A whole-link closed-loop treatment and work order management method for an old person falling early warning event

Abstract

The application is particularly a whole-link closed-loop disposal and work order management method for old people falling early warning events, relates to the information technology field of old people service, and comprises the following steps: matching resource requirements of each task node with state information of available resources, calculating a resource matching scheme by using a multi-objective optimization algorithm, and pushing a scheduling instruction to an execution personnel; receiving task progress information reported by the execution personnel, comparing actual execution time of each task node with a preset execution time limit in real time, and triggering an intervention process when a time deviation value exceeds an allowed deviation threshold. In the application, a risk level is associated with a disposal task chain template, falling events with different severities can automatically trigger a disposal process and time limit requirement matched therewith, and resource waste or insufficient disposal caused by insufficient response granularity in a unified disposal mode is avoided.

Description

Technical Field

[0001] This invention relates to the field of information technology for elderly care services, and in particular to a closed-loop method for handling and managing work orders for fall warning events in the elderly. Background Technology

[0002] In elderly care institutions and community-based home care settings, falls among the elderly are characterized by their suddenness and the urgent need for timely response. Existing fall monitoring solutions mainly focus on the detection and alarm stages of fall events, using sensors such as video surveillance and wearable devices to identify fall movements and trigger alarms.

[0003] However, existing solutions lack systematic management of the response process after an alarm is triggered. Throughout the entire process, from alarm triggering to rescue personnel arriving at the scene, on-site response execution, and subsequent medical follow-up, each task relies on manual coordination, lacking automatic task breakdown, status tracking, and timeliness control capabilities. When a task is delayed or obstructed, alternative response paths cannot be identified and activated in a timely manner. Furthermore, the execution data from each response is not systematically accumulated and utilized, preventing the continuous adjustment of process parameters based on actual execution conditions. These deficiencies limit response efficiency and make it difficult to quantitatively assess response quality. Summary of the Invention

[0004] The purpose of this invention is to provide a closed-loop handling and work order management method for elderly fall warning events in order to solve the above-mentioned problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A closed-loop approach to handling and managing work orders for fall warning events among the elderly, including: Collect multi-source sensor data on fall events, and output risk levels based on physiological characteristic data, fall posture data, location information and past medical history information, according to risk level classification rules; Based on the risk level, a task chain template is matched from the task chain template library and instantiated to generate a structured task order containing multiple task nodes; The resource requirements of each task node are matched with the status information of available resources, a multi-objective optimization algorithm is used to calculate the resource matching scheme, and the scheduling instructions are pushed to the executors. Receive task progress information reported by the executors, compare the actual execution time of each task node with the preset execution time limit in real time, and trigger intervention when the time deviation exceeds the allowable deviation threshold. Once all task nodes are completed, the audit logs are compiled to generate a full-chain handling report, and the task chain parameters and scheduling parameters are updated based on historical handling sample data.

[0006] Preferably, the risk level classification rule divides fall events into high-risk, medium-risk, and low-risk levels; wherein, when the physiological characteristic data indicates any of the following states: loss of consciousness, massive bleeding, or abnormal breathing, the risk level is determined to be high-risk; when the fall posture data indicates obvious external injury and the physiological characteristic data indicates consciousness, the risk level is determined to be medium-risk; when the fall posture data indicates no obvious external injury and the elderly person can move independently, the risk level is determined to be low-risk; different risk levels correspond to different timeliness requirements for treatment and resource allocation strategies.

[0007] Preferably, each task node in the processing work order includes the following attributes: task type, responsible role, execution time limit, required resource list, and completion criteria; the structured processing work order assigns a unique task identifier to each task node.

[0008] Preferably, the inputs to the multi-objective optimization algorithm are the resource requirements of each task node, the current location and status of each available resource, and the location information of the fall event; the optimization objectives of the multi-objective optimization algorithm include: minimizing the expected response time of each task node, minimizing the number of resource contention conflicts between key task nodes, and maximizing the number of task nodes that can be executed in parallel; the resource matching scheme specifies the executor, calling device, and expected arrival time for each task node.

[0009] Preferably, when the risk level is high-risk, a countdown to the golden rescue time is automatically started while the dispatch instruction is pushed out. The countdown duration is determined according to the execution time limit of the immediate rescue task node. During the countdown, the execution status of the immediate rescue task node and the medical support task node is continuously tracked.

[0010] Preferably, the calculation process of the resource matching scheme further includes: reading the list of currently occupied resources and marking resources that have been occupied by other handling work orders as unavailable; if there are multiple fall events to be handled in the same period, resource matching is performed in order of priority from high to low risk level, and high-priority handling work orders occupy scarce resources first.

[0011] Preferably, the intervention process includes: when the time deviation value exceeds the allowable deviation threshold, sending a reminder to the corresponding executor; when the time deviation value continues to increase or the executor reports an obstruction, triggering a task reassignment process, the task reassignment process including: querying the list of currently available substitutes, reading the current location and task load of each candidate substitute; calculating the availability score of each candidate substitute based on the distance between each candidate substitute and the fall location and the current task load, selecting the candidate substitute with the highest availability score as the reassignment target; reassigning the obstructed task to the selected substitute and pushing a new scheduling instruction; if there are no available substitutes, escalating the notification to higher-level management personnel for intervention and coordination.

[0012] Preferably, the calculation of the availability score incorporates the skill and qualification information of the candidate replacements, and combines the overall suitability score. Calculate using the following formula: ; in, The distance score is obtained by inverting the normalized actual distance between the candidate replacement and the fall location. The load score is obtained by normalizing and inverting the current task load of the candidate replacement. The skills matching score is determined by the degree of matching between the qualification certificate category held by the candidate replacement and the skill category required for the obstructed task. The highest value of 1 is taken when there is a complete match, the value is taken as the proportion of the number of matched items to the total number of required skills when there is a partial match, and zero is taken when there is no match. , , The weighting coefficients for distance score, load score, and skill matching score are respectively, and the sum of the three is 1. The actual distance value on which the distance score depends and the task load value on which the load score depends are respectively normalized to the mean based on the range before being included in the weighted calculation; a comprehensive fit score is selected. The highest-ranking candidate for replacement will be the one assigned.

[0013] Preferably, updating the task chain parameters and scheduling parameters based on historical handling sample data includes: storing the complete data of this handling as a sample in the fall handling knowledge base; based on the historical handling sample data accumulated in the fall handling knowledge base, statistically analyzing the distribution of the actual completion time of each task node under each risk level; when the deviation between the average historical actual completion time of a task node and the current preset time limit exceeds a preset update threshold, updating the preset time limit of the task node to an adjustment value calculated based on the historical handling sample data; adjusting the priority weight values ​​of each resource type in the multi-objective optimization algorithm based on the scheduling success rate and response time performance of various resources in the historical handling sample data; and adjusting the weight coefficients based on the actual response time and task completion rate after each task reassignment in the historical handling sample data. , , The value of .

[0014] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention maps risk levels to task chain templates, so that fall events of different severity can automatically trigger matching handling processes and time requirements, avoiding resource waste or inadequate handling caused by insufficient response granularity under a uniform handling method.

[0015] 2. This invention, by pre-setting clear execution time limits for each task node in the task chain and monitoring deviations in real time, enables timely identification of delays and obstructions during task execution. Through a task reassignment process, obstructed tasks are transferred to available replacement personnel, reducing the risk of task chain interruption due to obstruction of a single execution path. Attached Figure Description

[0016] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0017] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0018] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0019] Example 1 Its specific implementation method is combined with the appendix Figure 1 Please provide a detailed explanation.

[0020] In this embodiment, it includes: In elderly care institutions and community-based home care settings, falls among the elderly are frequent and require timely response. Existing fall monitoring solutions mainly focus on the detection and alarm stages of falls, lacking systematic closed-loop management of the response process after an alarm is triggered. Specifically, from alarm triggering to the arrival of rescue personnel, on-site response execution, and subsequent medical follow-up, there is a lack of capabilities for automatically breaking down each response task, tracking its status, and controlling its timeliness. This results in limited response efficiency and makes it difficult to quantitatively assess and continuously improve the quality of response.

[0021] The execution entity in this embodiment is the fall incident handling work order management system. This system is deployed on the server side of the elderly care facility and communicates with various sensors, mobile terminals, and an information push platform.

[0022] A method for managing work orders related to elderly fall incidents according to an embodiment of the present invention includes the following steps: Step 1: Collect multi-source sensor data on fall events and generate risk levels. Real-time monitoring data of the elderly's activity area is collected through pre-deployed multi-source sensors. When the fall detection features in the multi-source sensor data meet the preset fall judgment conditions, a fall event record is generated. Based on the fall event record, physiological characteristic data and fall posture data of the elderly person who fell are collected, the elderly person's current location information is obtained, and the elderly person's past medical history and medication contraindications information are retrieved. The physiological characteristic data, fall posture data, location information, and past medical history information are used as risk assessment inputs, and the risk level corresponding to the fall event is output according to the preset risk level classification rules.

[0023] It should be noted that the aforementioned multi-source sensors include at least two of the following: video surveillance equipment, millimeter-wave radar, wearable devices, and ground vibration sensors. The aforementioned physiological characteristic data includes heart rate, blood pressure, blood oxygen saturation, and respiratory rate. The aforementioned fall posture data includes the type of body posture at the time of the fall and an estimate of the impact force. The aforementioned location information is accurate to the room or bed level.

[0024] It should be noted that the above risk level classification rules divide fall events into three levels: high-risk, medium-risk, and low-risk. A fall event is classified as high-risk when physiological data indicates any of the following: loss of consciousness, severe bleeding, or abnormal breathing. A fall event is classified as medium-risk when fall posture data indicates obvious external injury and physiological data indicates consciousness. A fall event is classified as low-risk when fall posture data indicates no obvious external injury and the elderly person is able to move independently. Different risk levels correspond to different response time requirements and resource allocation strategies.

[0025] Step 2: Decompose the task chain based on risk level and generate structured disposal work orders. Based on the risk level, a corresponding handling task chain template is matched from the pre-stored task chain template library. The matched handling task chain template is instantiated to generate a handling task chain containing multiple task nodes. Each task node includes the following attributes: task type, responsible role, execution time limit, required resource list, and completion criteria. All task nodes in the handling task chain are encapsulated into a structured handling work order, and a unique task identifier is assigned to each task node.

[0026] It should be noted that, taking the response task chain corresponding to a high-risk level as an example, the response task chain for a high-risk level includes the following task nodes: Immediate Rescue Task Node: The responsible role is the caregiver or medical staff closest to the fall location, and the task is to rush to the scene with first-aid equipment. Medical Support Task Node: The responsible role is the resident doctor or emergency center, and the task is to receive the elderly person's basic information and physiological characteristics data, and provide medical guidance or dispatch emergency medical personnel. Management Notification Task Node: The responsible role is the institution's management personnel, and the task is to notify the institution's management and the elderly person's family. On-site Control Task Node: The responsible role is the security personnel, and the task is to evacuate onlookers and keep passageways clear. Follow-up Task Node: The responsible role is the medical coordination personnel, and the task is to arrange hospital beds and contact specialist doctors after the elderly person is admitted to the hospital.

[0027] It should be noted that the task chain for handling medium-risk cases differs from that for high-risk cases in that: the task chain for medium-risk cases does not include on-site control task nodes, and the execution of the medical support task node is adjusted to notify the resident physician for on-site diagnosis and treatment, without needing to call the emergency center. The task chain for low-risk cases only includes immediate rescue task nodes and management notification task nodes, where the immediate rescue task node involves a caregiver arriving on-site to confirm the patient's condition and provide basic care.

[0028] Step 3: Coordinate and schedule the sub-tasks in the work order and obtain resource matching solutions. Retrieve the status information of currently available resources within the organization. Available resources include caregivers, medical staff, security personnel, management personnel, emergency equipment, and vehicles. Match the resource requirements of each task node in the work order with the status information of available resources, and use a multi-objective optimization algorithm to calculate the resource matching scheme. The input of the multi-objective optimization algorithm is the resource requirements of each task node, the current location and status of each available resource, and the location information of the fall event. The optimization objectives of the multi-objective optimization algorithm include: minimizing the expected response time of each task node, minimizing the number of resource contention conflicts between critical task nodes, and maximizing the number of task nodes that can be executed in parallel. The output of the multi-objective optimization algorithm is the resource matching scheme, which specifies the executor, the equipment to be called, and the expected arrival time for each task node.

[0029] The scheduling instructions in the resource matching plan are synchronously pushed to the corresponding personnel through multiple terminals. The push terminals include caregiver mobile terminals, nurse station display terminals, and management personnel back-end terminals.

[0030] It should be noted that when the risk level is high, the fall incident response work order management system automatically starts a countdown to the golden rescue time while sending out dispatch instructions. The countdown duration is determined based on the execution time limit of the immediate rescue task node. During the countdown, the fall incident response work order management system continuously tracks the execution status of the immediate rescue task node and the medical support task node.

[0031] In this embodiment of the application, in order to reduce resource contention conflicts when multiple fall events occur concurrently, the calculation process of the resource matching scheme further includes: reading the list of currently occupied resources and marking resources already occupied by other handling work orders as unavailable. If there are multiple fall events to be handled within the same time period, resource matching is performed sequentially according to the priority order from high to low risk level, with high-priority handling work orders taking priority in occupying scarce resources.

[0032] Step 4: Track the execution status of each subtask in real time and intervene in abnormal tasks. Receive task progress information reported by each task executor via mobile terminal. Task progress information includes the following status types: Departed, Arrived, In Progress, Completed, and Encountered Obstacles.

[0033] The actual execution time of each task node is compared in real time with the preset execution time limit in the work order to calculate the time deviation value of each task node. When the time deviation value of a task node exceeds the preset allowable deviation threshold, an reminder is sent to the person responsible for executing that task node.

[0034] When the time deviation of a task node continues to increase, or when the executor of that task node reports an obstruction, the task reassignment process is triggered. Step 4 of the task reassignment process may specifically include: Step 401: Query the list of currently available replacements and read the current location and task load of each candidate replacement.

[0035] Step 402: Based on the distance between each candidate replacement and the fall location and the current task load, calculate the availability score of each candidate replacement, and select the candidate replacement with the highest availability score as the transfer target.

[0036] Step 403: The blocked task is reassigned to a selected replacement, and a new scheduling instruction is pushed to the selected replacement's mobile terminal. If no replacement is available, the task is escalated to a higher-level management position for coordination.

[0037] Write all status change records and intervention operation records into the audit log of the handling work order.

[0038] In this embodiment, to obtain more accurate replacement personnel matching results during task reassignment, the availability score calculation in step 402 also incorporates the skill and qualification information of candidate replacement personnel. The actual distance value upon which the distance score depends and the task load value upon which the load score depends are respectively processed using mean normalization based on the range, so that both are mapped to the same numerical range before participating in the weighted calculation. Let the comprehensive suitability score of a candidate replacement personnel be... ,but: ; in, The distance score is obtained by normalizing and inverting the actual distance between the candidate replacement and the fall location; the closer the distance, the higher the score. The higher; The load score is obtained by normalizing and inverting the current task load of the candidate replacement; the lower the load, the higher the score. The higher; The skills matching score is determined by the degree of matching between the candidate's qualification certificate category and the skills required for the obstructed task. A perfect match is scored as the highest value of 1, a partial match is scored as the proportion of the number of matched items to the total number of required skills, and no match is scored as zero. These are the weighting coefficients corresponding to the distance rating; The weighting coefficients corresponding to the load score; Weighting coefficients corresponding to skill matching scores; , , The sum of the three is 1. The initial values ​​of each weight coefficient are determined by a preset configuration and are continuously updated based on the historical treatment sample data in step 5. Select the comprehensive adaptation score. The highest-ranking candidate for replacement will be the one assigned.

[0039] Step 5: Generate a full-link handling assessment report and update the task chain parameters and scheduling parameters. In this embodiment of the application, based on step 4, when the status of all task nodes in the work order is updated to "completed", the following steps are also included: Step 501: Summarize the audit logs of the handling work orders and generate a full-chain handling report. The full-chain handling report includes the following: a complete timeline from the moment the fall occurred to the completion time of each task node; comparison data between the actual response time of each task node and the preset execution time limit; records of anomalies and intervention operations triggered during the handling process; and the final treatment result and prognosis information of the elderly person.

[0040] Step 502: Based on the data in the end-to-end handling report, calculate the following key performance indicators: average response time of each task node, on-time task completion rate, and resource scheduling success rate.

[0041] Step 503: Store the complete data of this incident as a sample in the fall handling knowledge base. Based on the historical incident sample data accumulated in the fall handling knowledge base, statistically analyze the distribution of the actual completion time of each task node under each risk level. When the deviation between the average historical actual completion time of a task node and the current preset time limit exceeds the preset update threshold, update the preset time limit of the task node to the adjusted value calculated based on the historical incident sample data. Simultaneously update the resource type priority weights in the multi-objective optimization algorithm in Step 3: Based on the scheduling success rate and response time performance of various resources in the historical incident sample data, adjust the priority weight values ​​of each resource type in the multi-objective optimization algorithm. Simultaneously update the comprehensive adaptation score in Step 4. Weighting coefficients , , Based on the actual response time and task completion rate of each task reallocation in the historical processing sample data, the weighting coefficients are adjusted. , , The value of .

[0042] Example 2 A certain elderly care facility (hereinafter referred to as "Institution A") has 180 beds, with 162 resident elderly people, and is staffed with 14 caregivers, 6 medical staff, 4 security personnel, and 3 management personnel. On a morning in March 20XX, an elderly person fell in the restroom area on the second floor of the east wing of Institution A. The elderly person who fell was identified as RES-047, a 78-year-old male, with a medical history including hypertension and type 2 diabetes, and an allergy to aspirin listed in his medication contraindications record. The fall incident handling work order management system is deployed on Institution A's server and maintains real-time communication with the area's video surveillance equipment, wearable devices, and ground vibration sensors.

[0043] At 09:14:22 on the same day, the risk assessment module received abnormal signals from three sensors. The video surveillance equipment detected that RES-047's posture suddenly changed from standing to lying down, the ground vibration sensor recorded an impact waveform with an amplitude exceeding the threshold, and the wearable device simultaneously reported an acceleration mutation event. All three signals met the fall detection criteria within the same time window, and the system generated a fall event record with the event number EVT-20XX-0312-001.

[0044] Subsequently, the risk assessment module collected RES-047's current physiological characteristics data, read its fall posture data, and located its position as the second-floor restroom in the East Zone. Simultaneously, it retrieved RES-047's past medical history and medication contraindications. This data was then input into the risk level classification rules: a wearable device reporting a decrease in blood oxygen saturation and a sudden increase in heart rate, along with physiological characteristics indicating abnormal breathing, triggers the high-risk level determination criteria. The system classified EVT-20XX-0312-001 as high-risk.

[0045] Table 1 Input data and assessment results for fall event risk assessment

[0046] Based on the high-risk level, the task decomposition module retrieves a high-risk response task chain template from the task chain template library, instantiates it, and fills in the event number, location information, and elderly person's file information of EVT-20XX-0312-001 into each task node, generating a response work order with the work order number WO-20XX-0312-001. This work order contains five task nodes: immediate rescue, medical support, management notification, on-site control, and follow-up, with each node assigned a unique task identifier.

[0047] Table 2 Task Node Structure for Work Orders

[0048] At 09:14:35, the collaborative scheduling module reads the status information of all currently available resources in Organization A. At this time, Caregiver A (NW-03) is located on the first floor of the East Wing, performing a routine patrol task with a low task load; Caregiver B (NW-07) is located on the third floor of the West Wing, currently without a task; the resident doctor (MD-01) is located in the medical room; and Security Personnel C (SG-02) is located at the gate on the first floor of the East Wing. The multi-objective optimization algorithm takes the resource requirements of each task node, the current location and status of each person, and the location of the fall (the restroom on the second floor of the East Wing) as input, and aims to minimize the expected response time, minimize resource contention conflicts, and maximize the number of tasks executed in parallel, outputting a resource matching scheme.

[0049] The system simultaneously detected another low-risk fall event, EVT-20XX-0312-002, which was being handled and had already been assigned to caregiver Ding (NW-11). The multi-objective optimization algorithm marked NW-11 as unavailable, and the high-risk work order WO-20XX-0312-001 prioritized the use of scarce emergency medical equipment resources. A resource matching plan was generated at 09:14:38, and dispatch instructions were simultaneously pushed to NW-03's mobile terminal, the nurse station display terminal, and the administrator's backend terminal. A countdown for the high-risk level's golden rescue time immediately started, with a countdown duration of 3 minutes, corresponding to the execution time limit of T-001.

[0050] Table 3 Output Results of Resource Matching Scheme

[0051] The status tracking module continuously receives progress information reported by each executor via mobile terminal after the scheduling instruction is pushed. At 09:17:08, NW-03 reported "Arrived," and T-001's actual arrival time was 2 minutes and 33 seconds, which did not exceed the 3-minute execution time limit, so no follow-up was needed. At 09:18:45, SG-02 reported "Obstructed," due to an elevator malfunction on the second floor of the East Zone, making it impossible to bring the isolation equipment to the site. T-004's time deviation continued to increase, triggering the task reassignment process.

[0052] Step 401: The status tracking module queries the list of currently available replacement personnel and reads the location and task load of security personnel Ding (ID SG-04, located at the third floor stairwell of the East Zone) and security personnel Wu (ID SG-05, located at the South Zone entrance).

[0053] Step 402: Using SG-04 and SG-05 as candidates, apply the comprehensive fit score formula. Calculate the overall suitability score for each candidate, where , , These are the weighting coefficients. Rate the distance. To score the load, Skill matching rating. SG-04 is located on the third floor of the East Wing, close to the restrooms on the second floor of the East Wing, within walking distance. Distance rating. After normalization of the range and mean, the score is 0.87, the task load is zero, and the load score is [not specified]. A score of 1.00 indicates the candidate holds a basic first aid qualification certificate and possesses skills that perfectly match the requirements for on-site control tasks. (Skills matching score) The rating is 1.00; SG-05 is located in the South District, which is quite far away, hence the distance rating. The score is 0.21, the task load is zero, and the load score is [missing information]. The score is 1.00, with no relevant qualifications, and the skill matching score is [not specified]. The score is 0.00. Substituting each score into the formula, the overall fit score for SG-04 is... Significantly higher than SG-05, SG-04 was selected as the transfer target.

[0054] Step 403: The system pushes a new scheduling instruction to the mobile terminal of SG-04. SG-04 reports "departed" at 09:19:20 and "arrived" at 09:20:15. The completion time of T-004 exceeds the original preset time limit but does not exceed the overall timeliness requirements for high-risk handling. All status change records and intervention operation records are written to the audit log of WO-20XX-0312-001.

[0055] Table 4 Task Execution Status Tracking and Intervention Records

[0056] Table 5 Input and Output of Comprehensive Suitability Score Calculation for Candidate Replacements

[0057] T-005 was updated to "Completed" at 09:46:30. All task nodes of WO-20XX-0312-001 were completed, and the closed-loop evaluation module was started.

[0058] Step 501: The closed-loop assessment module summarizes the audit logs and generates a full-chain response report. The report includes a complete timeline from the fall incident at 09:14:22 to the completion of all tasks at 09:46:30, records the obstacles encountered by T-004 and the reassignment intervention, and includes the final medical treatment result of RES-047.

[0059] Step 502: Calculate key performance indicators based on the end-to-end handling report. T-001 to T-003 and T-005 were all completed on time, and T-004 was completed after being reassigned, with an on-time completion rate of 4 / 5; in the resource scheduling success rate statistics, 4 items were successfully matched on the first attempt, and 1 item was successfully matched after being reassigned.

[0060] Step 503: Store the data from this incident as a sample in the fall handling knowledge base. The historical average completion time for high-risk level T-004 (on-site control) in the knowledge base deviates from the preset 4-minute time limit beyond the update threshold; therefore, the system adjusts the preset time limit for T-004 to 5 minutes and 30 seconds. Simultaneously update the priority weights of security personnel resource types in the multi-objective optimization algorithm, increasing their scheduling priority in the East Zone restroom scenario. Simultaneously update the comprehensive adaptation score. Weighting coefficient: After this reassignment, the actual response time of SG-04 was good, and the task completion rate was 1. The weight will be slightly increased in the next update. remain unchanged. Slightly lowered.

[0061] Table 6 Summary of Key Performance Indicators for End-to-End Processing

[0062] The entire process exhibits a clear data flow logic. Step 1, after risk assessment, outputs a high-risk level from the sensor data. This level serves as input for Step 2, driving the instantiation of the high-risk task chain template into a structured work order with five nodes. The resource requirements of each task node in the work order are then processed in Step 3, where they, along with real-time resource status data, drive a multi-objective optimization algorithm to output a resource matching solution and trigger a countdown to the golden rescue time. Step 4 continuously receives progress reports from each executor. The obstruction status of T-004 triggers the calculation of the comprehensive adaptation score, and the scores for SG-04 and SG-05 are substituted into the formula. Subsequently, SG-04's comprehensive fit score was significantly higher, driving the transfer decision, and all intervention records were written to the audit log. Step 5 summarizes the audit log to generate a full-link handling report, calculates performance indicators, stores the handling data in the knowledge base, and updates the time limit parameters of Step 2, the priority weight of Step 3, and the weight coefficient of Step 4 in reverse, forming a complete data-driven closed loop.

[0063] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0064] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0065] 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 closed-loop handling and work order management of elderly fall warning events, characterized in that, include: Collect multi-source sensor data on fall events, and output risk levels based on physiological characteristic data, fall posture data, location information and past medical history information, according to risk level classification rules; Based on the risk level, a task chain template is matched from the task chain template library and instantiated to generate a structured task order containing multiple task nodes; The resource requirements of each task node are matched with the status information of available resources, a multi-objective optimization algorithm is used to calculate the resource matching scheme, and the scheduling instructions are pushed to the executors. Receive task progress information reported by the executors, compare the actual execution time of each task node with the preset execution time limit in real time, and trigger intervention when the time deviation exceeds the allowable deviation threshold. Once all task nodes are completed, the audit logs are compiled to generate a full-chain handling report, and the task chain parameters and scheduling parameters are updated based on historical handling sample data.

2. The method for closed-loop handling and work order management of elderly fall warning events according to claim 1, characterized in that, The risk level classification rules divide fall events into high-risk, medium-risk, and low-risk levels. Specifically, when the physiological characteristic data indicates any of the following states: loss of consciousness, massive bleeding, or abnormal breathing, the risk level is determined to be high-risk; when the fall posture data indicates obvious external injury and the physiological characteristic data indicates that the person is conscious, the risk level is determined to be medium-risk; when the fall posture data indicates no obvious external injury and the elderly person is able to move independently, the risk level is determined to be low-risk. Different risk levels correspond to different timeliness requirements for treatment and resource allocation strategies.

3. The method for closed-loop handling and work order management of elderly fall warning events according to claim 1, characterized in that, Each task node in the processing work order contains the following attributes: task type, responsible role, execution time limit, required resource list, and completion criteria; the structured processing work order assigns a unique task identifier to each task node.

4. The method for closed-loop handling and work order management of elderly fall warning events according to claim 1, characterized in that, The inputs to the multi-objective optimization algorithm are the resource requirements of each task node, the current location and status of each available resource, and the location information of the fall event. The optimization objectives of the multi-objective optimization algorithm include: minimizing the expected response time of each task node, minimizing the number of resource contention conflicts between critical task nodes, and maximizing the number of task nodes that can be executed in parallel. The resource matching scheme specifies the executor, calling device, and expected arrival time for each task node.

5. The method for closed-loop handling and work order management of elderly fall warning events according to claim 1, characterized in that, When the risk level is high, the countdown to the golden rescue time will be automatically started at the same time as the dispatch instruction is pushed. The countdown duration is determined according to the execution time limit of the immediate rescue task node. During the countdown, the execution status of the immediate rescue task node and the medical support task node will be continuously tracked.

6. The method for closed-loop handling and work order management of elderly fall warning events according to claim 4, characterized in that, The calculation process of the resource matching scheme also includes: reading the list of currently occupied resources and marking resources that have been occupied by other handling work orders as unavailable; if there are multiple fall events to be handled in the same period, resource matching is performed in order of priority from high to low risk level, and high-priority handling work orders take priority in occupying scarce resources.

7. The method for closed-loop handling and work order management of elderly fall warning events according to claim 1, characterized in that, The intervention process includes: when the time deviation value exceeds the allowable deviation threshold, sending a reminder to the corresponding executor; when the time deviation value continues to increase or the executor reports an obstruction, triggering a task reassignment process, which includes: querying the list of currently available substitutes, reading the current location and task load of each candidate substitute; calculating the availability score of each candidate substitute based on the distance between each candidate substitute and the fall location and the current task load, selecting the candidate substitute with the highest availability score as the reassignment target; reassigning the obstructed task to the selected substitute and pushing a new scheduling instruction; if there are no available substitutes, escalating the notification to higher-level management personnel for intervention and coordination.

8. The method for closed-loop handling and work order management of elderly fall warning events according to claim 7, characterized in that, The usability score calculation incorporates the skills and qualifications of candidate replacements, along with a comprehensive suitability score. Calculate using the following formula: ； in, The distance score is obtained by inverting the normalized actual distance between the candidate replacement and the fall location. The load score is obtained by normalizing and inverting the current task load of the candidate replacement. The skills matching score is determined by the degree of matching between the qualification certificate category held by the candidate replacement and the skill category required for the obstructed task. The highest value of 1 is taken when there is a complete match, the value is taken as the proportion of the number of matched items to the total number of required skills when there is a partial match, and zero is taken when there is no match. , , The weighting coefficients for distance score, load score, and skill matching score are respectively, and the sum of the three is 1. The actual distance value on which the distance score depends and the task load value on which the load score depends are respectively normalized to the mean based on the range before being included in the weighted calculation; a comprehensive fit score is selected. The highest-ranking candidate for replacement will be the one assigned.

9. The method for closed-loop handling and work order management of elderly fall warning events according to claim 8, characterized in that, Updating task chain parameters and scheduling parameters based on historical handling sample data includes: storing the complete data of this handling as a sample in the fall handling knowledge base; based on the historical handling sample data accumulated in the fall handling knowledge base, statistically analyzing the distribution of the actual completion time of each task node under each risk level; when the deviation between the average historical actual completion time of a task node and the current preset time limit exceeds a preset update threshold, updating the preset time limit of the task node to an adjustment value calculated based on the historical handling sample data; adjusting the priority weight values ​​of each resource type in the multi-objective optimization algorithm based on the scheduling success rate and response time performance of various resources in the historical handling sample data; and adjusting the weight coefficients based on the actual response time and task completion rate after the reallocation of each task in the historical handling sample data. , , The value of .

Images