Remote consultation real-time scheduling method and system based on uncertainty

Abstract

The application relates to the technical field of medical dispatching, and provides a remote consultation real-time scheduling method and system based on uncertainty, which can generate a static preliminary scheduling scheme of remote consultation based on a first static allocation constraint according to a reservation request submitted by an AP of a basic hospital in a determined environment without considering random factors, and can optimize the allocation of AP patients in the static preliminary scheduling scheme based on a second static allocation constraint under the influence of random factors to obtain a static optimization scheduling scheme of remote consultation; and can allocate remote consultation sequence positions for WP patients on the basis of the static optimization scheduling scheme, and dynamically update the allocation of AP patients affected in the static optimization scheduling scheme to generate a dynamic real-time scheduling scheme of remote consultation, so that the timeliness of consultation information is ensured, and the response capability of remote consultation is improved.

Description

Technical Field

[0001] This application relates to the field of medical scheduling technology, and in particular to a method and system for real-time scheduling of remote consultations based on uncertainty. Background Technology

[0002] With the advent of the digital age, telemedicine services have facilitated the downward flow of high-quality medical resources. Telemedicine services mainly include remote consultation, remote monitoring, remote pathology diagnosis, and remote medical education.

[0003] Remote consultation is a core service of telemedicine, requiring advance appointments. However, in daily operations, ad-hoc remote consultation requests often disrupt the original remote consultation schedule. As the number of users participating in remote consultations increases, telemedicine centers face significant uncertainty when scheduling remote consultations. If this uncertainty is not taken into account during the initial allocation of consultation slots, a large discrepancy may occur between the allocated appointment time and the actual consultation time. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for real-time scheduling of remote consultations based on uncertainty, so as to solve or alleviate the problems existing in the prior art.

[0005] To achieve the above objectives, this application provides the following technical solution: This application provides a real-time scheduling method for remote consultations based on uncertainty, comprising: under a deterministic environment that does not consider random factors, generating a static preliminary scheduling scheme for remote consultations based on appointment requests submitted by access patients (APs) from primary hospitals and a first static allocation constraint; under the influence of random factors, optimizing the allocation of APs in the static preliminary scheduling scheme based on a second static allocation constraint to obtain a static optimized scheduling scheme for remote consultations; assigning remote consultation order positions to patients with patient benefits (WPs) and dynamically updating the allocation of affected APs in the static optimized scheduling scheme to generate a dynamic real-time scheduling scheme for remote consultations. Preferably, by means of the first static allocation constraint: Generate a static preliminary scheduling plan for remote consultation; In the formula, Indicated as APs patients Corresponding remote consultation experts Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; This refers to the scheduling cycle duration in the static preliminary scheduling scheme; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients The allocation of consultation service time, For APs patients Assigned remote consultation experts The buffer threshold.

[0006] Preferably, through the second static allocation constraint: The allocation of APs patients in the static preliminary scheduling plan was optimized by using preset performance indicators under the influence of random factors. In the formula, Indicated as APs patients Corresponding remote consultation experts Consultation room and consultation location , hour, This represents the APs patients in the static preliminary scheduling scheme. Corresponding remote consultation experts Consultation room and the location of the first consultation; For AP patients in the first consultation position in the static preliminary scheduling plan Waiting time; For APs patients The unpunctual arrival time The total number of APs that submitted reservation requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as WPs patient Corresponding remote consultation experts Consultation room and consultation location , This refers to the number of unappointed patients who arrive randomly on the day of the remote consultation.

[0007] Preferably, under the influence of random factors, a greedy scheduling method is used to assign remote consultation sequence positions to WPs patients, and the allocation of affected APs patients in the static optimization scheduling scheme is dynamically updated to generate a dynamic real-time scheduling scheme for remote consultation.

[0008] Preferably, only under the influence of random arrival of WPs patients, the remote consultation room with the shortest waiting time and closest availability in the static optimization scheduling scheme is selected based on the greedy scheduling strategy and assigned to WPs patients. By constructing a global update model, the allocation of affected APs patients in the static optimization scheduling scheme is updated under the global update allocation constraint, thereby generating a dynamic real-time scheduling scheme for remote consultation scheduling.

[0009] Preferably, the allocation of affected APs patients in the static optimization scheduling scheme is updated according to the global update allocation constraints; wherein, the global update allocation constraints include: APs update allocation constraints and WPs update allocation constraints, and the APs update allocation constraints are as follows: In the formula, Indicated as APs patients Corresponding remote consultation experts Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; This refers to the scheduling cycle duration in the static preliminary scheduling scheme. Consultation room The time limit during consultations; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients Assigned remote consultation experts The buffer threshold; WPs update assignment constraints: In the formula, For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts Consultation room Consultation location ; For WPs patients The allocation of consultation service time; The number of unappointed patients who arrive randomly on the day of the remote consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts were assigned. Consultation room Consultation location ; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation.

[0010] Preferably, the greedy strategy-based rolling time-domain optimization freezes the remote consultation scheduling of the current rolling window in the static optimization scheduling scheme, and follows the formula: The remote consultation scheduling in the static optimization scheduling scheme is sequentially and continuously optimized to generate a dynamic real-time scheduling scheme for remote consultation scheduling; In the formula, Remote consultation experts in static optimization scheduling scheme Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For APs patients The arrival deviation, For APs patients The actual consultation service duration For APs patients The allocation of consultation service time, For WPs patients The allocation of consultation service time; The rolling step size for the rolling time-domain optimization strategy.

[0011] Preferably, in the case where no WPs patients arrive and there are random factors for APs patients, wherein the random factors for APs patients include at least the uncertainty of APs patients' arrival and / or the randomness of APs patients' service duration; According to the rolling strategy: The consultation start time for affected APs in the static optimization scheduling scheme is updated within each rolling window; In the formula, For APs patients The actual arrival time For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; Characterized as AP patients The assigned consultation start times follow a uniform distribution; Characterizing AP patients The actual consultation service duration follows a uniform distribution. For APs patients The allocation of consultation service time, It is a fixed constant; For APs patients The deviation was achieved. For APs patients Assigned remote consultation experts The buffer threshold; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation.

[0012] Preferably, in the case of random arrival of WPs patients and random factors of APs patients, wherein the random factors of APs patients include at least the uncertainty of APs patients' arrival and / or the randomness of APs patients' service duration; According to the model: Assigning remote consultation rooms to WPs patients and consultation location In the formula, The total number of APs that submitted reservation requests. For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Arrival time; And according to the model: Update the consultation start time for affected APs patients in the static optimization scheduling scheme; In the formula, In cases where WP patients arrive randomly and AP patients have random factors, remote consultation experts are provided. Consultation room Corresponding APs patients Reassignment of consultation positions , To reallocate consultation positions Updated consultation start time for APs patients. For WPs patients The allocation of consultation service duration; A collection of WPs patients.

[0013] This embodiment also provides a real-time scheduling system for remote consultations based on uncertainty, which dynamically schedules remote consultations using any of the aforementioned real-time scheduling methods for remote consultations based on uncertainty. The system includes: The static scheduling unit is configured to generate a preliminary static scheduling scheme for remote consultation based on the appointment requests submitted by APs of the basic hospital under a deterministic environment that does not consider random factors; and to optimize the allocation of APs patients in the preliminary static scheduling scheme based on the second static allocation constraint under the influence of random factors, so as to obtain a static optimized scheduling scheme for remote consultation. The dynamic scheduling unit is configured to assign remote consultation sequence positions to WPs patients and dynamically update the allocation of affected APs patients in the static optimized scheduling scheme, generating a dynamic real-time scheduling scheme for remote consultations.

[0014] Beneficial effects: This application provides a method and system for real-time scheduling of remote consultations based on uncertainty. Under a deterministic environment that does not consider random factors, a static preliminary scheduling scheme for remote consultations is generated based on appointment requests submitted by access points (APs) from basic hospitals, and based on a first static allocation constraint, the allocation of APs in the static preliminary scheduling scheme is optimized under the influence of random factors, resulting in a static optimized scheduling scheme for remote consultations. Based on the static optimized scheduling scheme, remote consultation order positions are assigned to patients who have been treated (WPs), and the allocation of affected APs in the static optimized scheduling scheme is dynamically updated to generate a dynamic real-time scheduling scheme for remote consultations.

[0015] Therefore, after initially allocating AP appointment requests without considering the influence of random factors, the allocation of AP appointment requests is optimized through performance indicators in combination with random factors to complete static scheduling; on the basis of static scheduling, random arrival of WP patients is processed, and the scheduling allocation of APs affected in static scheduling is updated and optimized. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. Wherein: Figure 1 This is a schematic diagram illustrating a scenario of a real-time scheduling method for remote consultation based on uncertainty, provided according to some embodiments of this application. Figure 2 This is a flowchart illustrating a real-time scheduling method for remote consultation based on uncertainty, according to some embodiments of this application. Figure 3 This is a schematic diagram of a global update model affected only by WPs insertion, provided according to some embodiments of this application; Figure 4 This is a schematic diagram illustrating the distribution of waiting times for APs patients with different numbers of WPs according to an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a real-time scheduling system for remote consultation based on uncertainty, according to some embodiments of this application. Detailed Implementation

[0017] The present application will now be described in detail with reference to the accompanying drawings and embodiments. Various examples are provided by way of explanation and not by way of limitation. In fact, those skilled in the art will understand that modifications and variations can be made to the present application without departing from the scope or spirit of the present application. For example, a feature shown or described as part of one embodiment may be used in another embodiment to produce yet another embodiment. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention should fall within the scope of protection of the embodiments of the present invention.

[0018] The main uncertainties in the actual operation of telemedicine consultations include: random consultation service times, late arrivals of patients and specialists, and the random arrival of walk-in patients (WPs) on the day of the consultation. Dynamically updating appointment allocations during the telemedicine service can significantly improve operational efficiency. Currently, telemedicine scheduling, especially the scheduling of appointment patients (APs), is still largely decided by scheduling personnel based on experience. When WPs arrive, scheduling personnel assign them to locations for the telemedicine consultation and adjust the appointments of subsequent APs. While this manual method is flexible, its efficiency decreases sharply as the demand for telemedicine consultations continues to grow. Improving the accuracy of appointment times for APs at primary hospitals while minimizing unnecessary waiting times is a significant challenge for telemedicine centers in scheduling consultations.

[0019] Based on this, this embodiment provides a real-time scheduling method for remote consultations based on uncertainty. It achieves and evaluates real-time scheduling of remote consultations under the influence of uncertainty through one or two-stage static scheduling: First, in a deterministic environment without considering random factors, remote consultations are initially allocated according to the appointment requests submitted by APs from basic hospitals; then, random factors are introduced to optimize the initial allocation of APs through performance indicators; furthermore, when WPs arrive, remote consultation order positions are assigned to WPs patients, and the allocation of affected APs patients is dynamically updated.

[0020] like Figures 1 to 4 As shown, this real-time scheduling method for remote consultations based on uncertainty includes: Step S101: Under a deterministic environment that does not consider random factors, based on the appointment requests submitted by APs of the basic hospital, a static preliminary scheduling scheme for remote consultation is generated based on the first static allocation constraint; and under the influence of random factors, the allocation of APs patients in the static preliminary scheduling scheme is optimized based on the second static allocation constraint to obtain a static optimized scheduling scheme for remote consultation.

[0021] In this embodiment, a two-stage stochastic mixed-integer programming model is established within a static scheduling framework to generate an initial allocation scheme for AP (Advanced Patient Services) appointment scheduling. Here, all patients are scheduled for remote consultations in an appointment order. The service duration of each AP is a random variable within a predefined interval, and the actual arrival time of an AP relative to its assigned consultation start time is also random, with the deviation from the assigned start time being a random interval variation. Furthermore, consultation experts typically do not arrive earlier than their assigned consultation start time. Therefore, during the consultation scheduling process, a buffer time for consultation experts is added to mitigate the impact of late arrivals on the remote consultation period.

[0022] In the static scheduling phase, the objective is to minimize the weighted total expected cost, including: the cost of opening the remote consultation room, the cost of patient allocation, the cost of idle time, the cost of APs waiting, and the cost of overtime work in the remote consultation room. Based on this, a static scheduling objective model is established: in, The unit cost for opening a remote consultation room; This indicates whether the remote consultation room is open. When the remote consultation room is open... When the remote consultation room is not open, ; The penalty unit cost for unused time The duration of the scheduling cycle; The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. For APs patients The allocation of consultation service time; Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location ; The unit cost of remote consultation scheduling; Expressing expectations; The unit cost for patients to wait; The unit cost of timeout Consultation room The time limit during the consultation.

[0023] In the first phase of static scheduling (generating a preliminary static scheduling scheme for remote consultations based on the appointment requests submitted by APs from primary hospitals and the first static allocation constraint), APs are assigned remote consultation experts, consultation rooms, and consultation positions. Each AP must and can only be assigned to one consultation position in one remote consultation room, and can only be assigned to one consultation expert. At any given time, each consultation expert can only consult for one patient, and each consultation room can only accommodate one patient. All consultations are conducted sequentially according to the assigned consultation positions. Furthermore, remote consultation rooms are not all activated at the beginning; they are only opened for use when remote consultations are actually allocated based on consultation demand. Simultaneously, all remote consultations allocated to each remote consultation room must be completed before the end of working hours.

[0024] Furthermore, by constructing the first static allocation constraint: Generate a static preliminary scheduling scheme for remote consultation; where, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; This refers to the scheduling cycle duration in the static preliminary scheduling scheme; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients The allocation of consultation service time, For APs patients Assigned remote consultation experts The buffer threshold.

[0025] In the second stage of static scheduling (under the influence of random factors, based on the second static allocation constraint, the allocation of APs patients in the initial static scheduling scheme is optimized by preset performance indicators to obtain the static optimized scheduling scheme for remote consultation), based on the waiting time of APs patients and the overtime hours of the remote consultation room, the second static allocation constraint is constructed as follows: The allocation of AP patients in the initial static scheduling scheme is optimized by using preset performance indicators under the influence of random factors. Where, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , hour, This represents the APs patients in the static preliminary scheduling scheme. Corresponding remote consultation experts were assigned. Consultation room and the location of the first consultation; For AP patients in the first consultation position in the static preliminary scheduling plan Waiting time; For APs patients The unpunctual arrival time The total number of APs that submitted reservation requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as WPs patient Corresponding remote consultation experts were assigned. Consultation room and consultation location , This refers to the number of unappointed patients who arrive randomly on the day of the remote consultation.

[0026] In this static scheduling target model, the actual service duration and arrival behavior of APs are random. To reduce the randomness caused by uncertainties, this embodiment first transforms the random problem into a deterministic problem using the sample average approximation method. That is, it linearizes the nonlinear constraints by applying the sample average approximation method, transforming the static scheduling model into a deterministic model. Then, the variable neighborhood search method is combined with the integer L-shaped method to solve the static scheduling target model. First, the following steps are taken: The random scenario approximates the uncertain distribution of the random variable, and sets a maximum number of iterations and a threshold without improvement; then, the basic assignment of APs is determined using an integer L-shaped method; then, a variable neighborhood search method is used to perturb the patient and consultation room assignments.

[0027] In this way, the opening cost of the first phase of remote consultation rooms can be controlled through static scheduling. Allocation costs for patients And the waiting costs of patients in the second phase of static scheduling. Overtime costs in remote consultation rooms The static scheduling of patients is measured and evaluated to minimize the expected total cost.

[0028] Step S102: Assign remote consultation sequence positions to WPs patients and dynamically update the allocation of affected APs patients in the static optimization scheduling scheme to generate a dynamic real-time scheduling scheme for remote consultation.

[0029] On the day of the telemedicine consultation, APs (Advanced Patients) conduct telemedicine consultations according to their assigned appointment times. When WPs (Patients Who Have Patients) arrive, the WPs are assigned to the appropriate telemedicine consultation sequence, and the assignments of all subsequently affected APs are updated. This update includes the start time of the telemedicine consultation, the assigned telemedicine consultation room, and the designated consultation specialist. Furthermore, each time a WP is inserted, the assignments of all subsequent patients in the same telemedicine consultation room need to be adjusted.

[0030] In this embodiment, based on static scheduling, a dynamic scheduling model constructed based on the influence of random factors is used to assign remote consultation order positions to WPs patients using greedy scheduling, and the allocation of affected APs patients in the static optimized scheduling scheme is dynamically updated to generate a dynamic real-time scheduling scheme for remote consultation.

[0031] In a specific example, under the influence of random arrival of WPs (Patients with Remote Consultation) patients, a greedy scheduling strategy selects the remote consultation room with the shortest waiting time and closest availability from the statically optimized scheduling scheme and assigns it to the WP patient. Here, for each WP, the least congested consultation room is identified, and an attempt is made to insert the WP patient at the location closest to its arrival location. Simultaneously, the original APs at that scheduling location are subsequently delayed, and the appointment information for that patient and all subsequent patients is updated.

[0032] Here, a global update model is constructed: Under global update allocation constraints, the allocation of affected APs in the statically optimized scheduling scheme is updated to generate a dynamic real-time scheduling scheme for remote consultation scheduling. The global update allocation constraints include APs update allocation constraints and WPs update allocation constraints. The APs update allocation constraints are as follows: In the formula, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; This refers to the scheduling cycle duration in the static preliminary scheduling scheme. Consultation room The time limit during consultations; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients Assigned remote consultation experts The buffer threshold.

[0033] When a WP is inserted before a given AP, the consultation start time is updated once; if multiple WPs are inserted, the update is performed multiple times. Here, the updated consultation start time is defined as... There are WPs update allocation constraints: In the formula, For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts were assigned. Consultation room Consultation location ; For WPs patients The allocation of consultation service time; The number of unappointed patients who arrive randomly on the day of the remote consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts were assigned. Consultation room Consultation location ; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation.

[0034] In another specific example, when allocating WPs, uncertainties such as patient arrival behavior, actual service time, and consultation expert delays are comprehensively considered. By combining the rolling time-domain optimization strategy with the greedy insertion method, positions are allocated to WPs, and a dynamic scheduling model based on the rolling time-domain optimization strategy is constructed. The scheduling allocation is updated within the moving time window to gradually optimize the remote consultation scheduling. In this process, uncertainties such as service time, patient arrival deviation, and arrival events on the same day are always taken into account.

[0035] Specifically, based on the greedy strategy for rolling time-domain optimization, the remote consultation scheduling of the current rolling window in the static optimization scheduling scheme is frozen, and the following formula is applied: The static optimization scheduling scheme for remote consultation scheduling is subjected to sequential rolling optimization to generate a dynamic real-time scheduling scheme for remote consultation scheduling; where, Remote consultation experts in static optimization scheduling scheme Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For APs patients The arrival deviation, For APs patients The actual consultation service duration For APs patients The allocation of consultation service time, For WPs patients The allocation of consultation service time; The rolling step size for the rolling time-domain optimization strategy.

[0036] In uncertain environments, dynamic factors are unpredictable, thus patient scheduling updates are continuous. Here, patient scheduling assignments are divided into rolling windows, with each scheduling update concentrated within the current rolling window, simplifying dynamic scheduling. As allocation time progresses, new dynamic scheduling tasks are added, the rolling window updates accordingly, and the remote consultation start time for APs is continuously adjusted. Within each rolling window, only the remote consultation start time for APs and related allocation information are adjusted. After completing the optimization within the current rolling window, the results are stored, and the rolling window is moved forward to continue optimizing the next rolling window.

[0037] In this example, each remote consultation is affected by uncertainties such as the arrival time of Access Points (APs), the randomness of the actual service duration, and the random arrival of WPs. When no WPs arrive, but the uncertainty of arrival or the randomness of service duration causes APs to arrive late or early, affecting subsequent remote consultations, optimization is initiated within the current rolling window. That is, in the case of no WPs arriving and random factors affecting APs (at least including the uncertainty of AP arrival and / or the randomness of AP service duration), the rolling strategy is followed: The consultation start time for affected APs patients in the static optimization scheduling scheme is updated within each rolling window; where, For APs patients The actual arrival time For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; Characterized as AP patients The assigned consultation start times follow a uniform distribution; Characterizing AP patients The actual consultation service duration follows a uniform distribution. For APs patients The allocation of consultation service time, It is a fixed constant; For APs patients The deviation was achieved. For APs patients Assigned remote consultation experts The buffer threshold; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation.

[0038] When a patient with multiple wards (WPs) arrives randomly, the start time of all affected remote consultations for all relevant remote consultations (APs) is immediately updated. Here, based on the current remote consultation room load, an appropriate remote consultation room and location are assigned to the randomly arriving WP. Specifically, according to the model: Assigning remote consultation rooms to WPs patients and consultation location In the formula, The total number of APs that submitted reservation requests. For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Arrival time.

[0039] Once a patient with WPs is added to the remote consultation queue, the remote consultation time for affected APs patients will be adjusted immediately. Specifically, according to the model: The consultation start time for affected APs patients in the static optimization scheduling scheme is updated; where, In cases where WP patients arrive randomly and AP patients have random factors, remote consultation experts are provided. Consultation room Corresponding APs patients Reassignment of consultation positions , To reallocate consultation positions Updated consultation start time for APs patients. For WPs patients The allocation of consultation service duration; A collection of WPs patients.

[0040] It should be noted that appropriate remote consultation rooms and locations can be allocated to WPs based on the minimum room load; that is, for each WP, the least congested consultation room is identified. And try to insert WPs patients in the place closest to their arrival location; or minimize the number of interruptions to APs when allocating WPs by adjusting the required allocation of APs. That is, for each AP patient, not only should the actual arrival time and actual service duration be considered to assess potential lateness, but also whether they are affected by WP insertion. During the adjustment process, if the AP update start time is too close to the end of the workday, try to reassign the WP patient to another teleconsultation room.

[0041] In this embodiment, the adjustment rate of APs ( The scheduling of remote consultations is evaluated using the resource utilization rate (RUR) of the remote consultation room. Specifically, In the formula, The APs adjustment rate is used to monitor scheduling changes caused by dynamic adjustments. A higher APs adjustment rate is desirable. This means that more APs can obtain updated information before arriving, which helps alleviate congestion in the remote consultation queue and improves the overall patient experience; The total number of APs that submitted reservation requests; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; To improve the resource utilization rate of remote consultation rooms, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location ,, This represents the total number of consultation locations. Consultation room Whether it is open or not.

[0042] In this embodiment, a static scheduling target model with a two-stage static scheduling framework is used. A feedforward control mechanism is introduced when allocating remote consultation rooms (WPs) to generate a statically optimized scheduling scheme. Based on this static scheduling, a dynamic scheduling model is constructed. On one hand, a global update is performed based on immediate patient arrival. When a WP arrives, the optimal consultation time slot is allocated to it, and the appointment information of the affected APs is updated synchronously (each update adjusts the allocation scheme for all subsequent patients in the same consultation room). On the other hand, when allocating WPs, uncertainties such as patient arrival behavior, actual service duration, and consultation expert delays are comprehensively considered. A rolling time-domain optimization based on a greedy strategy is adopted to allocate consultation rooms to WPs. and consultation location At the same time, the schedule for affected APs patients is updated to ensure the timeliness of consultation information and improve the responsiveness of remote consultations.

[0043] like Figure 5 As shown, this embodiment also provides a real-time scheduling system for remote consultations based on uncertainty, which dynamically schedules remote consultations using any of the aforementioned real-time scheduling methods for remote consultations based on uncertainty. The system includes: The static scheduling unit 501 is configured to generate a preliminary static scheduling scheme for remote consultation based on the appointment requests submitted by APs of the basic hospital under a deterministic environment without considering random factors; and to optimize the allocation of APs patients in the preliminary static scheduling scheme based on the second static allocation constraint under the influence of random factors, so as to obtain a static optimized scheduling scheme for remote consultation. The dynamic scheduling unit 502 is configured to assign remote consultation sequence positions to WPs patients and dynamically update the allocation of affected APs patients in the static optimized scheduling scheme, thereby generating a dynamic real-time scheduling scheme for remote consultations.

[0044] The uncertainty-based remote consultation real-time scheduling system provided in this embodiment can implement the steps and processes of the uncertainty-based remote consultation real-time scheduling method in any of the above embodiments and achieve the same technical effect, which will not be described in detail here.

[0045] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature.

[0046] In this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0047] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A real-time scheduling method for remote consultations based on uncertainty, characterized in that, include: In a deterministic environment that does not consider random factors, a static preliminary scheduling scheme for remote consultation is generated based on the appointment requests submitted by APs of basic hospitals and the first static allocation constraint. Under the influence of random factors, based on the second static allocation constraint, the allocation of APs patients in the static preliminary scheduling scheme is optimized to obtain the static optimized scheduling scheme for remote consultation. Assign remote consultation sequence positions to WPs patients and dynamically update the allocation of affected APs patients in the static optimization scheduling scheme to generate a dynamic real-time scheduling scheme for remote consultations.

2. The method according to claim 1, characterized in that, By passing the first static allocation constraint: Generate a static preliminary scheduling plan for remote consultation; In the formula, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; This refers to the scheduling cycle duration in the static preliminary scheduling scheme; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients The allocation of consultation service time, For APs patients Assigned remote consultation experts The buffer threshold.

3. The method according to claim 1, characterized in that, By using the second static allocation constraint: The allocation of APs patients in the static preliminary scheduling plan was optimized by using preset performance indicators under the influence of random factors. In the formula, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , hour, This represents the APs patients in the static preliminary scheduling scheme. Corresponding remote consultation experts were assigned. Consultation room and the location of the first consultation; For AP patients in the first consultation position in the static preliminary scheduling plan Waiting time; For APs patients The unpunctual arrival time The total number of APs that submitted reservation requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as WPs patient Corresponding remote consultation experts were assigned. Consultation room and consultation location , This refers to the number of unappointed patients who arrive randomly on the day of the remote consultation.

4. The method according to claim 1, characterized in that, Under the influence of random factors, a greedy scheduling method is used to assign remote consultation order positions to WPs patients, and the allocation of affected APs patients in the static optimization scheduling scheme is dynamically updated to generate a dynamic real-time scheduling scheme for remote consultation.

5. The method according to claim 4, characterized in that, Under the influence of random arrival of WPs patients, the remote consultation room with the shortest waiting time and closest availability in the statically optimized scheduling scheme is selected and assigned to WPs patients based on a greedy scheduling strategy. By constructing a global update model, the allocation of affected APs patients in the static optimization scheduling scheme is updated under the global update allocation constraint, thereby generating a dynamic real-time scheduling scheme for remote consultation scheduling.

6. The method according to claim 5, characterized in that, The allocation of affected APs in the static optimization scheduling scheme is updated according to global update allocation constraints; wherein, the global update allocation constraints include: APs update allocation constraints and WPs update allocation constraints. The APs update assignment constraint is: In the formula, Indicated as APs patients Corresponding remote consultation experts were assigned. Consultation room and consultation location , The total number of APs who submitted appointment requests. The total number of experts providing remote consultations. This refers to the total number of consultation rooms. This represents the total number of consultation locations. Indicated as APs patients Assigning remote consultation experts Consultation room Consultation location ; Consultation room Whether it is open or not, For APs patients The allocation of consultation service time; This refers to the scheduling cycle duration in the static preliminary scheduling scheme. Consultation room The time limit during consultations; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; hour, Characterizing remote consultation experts in the static preliminary scheduling scheme Consultation room and the first consultation location corresponding to APs patients The start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The end time of the consultation; For APs patients Assigned remote consultation experts The buffer threshold; WPs update assignment constraints: In the formula, For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts were assigned. Consultation room Consultation location ; For WPs patients The allocation of consultation service time; The number of unappointed patients who arrive randomly on the day of the remote consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Corresponding remote consultation experts were assigned. Consultation room Consultation location ; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation.

7. The method according to claim 4, characterized in that, Based on a greedy strategy, the rolling time-domain optimization freezes the remote consultation scheduling of the current rolling window in the static optimization scheduling scheme, and follows the formula: The remote consultation scheduling in the static optimization scheduling scheme is sequentially and continuously optimized to generate a dynamic real-time scheduling scheme for remote consultation scheduling; In the formula, Remote consultation experts in static optimization scheduling scheme Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For APs patients The arrival deviation, For APs patients The actual consultation service duration For APs patients The allocation of consultation service time, For WPs patients The allocation of consultation service time; The rolling step size for the rolling time-domain optimization strategy.

8. The method according to claim 7, characterized in that, In the case of no WPs patients arriving and APs patients having random factors, where APs patients have at least the uncertainty of APs patients’ arrival and / or the randomness of APs patients’ service duration; According to the rolling strategy: The consultation start time for affected APs in the static optimization scheduling scheme is updated within each rolling window; In the formula, For APs patients The actual arrival time For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; Characterized as AP patients The assigned consultation start times follow a uniform distribution; Characterizing AP patients The actual consultation service duration follows a uniform distribution. For APs patients The allocation of consultation service time, It is a fixed constant; For APs patients The deviation was achieved. For APs patients Assigned remote consultation experts The buffer threshold; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation.

9. The method according to claim 7, characterized in that, In the case of random arrival of WPs patients and random factors of APs patients, wherein the random factors of APs patients include at least the uncertainty of APs patients' arrival and / or the randomness of APs patients' service duration; According to the model: Assigning remote consultation rooms to WPs patients and consultation location In the formula, The total number of APs that submitted reservation requests. For remote consultation experts in the static preliminary scheduling plan Consultation room Consultation location Corresponding APs patients The start time of the consultation; For remote consultation experts Consultation room Consultation location Corresponding APs patients Optimize the start time of the consultation; For WPs patients Arrival time; And according to the model: Update the consultation start time for affected APs patients in the static optimization scheduling scheme; In the formula, In cases where WP patients arrive randomly and AP patients have random factors, remote consultation experts are provided. Consultation room Corresponding APs patients Reassignment of consultation positions , To reallocate consultation positions Updated consultation start time for APs patients. For WPs patients The allocation of consultation service duration; A collection of WPs patients.

10. A real-time scheduling system for remote consultation based on uncertainty, characterized in that, The system employs any one of the uncertainty-based real-time scheduling methods of claims 1-9 to dynamically schedule remote consultations, and comprises: The static scheduling unit is configured to generate a preliminary static scheduling scheme for remote consultation based on the appointment requests submitted by APs of the basic hospital under a deterministic environment that does not consider random factors; and to optimize the allocation of APs patients in the preliminary static scheduling scheme based on the second static allocation constraint under the influence of random factors, so as to obtain a static optimized scheduling scheme for remote consultation. The dynamic scheduling unit is configured to assign remote consultation sequence positions to WPs patients and dynamically update the allocation of affected APs patients in the static optimized scheduling scheme, generating a dynamic real-time scheduling scheme for remote consultations.

Images