine if there are any conflicts; If two rules do not conflict, then the next rule is compared; if two rules conflict, then the number, priority level and priority value of the two rules are recorded as a conflict item and stored in the conflict set. If all scheduling rules have been compared, then it is determined whether the conflict set is empty. If the conflict set is empty, it means that there is no conflict in the scheduling rules. If the conflict set is not empty, it means that there is a conflict in the scheduling rules.
[0009] In some implementations, the conflict resolution module is configured to execute the following algorithm: For each conflict item in the conflict set, it is processed in descending order of priority level and priority value. For each conflict, find the corresponding scheduling rule based on its rule number, analyze the cause of the conflict, and generate corresponding solution suggestions; The solution suggestions are output to the user interaction module.
[0010] In some implementations, the user interaction module is configured to execute the following algorithm: When the conflict resolution module outputs a solution suggestion, the user interaction module displays it to the scheduling administrator in the form of a graphical interface. The displayed content includes: information about the conflict item, a list of solution suggestions, an OK button, and a Modify button. The scheduling administrator selects one or more solutions from the list of suggested solutions, clicks the confirmation button, and the system modifies the scheduling rules according to the selected solutions, and then returns to the conflict detection module to re-detect; Alternatively, the scheduling administrator can click the "Modify" button, and the system will redirect to the scheduling rule settings interface. The scheduling administrator can then modify the scheduling rules or rule priorities in the scheduling rule settings interface, and then return to the conflict detection module to re-detect.
[0011] In some implementations, the reinforcement learning optimization module is configured to execute the following algorithm: Taking into account scheduling rules, employee preferences, and work requirements, an initial scheduling plan is generated. The initial scheduling scheme is iteratively updated using the Q-learning algorithm to obtain the optimal scheduling scheme.
[0012] Secondly, this application proposes a scheduling method for radiology medical, technical, and nursing staff, the scheduling method comprising: Convert scheduling rules from text format to a computable format; Set priorities for scheduling rules; Conflict detection is performed on computable scheduling rules to identify conflicting rules; Based on the priority of scheduling rules, provide suggested solutions for identified conflicting rules; Visualize the proposed solutions to prompt scheduling administrators to modify scheduling rules to resolve conflicts. Taking into account scheduling rules, employee preferences, and work requirements, the scheduling scheme is continuously optimized using an iterative learning algorithm to obtain the optimal scheduling scheme.
[0013] Thirdly, this application proposes an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.
[0014] Fourthly, this application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.
[0015] This application proposes a scheduling system, method, equipment, and storage medium for radiology medical, technical, and nursing staff. It can automatically detect and quickly resolve scheduling rule conflicts, and comprehensively consider employee preferences and work needs. It can also automatically and dynamically adjust the scheduling plan according to the actual situation, thereby quickly obtaining a more reasonable scheduling plan, improving work quality and employee satisfaction. Attached Figure Description
[0016] The accompanying drawings, which are included to provide a further understanding of the embodiments of this application and form part of this application, do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 This is a system principle block diagram proposed in the embodiments of this application; Figure 2 This is a suggested example of rule conflict handling in an embodiment of this application. Figure 3 This is a flowchart of the method proposed in an embodiment of this application; Figure 4 This is a schematic diagram of the interface for the system setting rule R1 proposed in the embodiments of this application; Figure 5 This is a schematic diagram of the interface for system setting rule R2 proposed in the embodiments of this application; Figure 6 This is a schematic diagram of the interface for system setting rule R3 proposed in the embodiments of this application; Figure 7 This is a schematic diagram of the interface for system setting rule R4 proposed in the embodiments of this application; Figure 8 This is a schematic diagram illustrating the results of collision detection using the system proposed in the embodiments of this application; Figure 9 A schematic diagram illustrating conflict resolution using the system proposed in the embodiments of this application; Figure 10 This is a schematic diagram showing the result of performing conflict detection again using the system proposed in the embodiments of this application; Figure 11 To utilize the scheduling scheme generated by the system proposed in the embodiments of this application. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.
[0018] Example: Existing automatic scheduling software suffers from numerous scheduling rules, which exacerbate conflicts between these rules. This leads to a significant discrepancy between the software's reliability and rationality and actual needs, thus impacting staff performance and satisfaction. To address this, this embodiment proposes a scheduling system for medical, technical, and nursing staff in the radiology department.
[0019] like Figure 1 As shown, the scheduling system proposed in this embodiment includes: The rule modeling module converts scheduling rules from textual form into a computable form, such as mathematical expressions or logical rules, to facilitate subsequent conflict detection and resolution.
[0020] The rule priority module sets the priority for scheduling rules so that, in the event of a conflict, certain rules can be abandoned or retained based on their priority.
[0021] The conflict detection module performs conflict detection on the computable scheduling rules generated by the rule modeling module, identifying conflicting (inconsistent or contradictory) rules.
[0022] The conflict resolution module provides suggested solutions for rule conflicts identified by the conflict detection module based on the priority set by the rule priority module, such as deleting or modifying certain rules to eliminate the conflict.
[0023] The user interaction module allows the user to interact with the scheduling administrator, prompting them to modify scheduling rules to resolve conflicts. The scheduling administrator can adjust the rules based on the system's suggestions, or modify the rules or rule priorities themselves, and then resubmit the scheduling rules until there are no conflicts.
[0024] In addition, there is a reinforcement learning optimization module, which comprehensively considers scheduling rules, employee preferences and work needs, and uses iterative learning algorithms to continuously optimize the scheduling scheme to obtain the optimal scheduling scheme to adapt to the ever-changing work needs and employee preferences.
[0025] Furthermore, the rule modeling module is specifically configured to implement the following algorithms: For each scheduling rule, it is first broken down into several sub-rules, each of which represents a specific constraint or objective function; For each sub-rule, based on the variables, constants, operators, logical symbols, etc. involved, it is converted into a corresponding mathematical expression or logical rule; For each sub-rule, it is categorized and stored in different data structures according to its type (constraint or objective function); for example, constraints can be stored in a matrix, and objective functions can be stored in a vector. All sub-rules are integrated into a complete scheduling rule model, which can be represented as an optimization problem, namely, maximizing or minimizing a certain objective function while satisfying all constraints.
[0026] For example, suppose there are the following three scheduling rules: Rule 1: Each medical technician or nurse may work a maximum of 5 days per week; Rule 2: Each medical technician or nurse must work at least one night shift per week; Rule 3: Try to balance the workload of each medical technician and nurse, that is, minimize the difference in working hours among medical technicians and nurses.
[0027] The rule modeling module then decomposes the three rules into the following sub-rules: Sub-rule 1: For all i = 1, 2, ..., n, x_i ≤ 5, where n is the total number of medical and technical personnel and x_i is the number of working days of the i-th medical and technical personnel. This is a constraint condition. Sub-rule 2: ∑(j=1…m) y_j ≥ 1, where m is the total number of shifts, and y_j is whether the j-th shift is a night shift. If it is, the value is 1, otherwise it is 0. This is a constraint condition. Sub-rule 3: min(max(z_i)-min(z_i)), where z_i is the working time of the i-th medical technician, which is an objective function.
[0028] The three sub-rules mentioned above are stored in different data structures, as follows: Constraint matrix: [x_1,x_2, …,x_n], where x_i≤ 5 indicates a constraint that each employee's working days do not exceed 5 days; [y_1, y_2, …, y_m]],∑(j=1…m) y_j ≥ 1 means that each employee works at least one night shift; The objective function vector is [z_1, z_2, …, z_n], where min(max(z_i)-min(z_i)) represents minimizing the difference in working hours among medical technicians and nurses.
[0029] The data structure is integrated into a scheduling rule model, namely: min(max(z_i)-min(z_i)) st x_i ≤ 5 (i=1,2,…,n), ∑(j=1…m) y_j ≥ 1.
[0030] Furthermore, the rule priority module is specifically configured to execute the following algorithm: For each scheduling rule, a priority value is assigned based on factors such as its source, importance, and urgency. This value is a positive integer, and the larger the value, the higher the priority. For each scheduling rule, it is assigned to three priority levels based on its priority value: high, medium, and low. High-priority rules are mandatory, such as legal regulations and hospital policies. Low-priority rules are adjustable, such as the personal preferences of medical and technical staff. Medium-priority rules are between high and low, such as the workload balance of medical and technical staff. All scheduling rules are sorted according to their priority level, and rules with the same priority level are sorted according to their priority value, forming an ordered list of scheduling rules.
[0031] For example, suppose there are the following four scheduling rules: Rule 1: Each medical technician or nurse can work a maximum of 5 days per week. This rule is based on hospital policy, is of high importance and urgency, has a priority value of 10, and is of high priority level. Rule 2: Each medical technician or nurse must work at least one night shift per week. This rule is based on hospital policy, is of high importance, medium urgency, has a priority value of 8, and is of high priority level. Rule 3: Try to balance the workload of each medical technician and nurse. The source is the radiology department, the importance is medium, the urgency is low, the priority value is 5, and the priority level is medium. Rule 4: Try to satisfy the personal preferences of medical technicians and nurses. The source is the medical technician or nurse themselves. The importance and urgency are low. The priority value is 2 and the priority level is low.
[0032] The rule priority module can then sort the four rules according to their priority level and priority value, forming an ordered list of scheduling rules, as follows: Rule 1: Each medical technician or nurse can work a maximum of 5 days per week, with a priority value of 10, which is high priority level; Rule 2: Each medical technician / nursing staff member must work at least one night shift per week, with a priority value of 8, indicating a high priority level; Rule 3: Try to balance the workload of each medical technician and nurse, with a priority value of 5 and a priority level of medium; Rule 4: Try to satisfy the personal preferences of medical and technical staff. Priority value is 2, and priority level is low.
[0033] Furthermore, the collision detection module is specifically configured to execute the following algorithm: For each scheduling rule, it is compared with other related rules according to its type (constraint or objective function) to determine whether there are any inconsistencies or contradictions. If there is no inconsistency or contradiction between the two rules, then continue to compare the next rule; if there is an inconsistency or contradiction between the two rules, then record the number, priority level, priority value and other information of the two rules as a conflict item and store it in a conflict set. If all rules have been compared, then check if the conflict set is empty. If the conflict set is empty, it means that there is no conflict in the scheduling rules. If the conflict set is not empty, it means that there is a conflict in the scheduling rules.
[0034] Taking the above scheduling rule list as an example, conflict detection is performed on it, resulting in the following conflict combinations: Conflict 1: Rule 1 and Rule 2, because if each medical technician works a maximum of 5 days a week, some medical technicians may not be able to be scheduled for night shifts, which would result in not meeting Rule 2; Conflict Item 2: Rules 3 and 4, because if we try to satisfy the personal preferences of medical technicians and nurses as much as possible, it may lead to an imbalance in their workload, resulting in failure to meet Rule 3.
[0035] Furthermore, the conflict resolution module is specifically configured to execute the following algorithm: For each conflicting item in the conflict set, it is processed in descending order of priority level and priority value. For each conflict, find the corresponding scheduling rule based on its rule number, analyze the cause of the conflict, and generate corresponding solution suggestions; The proposed solution is then presented to the user interaction module, awaiting user confirmation or modification.
[0036] Taking the above set of conflicts as an example, the following solution suggestions can be generated for these two conflicting items: Solution 1: Modify rule 1, changing the maximum working days per week for each medical technician to a maximum working days per week of 6 days. This will increase the scheduling options for night shifts and satisfy rule 2. Solution 2: Delete rule 4, disregard the personal preferences of medical and technical staff, and schedule shifts solely based on the principle of workload balance, thus satisfying rule 3; Solution 3: Adjust the priority value of rule 4 to 1, so that it will be abandoned first in case of conflict, thus satisfying rule 3.
[0037] Furthermore, the user interaction module is configured to execute the following algorithm: When the conflict resolution module outputs suggested solutions, it will display them to the scheduling administrator in a graphical interface. The displayed content includes: information about the conflict, a list of suggested solutions, a confirmation button, and a modification button. Figure 2 As shown; The scheduling administrator can select one or more solutions from the list of suggested solutions, click the confirmation button, and the system will modify the scheduling rules according to the selected solutions, and then return to the conflict detection module to re-detect; The scheduling administrator can also click the "Modify" button, which will redirect the system to the scheduling rule settings interface. The scheduling administrator can modify the scheduling rules or rule priorities on this interface, and then return to the conflict detection module to re-detect.
[0038] Furthermore, the reinforcement learning optimization module is specifically configured to execute the following algorithms: Taking into account scheduling rules, employee preferences, and departmental work needs, an initial scheduling plan is generated. The initial scheduling plan is iteratively updated to obtain the optimal scheduling plan.
[0039] Taking a specific application scenario as an example: the radiology department has 5 employees (A, B, C, D, and E), working 5 days a week, with 2 employees working the morning shift and 1 employee working the evening shift each day. The goal is to optimize the shift schedule to balance the workload of employees while satisfying both individual employee preferences and the department's work needs. This embodiment uses the Q-learning algorithm for iterative updates to obtain a better shift schedule.
[0040] Employee preferences: A and B prefer not to work weekends; C, D, and E have no particular preference.
[0041] Initial scheduling strategy: Saturday: A is on the morning shift, B is on the evening shift, and C, D, and E are off. Sunday: D is on the morning shift, E is on the evening shift, and A, B, and C are off.
[0042] State space: Define state s as the employee's weekend shift schedule, for example: s = [A: Early shift B: Late shift C: Rest D: Rest E: Rest] Action space: Define action 'a' as adjusting an employee's shift. For example, a=(A,"Saturday evening shift") means adjusting employee A from the Saturday morning shift to the evening shift.
[0043] Reward function: The reward function, which considers employee satisfaction, scheduling compliance, and workload balance, is expressed as:
[0044] Here, satisfaction represents the degree to which employee preferences are met, violation represents the degree to which rules are violated (such as excessive workload), balance represents the degree to which workload is balanced among employees, and w1, w2 and w3 are their respective weights.
[0045] Strategy Update: The update strategy uses the Q-learning algorithm: ; Where α is the learning rate and γ is the discount factor.
[0046] Learning process: Initialization: Based on employee preferences and departmental work needs, randomly assign employees to weekend shifts and generate an initial strategy π; Iteration: Select action a based on the current policy π; Perform the action and update the state s; Calculate the reward R(s,a); Update Q value; Update strategy π based on Q value.
[0047] After several rounds of iterative learning, the optimal scheduling strategy is obtained, which can automatically take into account employee preferences and rules to provide more reasonable scheduling solutions. Ultimately, the system can automatically adjust the schedule so that employee A and employee B do not work on weekends, while ensuring that the workload of other employees remains balanced.
[0048] This embodiment detects and resolves conflicts in the scheduling rules to identify those without conflicts. It also learns and optimizes the scheduling rules by comprehensively considering factors such as employee preferences and departmental work needs, thereby obtaining a more realistic scheduling plan that improves the work quality and satisfaction of employees.
[0049] Based on the same technical concept described above, this embodiment also proposes a scheduling method for medical, technical, and nursing staff in the radiology department, such as... Figure 3 As shown, the scheduling method proposed in this embodiment includes: Step 100: Convert the scheduling rules from text to a computable form, such as mathematical expressions or logical rules, to facilitate subsequent conflict detection and resolution; the specific implementation process is as described in the employee preference and scheduling rule modeling module above, and will not be elaborated further here.
[0050] Step 200: Set priorities for scheduling rules so that when conflicts occur, some rules can be abandoned or retained based on their priority. The specific implementation process is as described in the rule priority module execution algorithm above, and will not be elaborated further here.
[0051] Step 300: Perform conflict detection on the computable scheduling rules to identify inconsistent or contradictory rules; the specific implementation process is as described in the conflict detection module algorithm above, and will not be elaborated further here.
[0052] Step 400: If conflicting rules exist, provide suggested solutions for the identified conflicts based on the priority of the scheduling rules, such as deleting or modifying certain rules to eliminate the conflicts; the specific implementation process is as described in the conflict resolution module execution algorithm above, and will not be elaborated further here.
[0053] Step 500: Visualize the proposed solution and prompt the scheduling administrator to modify the scheduling rules to resolve the conflict; the specific implementation process is the same as the algorithm executed in the user interaction module above, and will not be elaborated further here.
[0054] Step 600: Taking into account scheduling rules, employee preferences and departmental work needs, the scheduling scheme is continuously optimized using an iterative learning algorithm to obtain the optimal scheduling scheme; the specific implementation process is as described in the reinforcement learning optimization module above, and will not be elaborated further here.
[0055] Example 2: This embodiment uses the scheduling system proposed in the above embodiment to analyze the scheduling problem in the radiology department. The radiology department has 7 medical, technical, and nursing staff (A, B, C, D, E, F, G, H), who work 5 days a week, divided into two shifts (morning shift and evening shift) each day. The specific scheduling rules and their priorities are as follows: Rule R1: Each medical technician or nurse may work a maximum of 5 days per week (hospital policy, priority 10). Each employee can work no more than 5 days a week. This should be set in the system as follows: Figure 4 As shown.
[0056] Rule R2: Each medical technician or nurse can only work one night shift per week (hospital policy, priority 8). Each employee can only be scheduled for one night shift per week. This should be configured in the system as follows: Figure 5 As shown.
[0057] Rule R3: Strive to balance the workload of each medical, technical, and nursing staff (departmental requirement, priority 5). Workload balancing means that the workload of each employee (including morning and evening shifts) should be as close as possible. The morning shift is counted as 1 unit of workload, and the evening shift as 2 units of workload. This is configured in the system as follows: Figure 6 As shown.
[0058] Rule R4: Try to satisfy the personal preferences of medical technicians and nurses (personal requirements, priority 2). If employees have specific shift preferences, these should be considered whenever possible. This should be configured in the system. Figure 7 As shown.
[0059] Conflict detection: By using a collision detection algorithm, the detection results are shown in Figure 8: Conflict item: Shift inspection failed (general-purpose inspection) Problem: The minimum number of workers required for a shift is greater than the maximum workload per worker. Reason: The maximum workload of personnel is too low or the minimum workload of shifts is too high.
[0060] Conflict item: The workload of the specific shift group inspection failed (specific test). Problem: A CT day shift requires at least 42 staff members, but only 35 staff members were actually assigned to the shift.
[0061] Solution: Increase workload for additional staff or reduce workload for night shifts.
[0062] Conflict resolution is performed through the conflict resolution module, such as... Figure 9 As shown.
[0063] The results of the second check are as follows Figure 10 As shown: As shown in the figure, after resolving any conflicts, an iterative learning algorithm is executed to obtain the final scheduling scheme, as follows: Figure 11 As shown.
[0064] This embodiment, by considering the priority and urgency of rules, can propose reasonable solutions and allow administrators to make the final decision. This approach not only improves the efficiency and rationality of scheduling but also ensures employee job satisfaction and workload balance.
[0065] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0066] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0068] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0069] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A scheduling system for medical, technical, and nursing staff in the field of radiology, characterized in that, The scheduling system includes: A rule modeling module that converts employee preferences and scheduling rules from textual form into a computable form; The rule priority module sets the priority for scheduling rules; A conflict detection module performs conflict detection on the computable scheduling rules generated by the rule modeling module and identifies conflicting rules. A conflict resolution module provides solution suggestions for conflict rules identified by the conflict detection module based on the priority set by the rule priority module. The user interaction module is used to visually output solution suggestions and prompt the scheduling administrator to modify the scheduling rules to resolve conflicts. In addition, there is a reinforcement learning optimization module, which comprehensively considers scheduling rules, employee preferences and work requirements, and uses an iterative learning algorithm to continuously optimize the scheduling scheme to obtain the optimal scheduling scheme.
2. The scheduling system for medical, technical, and nursing staff in the radiology department according to claim 1, characterized in that, The rule modeling module is configured to execute the following algorithm: For each scheduling rule, it is first broken down into several sub-rules, each of which represents a specific constraint or objective function; Each sub-rule is converted into a corresponding mathematical expression or logical rule; For each sub-rule, it is categorized and stored in different data structures according to its type; Integrate all sub-rules into a complete scheduling rule model.
3. The scheduling system for medical, technical, and nursing staff in the radiology department according to claim 1, characterized in that, The rule priority module is configured to execute the following algorithm: Each scheduling rule is assigned a priority value, where the priority value is a positive integer, and the larger the value, the higher the priority. For each scheduling rule, it is assigned to three priority levels based on its priority value: high, medium, and low. High priority rules must be followed, low priority rules can be adjusted, and medium priority rules are between high and low priority. All scheduling rules are sorted according to their priority level, and rules with the same priority level are sorted according to their priority value, forming an ordered list of scheduling rules.
4. The scheduling system for medical, technical, and nursing staff in the radiology department according to claim 1, characterized in that, The collision detection module is configured to execute the following algorithm: For each scheduling rule, compare it with other related rules according to its type to determine if there are any conflicts; If two rules do not conflict, then the next rule is compared; if two rules conflict, then the number, priority level and priority value of the two rules are recorded as a conflict item and stored in the conflict set. If all scheduling rules have been compared, then it is determined whether the conflict set is empty. If the conflict set is empty, it means that there is no conflict in the scheduling rules. If the conflict set is not empty, it means that there is a conflict in the scheduling rules.
5. A scheduling system for medical, technical, and nursing staff in the radiology department according to claim 4, characterized in that, The conflict resolution module is configured to execute the following algorithm: For each conflict item in the conflict set, it is processed in descending order of priority level and priority value. For each conflict, find the corresponding scheduling rule based on its rule number, analyze the cause of the conflict, and generate corresponding solution suggestions; The solution suggestions are output to the user interaction module.
6. A scheduling system for medical, technical, and nursing staff in the radiology department according to claim 5, characterized in that, The user interaction module is configured to execute the following algorithm: When the conflict resolution module outputs a solution suggestion, the user interaction module displays it to the scheduling administrator in the form of a graphical interface. The displayed content includes: information about the conflict item, a list of solution suggestions, an OK button, and a Modify button. The scheduling administrator selects one or more solutions from the list of suggested solutions, clicks the confirmation button, and the system modifies the scheduling rules according to the selected solutions, and then returns to the conflict detection module to re-detect; Alternatively, the scheduling administrator can click the "Modify" button, and the system will redirect to the scheduling rule settings interface. The scheduling administrator can then modify the scheduling rules or rule priorities in the scheduling rule settings interface, and then return to the conflict detection module to re-detect.
7. A scheduling system for medical, technical, and nursing staff in the radiology department according to any one of claims 1-6, characterized in that, The reinforcement learning optimization module is configured to execute the following algorithm: Taking into account scheduling rules, employee preferences, and work requirements, an initial scheduling plan is generated. The initial scheduling scheme is iteratively updated using the Q-learning algorithm to obtain the optimal scheduling scheme.
8. A method for scheduling medical, technical, and nursing staff in the radiology department, characterized in that, The scheduling method includes: Convert scheduling rules from text format to a computable format; Set priorities for scheduling rules; Conflict detection is performed on computable scheduling rules to identify conflicting rules; Based on the priority of scheduling rules, provide suggested solutions for identified conflicting rules; Visualize the proposed solutions to prompt scheduling administrators to modify scheduling rules to resolve conflicts. Taking into account scheduling rules, employee preferences, and work requirements, the scheduling scheme is continuously optimized using an iterative learning algorithm to obtain the optimal scheduling scheme.
9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method of claim 8.