Efficiency-driven fluid path planning method and system based on digital twinning

Abstract

This invention discloses a performance-driven fluid path planning method and system based on digital twins. The system includes a digital twin construction module, an equipment status perception module, a gravitational field calculation module, and a path planning module. The method includes: constructing a digital twin based on a hospital building information model, abstracting the hospital space as a weighted directed dynamic graph, where nodes correspond to medical resource equipment or examination nodes, edge weights represent path travel costs, and the equipment utilization status of nodes dynamically affects the travel costs of edges; acquiring the real-time operating status of each device and patient queue information, and calculating the expected idling time of the devices accordingly; determining the gravitational value of the devices based on the expected idling time, with the gravitational value increasing as the expected idling time increases; calculating the total path cost to each candidate device for a patient with a planned path; and generating dynamic path guidance for the patient based on the total path cost, where the dynamic path guidance is an instruction pointing to the next target node. The system includes a digital twin construction module, an equipment status perception module, a gravitational field calculation module, and a path planning module.

Description

Technical Field

[0001] This invention relates to the fields of medical informatics and intelligent scheduling optimization, specifically to a performance-driven fluid path planning method and system based on digital twins. Background Technology

[0002] In the operation and management of large and complex public buildings, especially modern comprehensive medical institutions, the efficient flow of resources and the scientific planning of spatial paths are core issues. Current hospital path planning technologies mostly focus on solving for the shortest path in a geometric sense, ignoring the real-time load status of resources at the end of the path. This results in patients still having to wait in long queues after arriving at the shortest path, while remote equipment remains idle, creating a serious resource mismatch. In addition, existing systems lack the ability to quickly adapt and reconfigure when faced with dynamic disturbances such as sudden equipment failures or surges in patient flow, often relying on manual intervention and making it difficult to maintain the overall operational robustness of the system.

[0003] In view of the technical problems existing in the above-mentioned technologies, such as resource mismatch, poor anti-disturbance capability and system performance imbalance, there is an urgent need to provide a control scheme that can transform spatial pathfinding into dynamic resource scheduling. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a performance-driven fluid path planning method based on digital twins, comprising: A digital twin based on the hospital building information model is constructed, and the hospital space is abstracted as a weighted directed dynamic graph. The nodes in the graph correspond to medical resources and equipment or examination nodes, the weight of the edge represents the path travel cost, and the equipment utilization status of the node dynamically affects the travel cost of the edge. The system acquires the real-time operating status of each device and patient queue information, and calculates the expected idle time of the devices accordingly. The gravitational force value of the device is determined based on the expected idling time, and the gravitational force value increases with the increase of the expected idling time; For patients with a planned path, the total path cost to each candidate device is calculated. The total path cost includes physical distance cost, congestion time cost, and a negative gravity penalty term. This gravity penalty term reduces the total path cost as the gravity value of the target device increases. Dynamic path guidance is generated for patients based on the total path cost. The dynamic path guidance is an instruction pointing to the next target node.

[0005] In the preferred embodiment, when the device completes the current examination and the next scheduled patient has not yet arrived, resulting in an increase in the expected idle time, the gravity value of the device is increased; the gravity value is calculated through a preset gravitational field potential energy function.

[0006] In the preferred solution, the total path cost is calculated according to the following formula: Total path cost = Physical distance cost + Congestion time cost - k × Target device gravity value; Where k is a positive adjustment coefficient.

[0007] The preferred solution also includes: A dynamic game optimizer for both medical staff and patients is constructed, defining the patient utility function and the system utility function. The patient utility function is negatively correlated with walking distance, number of floor changes, and queuing time, while the system utility function is positively correlated with equipment utilization rate and total table turnover rate. By iteratively solving the Pareto optimal solution through multi-stage dynamic game theory, the patient pathway decision-making can improve system utility without significantly reducing patient utility. The dynamic path guide is the Nash equilibrium path output by the game optimizer.

[0008] In the preferred scheme, the game optimizer makes decisions to direct patients to similar devices that are physically farther away but have significantly shorter waiting times, thereby increasing the utilization rate of idle devices without increasing the overall time and physical exertion of patients, and achieving a Nash equilibrium at the system level.

[0009] The preferred scheme also includes a burst disturbance compensation step based on meta-learning: The system accesses the IoT operating parameters of medical devices. When abnormal parameters are detected, indicating that the device is about to fail, a replanning signal is triggered. A meta-learning model is adopted, which learns meta-knowledge of rerouting strategies under various historical fault scenarios in the offline outer loop stage; when a rerouting signal is received, the inner loop performs a small gradient update based on the current faulty equipment information and the distribution of patients en route, and outputs a backup equipment diversion plan and an examination sequence reorganization plan. Update the dynamic pathway guidance for the corresponding patients based on the triage plan and the time sequence reorganization plan.

[0010] The preferred scheme also includes a timing synchronization control step: Combine indoor positioning systems to estimate the patient's estimated arrival time at the target device; When the patient's estimated movement speed is lower than a preset threshold, the instantaneous gravitational value of the target device is reduced, and the device is released to prioritize serving nearby high-priority patients awaiting examination, thus achieving streamline decoupling.

[0011] This invention also proposes a performance-driven fluid path planning system based on digital twins, comprising: The digital twin building module is used to create a weighted directed dynamic graph based on the hospital building information model, where nodes represent medical equipment or examination nodes, and the weights of the edges change dynamically with the equipment utilization rate of the nodes. The equipment status sensing module is used to obtain the operating status, reservation queue and expected idle time of each device in real time; The gravitational field calculation module is used to calculate the gravitational value of the device based on the expected idling time, where the longer the expected idling time, the higher the gravitational value. The path planning module is used to calculate the total path cost when planning a path for a patient, taking into account the physical distance cost, congestion time cost, and negative equipment gravity penalty term, and to determine and issue dynamic path guidance for the next hop based on the total cost.

[0012] In the preferred embodiment, a game optimization module is also included, which is used to define the patient utility function and the system utility function, and solve the Nash equilibrium through multi-stage dynamic game to output the path decision that maximizes the system utility within the acceptable range of patient utility; the path planning module generates dynamic path guidance based on the path decision.

[0013] In the preferred embodiment, a meta-learning compensation module is also included, which is used to access the IoT operating parameters of medical devices, trigger replanning when an anomaly is detected, and use a pre-trained meta-learning model to quickly generate backup equipment diversion schemes and examination sequence reorganization schemes based on the current faulty equipment information and the distribution of patients en route, driving the path planning module to update patient path guidance.

[0014] Compared with the prior art, this application has the following technical effects: (1) By establishing a dynamic correlation between the equipment's gravity value and the expected idle time, the real-time status of the equipment is introduced into the path planning decision-making process. When high-value equipment shows signs of being idle, the system automatically increases the attractiveness of the equipment in terms of pathfinding costs, selects and guides suitable patients from all patients waiting for examination in the hospital, effectively reduces the time that the equipment is idle due to waiting for patients, and directly increases the proportion of the equipment's effective working time.

[0015] (2) Introducing a meta-learning mechanism to solve the replanning problem when equipment suddenly fails or its performance degrades. The model pre-learns patient triage and examination sequence adjustment strategies under various failure scenarios in the offline stage. When the failure actually occurs, only a very small amount of online computation is needed to quickly generate usable alternatives, which significantly shortens the time for the system to recover from the failure state to the normal operating state and reduces the impact on the hospital's diagnosis and treatment process.

[0016] (3) A dynamic game model based on the utility functions of both doctors and patients is adopted to avoid unilateral pursuit of equipment utilization or minimizing patient walking distance. By iteratively solving the Nash equilibrium, the system can allocate some patients to similar equipment with lower current utilization within the patient's acceptable walking distance or waiting time range, thereby improving the overall balance and throughput of hospital equipment resources without worsening the patient experience.

[0017] (4) Combining indoor positioning and predicted arrival time, the system dynamically adjusts resource allocation during patient movement. When a patient's movement speed is detected to be lower than expected, the system will reduce the priority of the patient's original target device and immediately release the device resource to other eligible patients nearby. This decouples the patient movement process from the device reservation status and reduces the secondary idleness of devices caused by individual patient movement delays.

[0018] (5) By using digital twin technology, the hospital's physical space and resources are mapped into a computable dynamic graph model, and path planning is transformed from static route guidance to resource scheduling and control based on global status. The system continuously adjusts the connection relationship according to equipment status, patient location and queue information, so that the overall operation of the hospital is transformed from relying on manual experience to a data-driven, automatically matching supply and demand refined operation process. Attached Figure Description

[0019] Figure 1 This is a flowchart of a performance-driven fluid path planning method based on digital twins. Detailed Implementation

[0020] like Figure 1 As shown, a performance-driven fluid path planning method based on digital twins can be applied to hospital environments, including: A digital twin based on the hospital building information model is constructed, and the hospital space is abstracted as a weighted directed dynamic graph. The nodes in the graph correspond to medical resources and equipment or examination nodes, the weight of the edge represents the path travel cost, and the equipment utilization status of the node dynamically affects the travel cost of the edge. The system acquires the real-time operating status of each device and patient queue information, and calculates the expected idle time of the devices accordingly. The gravitational force value of the device is determined based on the expected idling time, and the gravitational force value increases with the increase of the expected idling time; For patients with a planned path, the total path cost to each candidate device is calculated. The total path cost includes physical distance cost, congestion time cost, and a negative gravity penalty term. This gravity penalty term reduces the total path cost as the gravity value of the target device increases. Dynamic path guidance is generated for patients based on the total path cost. The dynamic path guidance is an instruction pointing to the next target node.

[0021] Preferably, when the device completes the current examination and the next scheduled patient has not yet arrived, resulting in an increase in the expected idle time, the gravity value of the device is increased; the gravity value is calculated through a preset gravitational field potential energy function.

[0022] Preferably, the total path cost is calculated according to the following formula: Total path cost = Physical distance cost + Congestion time cost - k × Target device gravity value; Where k is a positive adjustment coefficient.

[0023] Preferably, it further includes: A dynamic game optimizer for both medical staff and patients is constructed, defining the patient utility function and the system utility function. The patient utility function is negatively correlated with walking distance, number of floor changes, and queuing time, while the system utility function is positively correlated with equipment utilization rate and total table turnover rate. By iteratively solving the Pareto optimal solution through multi-stage dynamic game theory, the patient pathway decision-making can improve system utility without significantly reducing patient utility. The dynamic path guide is the Nash equilibrium path output by the game optimizer.

[0024] Preferably, the game optimizer makes decisions to direct patients to similar devices that are physically farther away but have significantly shorter waiting times, thereby increasing the utilization rate of idle devices without increasing the overall time and physical exertion of patients, and achieving a Nash equilibrium at the system level.

[0025] Preferably, it also includes a burst disturbance compensation step based on meta-learning: The system accesses the IoT operating parameters of medical devices. When abnormal parameters are detected, indicating that the device is about to fail, a replanning signal is triggered. A meta-learning model is adopted, which learns meta-knowledge of rerouting strategies under various historical fault scenarios in the offline outer loop stage; when a rerouting signal is received, the inner loop performs a small gradient update based on the current faulty equipment information and the distribution of patients en route, and outputs a backup equipment diversion plan and an examination sequence reorganization plan. Update the dynamic pathway guidance for the corresponding patients based on the triage plan and the time sequence reorganization plan.

[0026] Preferably, it further includes a timing synchronization control step: Combine indoor positioning systems to estimate the patient's estimated arrival time at the target device; When the patient's estimated movement speed is lower than a preset threshold, the instantaneous gravitational value of the target device is reduced, and the device is released to prioritize serving nearby high-priority patients awaiting examination, thus achieving streamline decoupling.

[0027] A performance-driven fluid path planning system based on digital twins, comprising: The digital twin building module is used to create a weighted directed dynamic graph based on the hospital building information model, where nodes represent medical equipment or examination nodes, and the weights of the edges change dynamically with the equipment utilization rate of the nodes. The equipment status sensing module is used to obtain the operating status, reservation queue and expected idle time of each device in real time; The gravitational field calculation module is used to calculate the gravitational value of the device based on the expected idling time, where the longer the expected idling time, the higher the gravitational value. The path planning module is used to calculate the total path cost when planning a path for a patient, taking into account the physical distance cost, congestion time cost, and negative equipment gravity penalty term, and to determine and issue dynamic path guidance for the next hop based on the total cost.

[0028] Preferably, it also includes a game optimization module, which is used to define the patient utility function and the system utility function, and solve the Nash equilibrium through multi-stage dynamic game to output the path decision that maximizes the system utility within the acceptable range of patient utility; the path planning module generates dynamic path guidance based on the path decision.

[0029] Preferably, it also includes a meta-learning compensation module, which is used to access the IoT operating parameters of medical devices, trigger replanning when an anomaly is detected, and use a pre-trained meta-learning model to quickly generate a backup device diversion plan and an examination sequence reorganization plan based on the current faulty device information and the distribution of patients en route, driving the path planning module to update the patient path guidance.

[0030] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A performance-driven fluid path planning method based on digital twins, characterized in that, include: A digital twin based on the hospital building information model is constructed, and the hospital space is abstracted as a weighted directed dynamic graph. The nodes in the graph correspond to medical resources and equipment or examination nodes, the weight of the edge represents the path travel cost, and the equipment utilization status of the node dynamically affects the travel cost of the edge. The system acquires the real-time operating status of each device and patient queue information, and calculates the expected idle time of the devices accordingly. The gravitational force value of the device is determined based on the expected idling time, and the gravitational force value increases with the increase of the expected idling time; For patients whose paths are to be planned, the total path cost to each candidate device is calculated. The total path cost includes physical distance cost, congestion time cost, and a negative gravity penalty term. The gravity penalty term reduces the total path cost as the gravity value of the target device increases. Dynamic path guidance is generated for patients based on the total path cost. The dynamic path guidance is an instruction pointing to the next target node.

2. The efficiency-driven fluid path planning method based on digital twins according to claim 1, characterized in that, When the device completes the current examination and the next scheduled patient has not yet arrived, resulting in an increase in the expected idle time, the device's gravity value is increased; the gravity value is calculated using a preset gravitational field potential energy function.

3. The efficiency-driven fluid path planning method based on digital twins according to claim 1, characterized in that, The total path cost is calculated using the following formula: Total path cost = Physical distance cost + Congestion time cost - k × Target device gravity value; Where k is a positive adjustment coefficient.

4. The efficiency-driven fluid path planning method based on digital twins according to claim 1, characterized in that, Also includes: A dynamic game optimizer for both medical staff and patients is constructed, defining the patient utility function and the system utility function. The patient utility function is negatively correlated with walking distance, number of floor changes, and queuing time, while the system utility function is positively correlated with equipment utilization rate and total table turnover rate. By iteratively solving the Pareto optimal solution through multi-stage dynamic game theory, the patient pathway decision-making can improve system utility without significantly reducing patient utility. The dynamic path guide is the Nash equilibrium path output by the game optimizer.

5. The efficiency-driven fluid path planning method based on digital twins according to claim 4, characterized in that, The game optimizer makes decisions to direct patients to similar devices that are physically farther away but have significantly shorter waiting times, thereby increasing the utilization rate of idle devices without increasing the overall time and physical exertion of patients, thus achieving a Nash equilibrium at the system level.

6. The efficiency-driven fluid path planning method based on digital twins according to claim 1 or 4, characterized in that, It also includes a burst perturbation compensation step based on meta-learning: The system accesses the IoT operating parameters of medical devices. When abnormal parameters are detected, indicating that the device is about to fail, a replanning signal is triggered. A meta-learning model is adopted, which learns meta-knowledge of rerouting strategies under various historical fault scenarios in the offline outer loop stage; when a rerouting signal is received, the inner loop performs a small gradient update based on the current faulty equipment information and the distribution of patients en route, and outputs a backup equipment diversion plan and an examination sequence reorganization plan. Update the dynamic pathway guidance for the corresponding patients based on the triage plan and the time sequence reorganization plan.

7. The efficiency-driven fluid path planning method based on digital twins according to claim 1, characterized in that, It also includes timing synchronization control steps: Combine indoor positioning systems to estimate the patient's estimated arrival time at the target device; When the patient's estimated movement speed is lower than a preset threshold, the instantaneous gravitational value of the target device is reduced, and the device is released to prioritize serving nearby high-priority patients awaiting examination, thus achieving streamline decoupling.

8. A performance-driven fluid path planning system based on digital twins, characterized in that, include: The digital twin building module is used to create a weighted directed dynamic graph based on the hospital building information model, where nodes represent medical equipment or examination nodes, and the weights of the edges change dynamically with the equipment utilization rate of the nodes. The equipment status sensing module is used to obtain the operating status, reservation queue and expected idle time of each device in real time; The gravitational field calculation module is used to calculate the gravitational value of the device based on the expected idling time, where the longer the expected idling time, the higher the gravitational value. The path planning module is used to calculate the total path cost when planning a path for a patient, taking into account the physical distance cost, congestion time cost, and negative equipment gravity penalty term, and to determine and issue dynamic path guidance for the next hop based on the total cost.

9. The efficiency-driven fluid path planning system based on digital twins according to claim 8, characterized in that, It also includes a game optimization module, which defines the patient utility function and the system utility function, and solves the Nash equilibrium through multi-stage dynamic game to output the path decision that maximizes the system utility within the acceptable range of patient utility; the path planning module generates dynamic path guidance based on the path decision.

10. A performance-driven fluid path planning system based on digital twins according to claim 8 or 9, characterized in that, It also includes a meta-learning compensation module, which is used to access the IoT operating parameters of medical devices, trigger replanning when an anomaly is detected, and use a pre-trained meta-learning model to quickly generate backup equipment diversion plans and examination sequence reorganization plans based on the current faulty equipment information and the distribution of patients en route, driving the path planning module to update patient path guidance.

Images