Physical examination scheduling and load balancing method oriented to regional resource collaboration

Abstract

The invention discloses a physical examination scheduling and load balancing method oriented to regional resource collaboration, and relates to the technical field of regional resource scheduling, and the method comprises the steps: capturing demand information through physical examination records, building a priority standard through physical examination demand urgency indexes, preferentially scheduling high urgency demands, and avoiding health risks and physical examination overdue problems from the source. And the medical service pertinence is improved. According to the method, a regional resource collaborative topology network is constructed, collaborative load edges are identified through collaborative load circulation coefficients, resource scheduling information islands are broken, all regional resources and demand association are integrated, overall resource planning is achieved, and collaborative efficiency and transparency are improved. Meanwhile, a risk link is positioned through a series load conduction link, and the load is shunted by combining an alternative path, so that the breakthrough of load balancing from single-point management to link optimization is realized, chain overload is avoided, full-region resource dynamic balancing is achieved, and the stability and anti-risk capability of a scheduling system are improved.

Description

Technical Field

[0001] This invention relates to the field of regional resource scheduling technology, and more specifically, to a system-wide scheduling and load balancing method for regional resource collaboration. Background Technology

[0002] With increasing health awareness, the demand for physical examinations continues to grow. However, traditional scheduling models for physical examinations lack a scientific mechanism for determining the urgency of demand. Existing scheduling relies heavily on manual sorting or simple appointment time priority, failing to effectively differentiate the health risks of different physical examination needs. It also fails to systematically consider the time window constraints and scale characteristics of the demand, resulting in delays for some high-priority physical examination needs due to a lack of clear identification. This increases health risks and reduces the targetedness and rationality of medical services, making it difficult to adapt to diverse and differentiated physical examination needs.

[0003] Currently, health checkup resources within the region are scattered, and there is a lack of efficient collaboration mechanisms among various checkup institutions, specialized medical equipment, and medical teams, resulting in significant information silos. Existing scheduling models are mostly limited to the allocation of resources from a single institution or in a localized area, failing to integrate the relationship between resources and demand across the entire region, and also unable to quantify the degree of synergy and adaptability between nodes. This makes the matching of resource supply and demand lack a scientific basis, often resulting in a situation where some resources are idle while others are overloaded, leading to low resource coordination efficiency and making it difficult to achieve the coordinated utilization of health checkup resources across the entire region.

[0004] Traditional load balancing solutions in the field of health checkups often focus on localized control of individual overloaded nodes, lacking a holistic analysis of load propagation relationships. Existing methods fail to identify load flow links between resource nodes, only temporarily allocating resources to isolated overloaded nodes. They fail to trace the root cause of load backlogs or predict the cascading overload risks caused by load propagation. This results in short-lived load balancing effects, susceptibility to risk transfer, inability to achieve dynamic resource balancing across the entire area, and insufficient stability and resilience of the overall scheduling system, making it difficult to cope with complex and ever-changing health checkup demands and resource states. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide a system for scheduling and load balancing of regional resources.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for health check scheduling and load balancing oriented towards regional resource collaboration includes the following steps: Step 1: Identify all physical examination resource nodes and demand nodes within the region. Whenever a new demand node appears, generate physical examination records for each demand node and obtain the physical examination demand urgency index for each demand node. When the physical examination demand urgency index of a demand node is higher than the physical examination demand urgency threshold, mark the corresponding demand node as the physical examination scheduling target node. Step 2: Construct a regional resource collaborative topology network based on all physical examination resource nodes and all physical examination scheduling target nodes, obtain the collaborative load transfer coefficient of each network edge in the regional resource collaborative topology network, and mark the corresponding network edge as a collaborative load edge when the collaborative load transfer coefficient is higher than the collaborative load transfer threshold. Connect multiple collaborative load edges in sequence according to the direction of the collaborative load edge to form multiple resource load transmission links. Step 3: Identify the risky links in the resource load transmission links, and divert part of the load on each risky link to an alternative path to complete the load diversion for the health check scheduling.

[0007] Furthermore, the node health check record includes the demand node ID, the remaining number of people waiting for a health check, the type of health check appointment, and the appointment time window.

[0008] Further, the steps to obtain the urgency index of physical examination demand at a demand node are as follows: Select a demand node, obtain the risk index G1 of the physical examination items, the remaining time window G2, and the proportion of the scale of physical examinations to be conducted at that demand node, using the formula: Demand node = G1*r1 + G2*r2 + G3*r3; r1, r2, and r3 are all weighting coefficients.

[0009] Further, the construction steps of the regional resource collaborative topology network are as follows: determine all physical examination resource nodes and all physical examination scheduling target nodes contained in the region, take all the interconnected physical examination resource nodes and physical examination scheduling target nodes as network nodes, and take the collaborative relationship between network nodes as network edges to form a regional resource collaborative topology network.

[0010] Furthermore, the steps for obtaining the collaborative load balancing coefficient of the network edge in the patient referral scenario are as follows: Mark the remaining number of patients awaiting physical examination at the target node for physical examination scheduling in the network edge as... Mark the remaining available appointments for physical examination institutions or professional medical equipment nodes in the network edge as... Through formula Calculate the resource supply fit degree Get the overlap duration of the appointment time windows. Obtain the total duration of the appointment time window for the target node of the physical examination scheduling. Through formula Calculate the time window matching degree Obtain the actual traffic distance between two nodes. Obtain the average traffic distance between demand nodes and physical examination resource nodes within the region. Through formula Calculate the distance fit Through formula The collaborative load flow coefficient of the network edge in the patient referral scenario is calculated. c1, c2, and c3 are all weighting coefficients.

[0011] Further, the process for determining the risk link in the physical examination scheduling is as follows: Extract the collaborative load transfer coefficients of all collaborative load edges in a single resource load transmission link, denoted as K1, K2, ..., Kn; calculate the comprehensive load deviation using the average absolute deviation. Set a comprehensive load deviation threshold. When the comprehensive load deviation is higher than the comprehensive load deviation threshold, mark the resource load transmission link as a health check scheduling risk link.

[0012] Further, the steps for obtaining the collaborative load transfer coefficient of the network edge are as follows: determine a collaborative scenario of the network edge (the collaborative scenario includes resource sharing scenario, medical support scenario, and patient referral scenario), and determine the collaborative load transfer coefficient of the network edge based on the collaborative scenario of the network edge.

[0013] Furthermore, the steps for obtaining the collaborative load balancing coefficient of the network edge in a resource-sharing scenario are as follows: Obtain the real-time load rate of each node at both ends of the network edge, and label them as Lx and Ly respectively, using the formula... Calculate the load matching degree Obtain the actual process connection time between two nodes. And obtain the benchmark connection time for the coordination of similar resources within the region. Through formula Calculate the process integration efficiency Obtain the duration of overlap of load peaks between two nodes. Through formula Peak synergy was calculated Through formula The collaborative load balancing coefficient of the network edge in the resource-sharing scenario is calculated. a1, a2, and a3 are all weighting coefficients.

[0014] Furthermore, the steps for obtaining the collaborative load balancing coefficient of the network edge in a healthcare support scenario are: obtaining the actual support response time of the two nodes at both ends of the network edge. and obtain the standard response time for similar medical support within the region. Through formula Calculate the support response efficiency Obtain the load conversion ratio between the two nodes. And then through the formula Get load capacity adaptability Obtain the number of physical examination resource nodes simultaneously served by the medical team node. Through formula The support dispersion was calculated. Through formula The collaborative load balancing coefficient of the network edge in the medical support scenario was calculated. b1 and b2 are both weighting coefficients.

[0015] Furthermore, the steps for obtaining the collaborative load balancing coefficient of the network edge in the patient referral scenario are as follows: Mark the remaining number of patients awaiting physical examination at the target node for physical examination scheduling in the network edge as... Mark the remaining available appointments for physical examination institutions or professional medical equipment nodes in the network edge as... Through formula Calculate the resource supply fit degree Get the overlap duration of the appointment time windows. Obtain the total duration of the appointment time window for the target node of the physical examination scheduling. Through formula Calculate the time window matching degree Obtain the actual traffic distance between two nodes. Obtain the average traffic distance between demand nodes and physical examination resource nodes within the region. Through formula Calculate the distance fit Through formula The collaborative load flow coefficient of the network edge in the patient referral scenario is calculated. c1, c2, and c3 are all weighting coefficients.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention's method captures core demand information comprehensively through physical examination records, establishes an objective priority standard based on the urgency index of the examination demand, and ensures that high-urgency demands are prioritized for scheduling. This avoids health risks and overdue examinations from the outset, allowing scheduling resources to be precisely tilted towards the most needed scenarios, improving the targeting of medical services. A regional resource collaborative topology network is constructed based on resource nodes and scheduling target nodes. The design of identifying collaborative load edges through a collaborative load flow coefficient breaks the information silo dilemma in physical examination resource scheduling. By integrating the resources and demands of the entire region through the topology network, and quantifying the collaborative adaptability between nodes using the flow coefficient, the method accurately identifies network edges for efficient collaboration, facilitating subsequent load propagation links. The analysis provides a scientific basis, upgrading resource coordination from fragmented scheduling to global planning, improving the transparency and efficiency of resource association. By connecting collaborative load edges to form resource load transmission links, locating risk links, and implementing load diversion, a breakthrough in load balancing for health checkups has been achieved, moving from single-point governance to link optimization. By tracing the root cause of load transmission through link analysis, the core bottlenecks of risk links are accurately located, and load is scientifically diverted by combining alternative paths. This not only fundamentally alleviates the pressure on risk links but also ensures that alternative paths have stable carrying capacity, avoiding cascading overload problems. It achieves dynamic balance and efficient utilization of health checkup resources across the entire region, improving the stability and risk resistance of the overall scheduling system. Attached Figure Description

[0017] Figure 1 A flowchart of a system for scheduling and load balancing based on regional resource collaboration; Figure 2 A flowchart for constructing a regional resource collaborative topology network. Detailed Implementation

[0018] Reference Figures 1-2 A method for health check scheduling and load balancing oriented towards regional resource collaboration includes the following steps: Step 1: Identify all health checkup resource nodes and demand nodes within the area (resource nodes include health checkup institutions (A1, A2, ..., An), professional medical equipment (such as CT scanners, color Doppler ultrasound machines, etc., B1, B2, ..., Bm), and medical teams (C1, C2, ..., Cp). Each health checkup institution, professional medical equipment, and medical team can be marked as a health checkup resource node). Whenever a new demand node appears, generate a node health checkup record for each demand node. The node health checkup record includes the demand node ID (community, enterprise / institution group, family, or individual can all represent a demand node; for example, if an enterprise / institution group undergoes health checkups together, then the enterprise / institution group is a demand node; if an individual undergoes health checkups, then the individual is a demand node), the remaining number of people waiting for health checkups, and health checkup appointment items. The system identifies the type of medical examination (e.g., pre-employment physical examination, chronic disease follow-up, cancer screening) and the appointment time window (i.e., the time range within which the medical examination can be performed as explicitly required by the demand node). It obtains the urgency index of the medical examination demand for each demand node. When the urgency index of a demand node exceeds the urgency threshold (the urgency threshold is set based on historical risk data statistics; for example, analyzing the correlation between the urgency index of medical examination demand in the past 6-12 months and the overdue risk / health risk within the region; e.g., when the urgency index is ≥0.7, more than 50% of nodes experience overdue medical examinations, or chronic disease follow-up nodes experience fluctuations in condition due to overdue examinations), the urgency index corresponding to a risk incidence rate reaching a preset threshold (e.g., 30%) is set as the urgency threshold for medical examination demand), and the corresponding demand node is marked as the target node for medical examination scheduling.

[0019] Steps to obtain the urgency index of physical examination demand at a demand node: Select a demand node and obtain the risk index G1, time window remaining time G2, and the proportion of pending physical examinations G3 for that demand node, using the formula: Demand node = G1*r1 + G2*r2 + G3*r3; where r1, r2, and r3 are weighting coefficients, and r1 + r2 + r3 = 1. Risk index G1: Assigning values ​​to different types of physical examination appointments (based on medical risk levels), for example: High-risk items (chronic disease follow-up, cancer screening): G1 = 1.0; Medium-risk items (routine physical examinations for middle-aged and elderly people): G1 = 0.7; Low-risk items (pre-employment / routine physical examinations): G1 = 0.3. Time window remaining time G2: G2 = 1 - current remaining time of the appointment time window / total appointment time window time; when the remaining time is 0, G2 = 1. Proportion of pending physical examinations... The average daily volume of similar projects in the region is based on historical data from the past 6 months. Since the risk level of physical examination projects is directly related to health risks and is the core basis for judging the urgency of demand, it has the greatest impact on scheduling priority. The remaining time window determines that the risk of overdue demand nodes is second only to health risks. The scale of pending physical examinations mainly affects the pressure on resource load and has a lower impact on the urgency of demand itself. Therefore, the value of r1 can be 0.5, the value of r2 can be 0.3, and the value of r3 can be 0.2.

[0020] Step 2: Construct a regional resource collaboration topology network based on all physical examination resource nodes and all physical examination scheduling target nodes. Obtain the collaboration load transfer coefficient of each network edge in the regional resource collaboration topology network. When the collaboration load transfer coefficient is higher than the collaboration load transfer threshold, mark the corresponding network edge as a collaboration load edge. (The collaboration load transfer threshold for network edges in different collaboration scenarios is not the same. For example, the core risk of resource sharing scenarios is "overload transmission causing chain overload", so the threshold should not be too high. When the collaboration load transfer coefficient is ≥0.7, it indicates that the load matching degree and connection efficiency between resources are high enough, and the load is easily transmitted from one node to another. It needs to be marked as a collaboration load edge for key monitoring.) The core objective of the medical support scenario is to "quickly and effectively take over the load of resource nodes," so the threshold needs to be set relatively high. Only when the collaborative load turnover coefficient is ≥0.8 can it be said that the response efficiency and adaptability of medical support meet the standards, and it is a key edge that can effectively alleviate the resource load, and should be marked as a collaborative load edge. The core objective of the patient referral scenario is to "the referral task can be actually implemented," so the threshold is set appropriately. When the collaborative load turnover coefficient is ≥0.6, it means that the adaptability of resource supply, time window, and distance has met the basic referral conditions, and it is a feasible referral edge, and should be marked as a collaborative load edge. Multiple collaborative load edges are sequentially connected according to the direction of the collaborative load edge to form multiple resource load transmission links.

[0021] Step 3: Identify the risky links in the multiple resource load transmission links, and divert part of the load on each risky link to an alternative path to complete the load diversion for the health check scheduling.

[0022] To offload a portion of the load on each health checkup scheduling risk link to an alternative path, the following example illustrates the load offloading process: Determine the offloading targets and risk aversion coefficients. The health checkup scheduling risk link is as follows: Health checkup scheduling target node D2 → Professional medical equipment B2 → Health checkup institution A1 → Medical team C1. Calculate the workload rate of professional medical equipment B2, health checkup institution A1, and medical team C1 (Workload rate of professional medical equipment B2: 30 appointments at D2 + other pre-booked appointments). Assuming a total of 17 people, the rated daily capacity is 20 people → 17 / 20 = 85%; The workload rate of medical examination institution A1: actual occupancy = 17 people from B2 + 33 people from other equipment = 50 people, rated daily capacity = 80 people → 50 / 80 = 62.5%; The workload rate of medical team C1: 17 people from B2 + 12 people from B1 = 29 people, rated daily workload = 30 people → 29 / 30 ≈ 96.7% (judged as "saturated"), then the core... Bottleneck: B2 load rate 85%, C1 workload saturated; Risk avoidance coefficient set: 0.3 (based on historical data, this coefficient can reduce the load of the risk link to a safe range); Load to be diverted: 30 chronic disease follow-up tasks in D2; Screening alternative paths: Select alternative paths according to the standard of collaborative load transfer coefficient ≥ 0.6 + not in the risk link of physical examination scheduling; Alternative path: D2 → B5 (backup color Doppler ultrasound machine in A2 institution) → A2 → C4 (idle medical team); Alternative Path collaborative load edge verification: D2→B5 (patient referral edge, collaborative load transfer coefficient 0.72≥0.6); B5→A2 (resource sharing edge, collaborative load transfer coefficient 0.8≥0.7); A2→C4 (medical support edge, collaborative load transfer coefficient 0.85≥0.8); Diversion ratio = risk avoidance coefficient × 100% = 0.3 × 100% = 30%; Diversion load = total number of people waiting for physical examination in D2 × diversion ratio = 30 × 30% = 10 people.

[0023] The process of obtaining the risk aversion coefficient: The region's historical maximum comprehensive load deviation is the maximum ΔZ value of all health check scheduling risk links in the past 6 months.

[0024] The process for determining the risk link in the physical examination scheduling is as follows: Extract the collaborative load transfer coefficients of all collaborative load edges in a single resource load transmission link, denoted as K1, K2, ..., Kn (n is the number of collaborative load edges included in the resource load transmission link); calculate the comprehensive load deviation using the average absolute deviation. Set a comprehensive load deviation threshold (the comprehensive load deviation threshold is set based on historical risk data, such as statistically analyzing resource load transmission links that have experienced overload / delay / patient backlog in the past 6-12 months, extracting the comprehensive load deviation corresponding to these links, and taking their average or median as the threshold benchmark). When the comprehensive load deviation is higher than the comprehensive load deviation threshold, the resource load transmission link is marked as a physical examination scheduling risk link (the comprehensive load deviation is the core indicator for measuring the stability of load transmission within the link (the larger the deviation, the more likely the load transmission is to fluctuate / overload risk)).

[0025] Multiple collaborative load edges are sequentially connected based on their directions to form multiple resource load transmission links. Example: Physical examination scheduling target nodes: D1 (group physical examination, 50 people), D2 (chronic disease follow-up, 30 people); Physical examination resource nodes: Institution A1 (including B1=CT scanner, B2=color Doppler ultrasound machine), Institution A2 (including B3=ECG machine); Medical team C1 (adapted to B1, B2), C2 (adapted to B3); Collaborative load edges (already determined): Patient referral edges: D1→A1, D1→B1, D1→B2; D2→A1, D2→B2, D2→B3; Resource sharing edges: B1→B2, A1→B1, A1→B2, A2→B3; Medical support edges: C1→B1, C1→B2, C2→B3; The direction of the collaborative load edges is the association direction between nodes (e.g., D1→B2, C1→B2, C2→B3). 1 indicates the path from D1 to B1. The paths must be connected according to the rule that the endpoint of the previous edge equals the starting point of the next edge. Here is a specific example of a link: Link 1: Load propagation link for the group physical examination in D1; Path: D1 (Patient referral edge) → B1 (Resource sharing edge) → B2 (Medical support edge) → C1; Corresponding collaborative load edges: D1 → B1 (Patient referral edge, coefficient 0.88 ≥ 0.6): The group physical examination needs of D1 are diverted to B1 (CT scanner), and the load is propagated to B1; B1 → B2 (Resource sharing edge, coefficient 1.0 ≥ 0.7): The physical examination patient process in B1 is connected to B2 (color Doppler ultrasound machine), and the load of B1 is propagated to B2; B2 → C1 (Medical support edge, coefficient 0.95 ≥ 0.8): The load of B2 is transformed into the workload of the medical team in C1, and the load of B2 is propagated to C1. Load propagation logic: The demand for physical examinations for 50 people in D1 → Increased load on B1 (CT scanner) → Passive increase in load on B2 (color Doppler ultrasound) → Saturation of workload for the C1 medical team. Link 2: Load propagation link for chronic disease follow-up in D2; Serial path: D2 (patient referral edge) → B2 (resource sharing edge) → A1 (medical support edge) → C1; Corresponding collaborative load edges: D2 → B2 (patient referral edge, coefficient 0.80 ≥ 0.6): The follow-up demand of D2 is diverted to B2 (color Doppler ultrasound), and the load is propagated to B2; B2 → A1 (resource sharing edge, coefficient 0.85 ≥ 0.7): The load of B2 is propagated to its affiliated institution A1, and the overall load of A1 increases; A1 → C1 (medical support edge, coefficient 0.92 ≥ 0.8): The load of A1 is transformed into the workload of C1, and the load of A1 is propagated to C1. Load propagation logic: The demand for follow-up examinations for 30 people in D2 → Increased load on B2 (color Doppler ultrasound machine) → Increased overall load on A1 institution → Further saturation of workload for C1 medical team.Link 3: Cross-institutional load transmission link; Serial path: D2 (patient referral edge) → B3 (resource sharing edge) → A2 (medical support edge) → C2. Corresponding collaborative load edges: D2 → B3 (patient referral edge, coefficient 0.75 ≥ 0.6): The ECG demand of D2 is diverted to B3 (ECG machine), and the load is transmitted to B3; B3 → A2 (resource sharing edge, coefficient 0.8 ≥ 0.7): The load of B3 is transmitted to its affiliated institution A2, and the overall load of A2 increases; A2 → C2 (medical support edge, coefficient 0.9 ≥ 0.8): The load of A2 is transformed into the workload of C2, and the load of A2 is transmitted to C2. Load transmission logic: ECG demand of D2 → B3 (ECG machine) load increases → overall load of institution A2 increases → workload of C2 medical team reaches saturation.

[0026] The steps to obtain the collaborative load transfer coefficient of the network edge are as follows: determine a collaborative scenario of the network edge (the collaborative scenario includes resource sharing scenario, medical support scenario, and patient referral scenario), and determine the collaborative load transfer coefficient of the network edge based on the collaborative scenario of the network edge.

[0027] Network edge in resource sharing scenarios: A network edge connecting two medical examination institutions, two professional medical devices, or one medical examination institution and one professional medical device. The core collaborative relationship is resource reuse, complementarity, or process connection. The load transfer logic is that the load change of a resource node (such as occupation or release) will be directly transmitted to the associated resource node, such as institution A1 and institution A2, B1 CT machine and B2 color Doppler ultrasound machine, institution A1 and B3 electrocardiograph. Example: institution A1 (a) → institution A2 (b) (cross-institutional equipment sharing, when the CT machine of A1 is overloaded, the backup CT machine of A2 can be scheduled); B1 CT machine (a) → B2 color Doppler ultrasound machine (b) (same-institutional process connection, after the CT examination, the patient needs to be transferred to the color Doppler ultrasound machine).

[0028] Network edge in healthcare support scenarios: This network edge connects healthcare team nodes with medical examination institutions or specialized medical equipment. The core collaborative relationship is that healthcare personnel provide professional operation and treatment support to resource nodes. The load transfer logic is that the load of resource nodes is transformed into the workload of healthcare teams. Changes in the load of resource nodes directly affect the busyness of healthcare teams. For example, C1 healthcare team (a) → B1 CT machine (b) (C1 team operates B1 CT machine exclusively. An increase in B1's appointment volume will lead to an increase in C1's workload); C2 healthcare team (a) → A2 institution (b) (C2 team provides multi-equipment support to A2 institution. The overall saturation of A2's medical examination volume will be transmitted to C2's load).

[0029] The network edge in the patient referral scenario connects the target node for physical examination scheduling with the physical examination institution or specialized medical equipment. The core collaborative relationship is that the physical examination tasks of the demand node are distributed and undertaken by the resource node. The load transfer logic is that the number of people waiting for physical examinations and the appointment demand of the demand node will directly translate into the load of the resource node. Changes in the scale of demand determine the load pressure of the resource node. For example, D1 group physical examination node (a) → A1 institution (b) (the physical examination task of 50 people in D1 is undertaken by A1 institution, and the number of people waiting for physical examinations in D1 directly translates into the load of A1); D2 chronic disease follow-up node (a) → B2 color Doppler ultrasound machine (b) (the follow-up demand of 30 people in D2 requires the use of B2 color Doppler ultrasound machine, and the appointment volume of D2 will occupy the available appointment time of B2, resulting in an increase in the load of B2).

[0030] Steps to obtain the collaborative load balancing coefficient of the network edge in a resource-sharing scenario: Obtain the real-time load rate of each node at both ends of the network edge (real-time load rate = actual node occupancy (e.g., current appointments for the device, current number of physical examinations at the institution) / rated node load (rated daily capacity of the device, rated daily capacity of the institution)), and label them as Lx and Ly respectively, using the formula... Calculate the load matching degree Obtain the actual process connection time between two nodes. (e.g., the average time it takes for a patient to move from node x to node y, and the average response time for cross-institutional resource scheduling), and obtain the benchmark connection time for the coordination of similar resources within the region. (Based on the average connection time of historical data over the past m months), using the formula Calculate the process integration efficiency Obtain the duration of overlap of load peaks between two nodes. (Unit: hours / day, i.e., the overlap length of time within a day when the load rate of each is ≥80%), using the formula. Peak synergy was calculated Through formula The collaborative load balancing coefficient of the network edge in the resource-sharing scenario is calculated. a1, a2, and a3 are all weighting coefficients, a1 + a2 + a3 = 1. Since load matching reflects the load balancing degree between two resource nodes, and process connection efficiency reflects the flow speed between resources, both are equally important and therefore have the same weight. Compared to load balancing and efficient flow, which directly affect load propagation, peak overlap is an indirect factor influencing load propagation over time, and its effect on immediate load propagation is weaker than the former two. Therefore, the values ​​for a1, a2, and a3 can be 0.4, 0.4, and 0.2, respectively. Result calibration: When... When, take 1; when When the time is right, take 0.

[0031] Steps to obtain the collaborative load balancing coefficient of the network edge in a healthcare support scenario: Obtain the actual support response time of the two nodes at both ends of the network edge. (The average time taken for a medical team to reach another node within the most recent time interval t), and obtain the standard response time for similar medical support within the region. (Based on historical data statistics), through the formula Calculate the support response efficiency Obtain the load conversion ratio between the two nodes. (Load conversion ratio = current load of resource node b / current available workload of medical team a, current available workload of medical team a = rated workload of medical team - workload already undertaken), and then through the formula Get load capacity adaptability Obtain the number of physical examination resource nodes simultaneously served by the medical team node. (i.e., the number of network edges of the physical examination resource nodes associated with the medical team node), expressed by the formula The support dispersion was calculated. Through formula The collaborative load balancing coefficient of the network edge in the medical support scenario was calculated. b1 and b2 are both weighting coefficients, b1 + b2 = 1. Since support response efficiency is a core prerequisite for the effectiveness of medical support, load capacity adaptability ensures that the workload of medical staff matches the resource load, and support dispersion prevents medical staff from being distracted by too many resource nodes. However, this is the effect after timely response, and its priority is slightly lower than the core prerequisite of timely arrival. Therefore, the value of b1 can be 0.6, and the value of b2 can be 0.4; Result calibration: When When, take 1; when When the time is right, take 0.

[0032] Steps for obtaining the collaborative load balancing coefficient of the network edge in a patient referral scenario: Mark the remaining number of patients awaiting physical examination at the target node of the physical examination scheduling in the network edge as... Mark the remaining available appointments for physical examination institutions or professional medical equipment nodes in the network edge as... (The number of people who can be accommodated for the required medical examination items for the corresponding medical examination scheduling target node), through the formula Calculate the resource supply fit degree Get the overlap duration of the appointment time windows. (The overlap between the opening hours of the medical examination institution or specialized medical equipment node and the appointment time window of the medical examination scheduling target node), obtain the total appointment time window duration of the medical examination scheduling target node. Through formula Calculate the time window matching degree Obtain the actual traffic distance between two nodes. Obtain the average traffic distance between demand nodes and physical examination resource nodes within the region. (Based on historical data statistics), through the formula Calculate the distance fit Through formula The collaborative load flow coefficient of the network edge in the patient referral scenario is calculated. c1, c2, and c3 are all weighting coefficients, and c1 + c2 + c3 = 1. Resource supply matching reflects whether a resource node can handle the number of patients waiting for medical examinations at a demand node, which is the core foundation for referrals. Time window matching reflects whether the opening hours of a resource node match the appointment time requirements of a demand node, ensuring the timeliness of referrals (e.g., if a demand node requires weekend medical examinations, but the resource node is only open on weekdays, it cannot match). Its importance is second only to resource supply. Distance matching reflects the ease of transportation from the demand node to the resource node, ensuring the actual implementation of referrals (excessive distance will reduce patients' willingness to seek medical treatment). It is a necessary condition for referral implementation, along with time window matching, but its priority is slightly lower than resource supply. Therefore, its weight is the same as time window matching, so c1 can be 0.4, c2 can be 0.3, and c3 can be 0.3.

[0033] The steps for constructing a regional resource collaborative topology network are as follows: First, identify all physical examination resource nodes and all physical examination scheduling target nodes within the region. Then, define all interconnected physical examination resource nodes and scheduling target nodes as network nodes (i.e., select physical examination resource nodes and scheduling target nodes with direct / indirect collaborative relationships, including: 1. all physical examination scheduling target nodes; 2. physical examination resource nodes (institutions, equipment, medical teams) that have resource matching or service support relationships with these scheduling target nodes; 3. physical examination resource nodes that have sharing or support relationships with each other). Finally, define the collaborative relationships between network nodes as network edges (network edges represent the actual business collaboration logic between nodes, serving as bridges for load transfer from one node to another; each edge corresponds to a specific collaborative scenario, such as resource sharing, medical support, or patient referral scenarios), thus forming a regional resource collaborative topology network.

[0034] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0035] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the steps or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0036] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0037] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0038] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for health check scheduling and load balancing oriented towards regional resource collaboration, characterized in that, Includes the following steps: Step 1: Identify all physical examination resource nodes and demand nodes within the region. Whenever a new demand node appears, generate physical examination records for each demand node and obtain the physical examination demand urgency index for each demand node. When the physical examination demand urgency index of a demand node is higher than the physical examination demand urgency threshold, mark the corresponding demand node as the physical examination scheduling target node. Step 2: Construct a regional resource collaborative topology network based on all physical examination resource nodes and all physical examination scheduling target nodes, obtain the collaborative load transfer coefficient of each network edge in the regional resource collaborative topology network, and mark the corresponding network edge as a collaborative load edge when the collaborative load transfer coefficient is higher than the collaborative load transfer threshold. Connect multiple collaborative load edges in sequence according to the direction of the collaborative load edge to form multiple resource load transmission links. Step 3: Identify the risky links in the resource load transmission links, and divert part of the load on each risky link to an alternative path to complete the load diversion for the health check scheduling.

2. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, The node health check record includes the demand node ID, the remaining number of people waiting for health check, the type of health check appointment, and the appointment time window.

3. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, Steps to obtain the urgency index of physical examination demand for a demand node: Select a demand node and obtain the risk index G1 of the physical examination items, the remaining time window G2, and the proportion of the scale of physical examinations to be conducted for that demand node, using the formula: Demand node = G1*r1 + G2*r2 + G3*r3; r1, r2, and r3 are all weighting coefficients.

4. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, The steps for constructing a regional resource collaborative topology network are as follows: Identify all physical examination resource nodes and all physical examination scheduling target nodes within the region; designate all interconnected physical examination resource nodes and physical examination scheduling target nodes as network nodes; and designate the collaborative relationships between network nodes as network edges to form a regional resource collaborative topology network.

5. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, Steps for obtaining the collaborative load balancing coefficient of the network edge in a patient referral scenario: Mark the remaining number of patients awaiting physical examination at the target node of the physical examination scheduling in the network edge as... Mark the remaining available appointments for physical examination institutions or professional medical equipment nodes in the network edge as... Through formula Calculate the resource supply fit degree Get the overlap duration of the appointment time windows. Obtain the total duration of the appointment time window for the target node of the physical examination scheduling. Through formula Calculate the time window matching degree Obtain the actual traffic distance between two nodes. Obtain the average traffic distance between demand nodes and physical examination resource nodes within the region. Through formula Calculate the distance fit Through formula The collaborative load flow coefficient of the network edge in the patient referral scenario is calculated. c1, c2, and c3 are all weighting coefficients.

6. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, The process for determining the risk link in the physical examination scheduling is as follows: Extract the collaborative load transfer coefficients of all collaborative load edges in a single resource load transmission link, denoted as K1, K2, ..., Kn; calculate the comprehensive load deviation using the average absolute deviation. Set a comprehensive load deviation threshold. When the comprehensive load deviation is higher than the comprehensive load deviation threshold, mark the resource load transmission link as a health check scheduling risk link.

7. The method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, The steps to obtain the collaborative load transfer coefficient of the network edge are as follows: determine a collaborative scenario of the network edge (the collaborative scenario includes resource sharing scenario, medical support scenario, and patient referral scenario), and determine the collaborative load transfer coefficient of the network edge based on the collaborative scenario of the network edge.

8. A method for system-wide resource coordination-oriented scheduling and load balancing according to claim 1, characterized in that, Steps to obtain the collaborative load balancing coefficient of the network edge in a resource-sharing scenario: Obtain the real-time load rate of each node at both ends of the network edge and label them as Lx and Ly respectively. Then, use the formula... Calculate the load matching degree Obtain the actual process connection time between two nodes. And obtain the benchmark connection time for the coordination of similar resources within the region. Through formula Calculate the process integration efficiency Obtain the duration of overlap of load peaks between two nodes. Through formula Peak synergy was calculated Through formula The collaborative load balancing coefficient of the network edge in the resource-sharing scenario is calculated. a1, a2, and a3 are all weighting coefficients.

9. A method for system check scheduling and load balancing oriented towards regional resource collaboration according to claim 1, characterized in that, Steps to obtain the collaborative load balancing coefficient of the network edge in a healthcare support scenario: Obtain the actual support response time of the two nodes at both ends of the network edge. and obtain the standard response time for similar medical support within the region. Through formula Calculate the support response efficiency Obtain the load conversion ratio between the two nodes. And then through the formula Get load capacity adaptability Obtain the number of physical examination resource nodes simultaneously served by the medical team node. Through formula The support dispersion was calculated. Through formula The collaborative load balancing coefficient of the network edge in the medical support scenario was calculated. b1 and b2 are both weighting coefficients.

10. A method for system-wide resource coordination-oriented scheduling and load balancing according to claim 1, characterized in that, Steps for obtaining the collaborative load balancing coefficient of the network edge in a patient referral scenario: Mark the remaining number of patients awaiting physical examination at the target node of the physical examination scheduling in the network edge as... Mark the remaining available appointments for physical examination institutions or professional medical equipment nodes in the network edge as... Through formula Calculate the resource supply fit degree Get the overlap duration of the appointment time windows. Obtain the total duration of the appointment time window for the target node of the physical examination scheduling. Through formula Calculate the time window matching degree Obtain the actual traffic distance between two nodes. Obtain the average traffic distance between demand nodes and physical examination resource nodes within the region. Through formula Calculate the distance fit Through formula The collaborative load flow coefficient of the network edge in the patient referral scenario is calculated. c1, c2, and c3 are all weighting coefficients.

Images