A digital preplan execution optimization method, device, equipment and storage medium
By breaking down emergency plans into task lists in smart parks and combining them with resource allocation to generate optimal execution paths, the problem of static emergency plans being unable to adapt to the actual environment is solved, thus improving disaster relief efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING YONGSHENG JIETAI TECH CO LTD
- Filing Date
- 2026-01-30
- Publication Date
- 2026-05-29
AI Technical Summary
In traditional smart park emergency response, static emergency plans cannot dynamically adapt to the actual environment, resulting in low disaster relief efficiency.
By combining emergency plans with park resources, the emergency plans are broken down into task lists, and the optimal execution path is generated based on resource allocation to optimize the order of disaster relief tasks.
It improves the response time and efficiency of disaster relief missions, dynamically adapts to the actual environment, and generates the optimal sequence of disaster relief mission execution.
Smart Images
Figure CN122114309A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart park technology, and in particular to a method, apparatus, equipment and storage medium for optimizing the execution of digital contingency plans. Background Technology
[0002] Currently, in the field of emergency management in smart parks, after a disaster occurs, the system often matches the emergency plan with pre-set plans to select the one most suitable for the current disaster, and then carries out disaster relief according to the execution measures in the matched emergency plan. However, this disaster relief approach is based on static emergency plans and cannot provide the optimal sequence of disaster relief tasks according to the actual environment, resulting in low disaster relief efficiency. In other words, the static emergency plans in traditional smart park emergency response cannot dynamically adapt to the actual environment and provide the optimal execution plan for the current situation.
[0003] Therefore, how to optimize the execution process of contingency plans and improve the efficiency of disaster relief has become an urgent problem to be solved by those skilled in the art. Summary of the Invention
[0004] In view of this, this application proposes a digital contingency plan execution optimization method, apparatus, equipment and storage medium, which can generate the optimal disaster relief task execution sequence based on an emergency plan that matches the current disaster when a disaster occurs, which can effectively improve the efficiency of disaster relief.
[0005] According to a first aspect of this application, a digital emergency plan execution optimization method is provided. This method generates the optimal disaster relief task execution sequence of the current emergency plan when a disaster occurs in a park. The method includes: Obtain current emergency plans and resources within the park; The current emergency plan is broken down to generate a task list containing at least two execution tasks; Based on the task list and the resources mentioned above, the optimal execution path of the emergency plan is generated, and the optimal execution path is the current optimal order of disaster relief tasks.
[0006] In one possible implementation, when breaking down the current emergency plan to generate a task list containing at least two execution tasks, the following is included: Based on the acquired current emergency plan, identify each step, each entity, and the relationships between them in the current emergency plan; Based on the relationships between the entities, each entity is bound to each step to generate a task description corresponding to each step, and each task description is identified to obtain a task identifier. The task list is generated based on the task identifier and the description of each task.
[0007] In one possible implementation, the entity includes at least one of an execution entity, a resource, and an object.
[0008] In one possible implementation, generating the optimal execution path for the emergency plan based on the task list and each of the resources includes: Obtain the task list and all resources within the current park; Based on preset constraints, resources are allocated to each task in the task list to obtain the currently executable task list; The optimal execution path is the task execution path whose execution time is the shortest when the execution time of each task in the currently executable task list is calculated and combined with the constraints.
[0009] In one possible implementation, the constraints include at least one of resource availability and temporal dependencies between tasks.
[0010] In one possible implementation, before generating the optimal order of disaster relief tasks for the current emergency plan, the process also includes confirming the emergency plan corresponding to the current disaster.
[0011] In one possible approach, when identifying an emergency plan that matches the current disaster, the following is included: Obtain real-time alarm information within the park; Calculate the similarity between the real-time alarm information and each emergency plan in the contingency plan database; Based on the similarity, an emergency plan matching the current disaster is identified.
[0012] According to a second aspect of this application, a digital contingency plan execution optimization device is provided, used to generate the optimal disaster relief task execution sequence of the current emergency plan when a disaster occurs in a smart park, including: The data acquisition module is used to acquire the current emergency plan and various resources in the current park; The task list generation module is used to break down the current emergency plan and generate a task list containing at least two execution tasks. The optimal execution path generation module is used to generate the optimal execution path of the emergency plan based on the task list and each of the resources. The optimal execution path is the current optimal order of execution of disaster relief tasks.
[0013] According to a third aspect of this application, a digital contingency plan execution optimization device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the method described in the first aspect of this application.
[0014] According to a fourth aspect of this application, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions, when executed by a processor, implement the method described in the first aspect of this application.
[0015] This application provides a digital emergency plan execution optimization method. This method generates the optimal disaster relief task execution sequence of the current emergency plan when a disaster occurs in a park. The method includes: acquiring the current emergency plan and the resources of the current park; decomposing the current emergency plan to generate a task list containing multiple execution tasks; and generating the optimal execution path of the emergency plan based on the task list and the resources, whereby the optimal execution path is the optimal disaster relief task execution sequence. When a disaster occurs in a park, this application decomposes the currently matching emergency plan into an executable task list, and then allocates resources based on the existing resources within the park. This ensures that each task requires the most suitable resources at the moment of execution, and based on the allocated resources, generates the task execution path with the shortest execution time for each task in the task list, thereby improving disaster relief response time and efficiency.
[0016] Other features and aspects of this application will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description
[0017] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this application together with the specification and serve to explain the principles of this application.
[0018] Figure 1 A flowchart illustrating a digital plan execution optimization method according to an embodiment of this application is shown; Figure 2 This is a schematic block diagram of a digital plan execution optimization apparatus according to an embodiment of the present application; Figure 3 A schematic block diagram of a digital scheme execution optimization device according to an embodiment of this application is shown. Detailed Implementation
[0019] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.
[0020] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.
[0021] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.
[0022] <Method Implementation> Figure 1 A flowchart illustrating a digital plan execution optimization method according to an embodiment of this application is shown. Figure 1 As shown, this method is used to generate the optimal disaster relief task execution order of the current emergency plan when a disaster occurs in the park. The method includes steps S1100-S1300: S1100, obtaining the current emergency plan and various resources in the current park; S1200, decomposing the current emergency plan to generate a task list containing at least two execution tasks; S1300, generating the optimal execution path of the current emergency plan based on the task list and various resources, wherein the optimal execution path is the current optimal disaster relief task execution order.
[0023] In the event of a disaster in the park, this application breaks down the current emergency plan into a list of executable tasks. Then, it allocates resources based on the existing resources within the park, ensuring that each task requires the most suitable resources at the moment. Based on the allocated resources, it generates the task execution path with the shortest execution time for each task in the task list, thereby improving the response time and efficiency of disaster relief.
[0024] It should be noted that before generating the optimal order of disaster relief tasks in the current emergency plan, the process also includes confirming the emergency plan corresponding to the current disaster.
[0025] In one possible implementation, the process of reconfirming an emergency plan matching the current disaster includes: acquiring real-time alarm information within the park; calculating the similarity between the real-time alarm information and each emergency plan in the plan database; and confirming an emergency plan matching the current disaster based on the calculated similarity. The real-time alarm information within the park can be directly obtained through the park's monitoring equipment. For example, alarm information obtainable during a fire could include smoke detector IDs, temperature sensor readings, and flame / ashes image feature values captured by cameras.
[0026] This application includes a pre-set contingency plan database, which stores various emergency plans developed for potential disasters in the park.
[0027] The similarity between real-time alarm information and each emergency plan in the contingency plan database is calculated by comparing the feature vectors of the alarm information with the feature vectors of each emergency plan. Specifically, the acquired real-time alarm information is first converted into a feature vector V_alert.
[0028] V_alert can be expressed by the formula: V_alert = [f1, f2, f3, ... fn] In the formula, Fn represents the confidence score of the nth sensor source.
[0029] It should be noted that the confidence scores of each sensor can be obtained through manual configuration.
[0030] After obtaining the feature vectors of the real-time alarm information, it is also necessary to obtain the feature vectors V_plan_k corresponding to each emergency plan. The feature vector V_plan_k corresponding to each emergency plan can be expressed by the formula: V_plan_k = [w1, w2, w3, ... wn] In the formula, k represents the kth plan in the plan library, and wn represents the weight of the nth feature in the current plan.
[0031] It should be noted that the weights of each feature in the plan are calculated using a preset algorithm. This application preferably uses the TF-IDF algorithm for calculation.
[0032] It should be noted that any vector conversion method can be used to convert real-time alarm information into feature vectors and to convert each contingency plan into its corresponding feature vector; no specific method is limited.
[0033] After obtaining the feature vectors of real-time alarm information and each contingency plan, an event classification model based on semantic understanding and feature weighting is used to perform vector matching between the real-time alarm information and each contingency plan in the contingency plan library. This is achieved by calculating the similarity between the feature vectors of the real-time alarm information and the corresponding feature vectors of each contingency plan. This application preferably uses the cosine similarity of the two vectors, which can be expressed by the following formula: Similarity = (V_alert · V_plan_k) / (||V_alert|| ||V_plan_k||) After obtaining the similarity scores between real-time alarm information and various contingency plans, the emergency plan matching the current disaster is selected from the calculated similarity scores. Specifically, similarity scores greater than a preset threshold are selected. When the selected result is unique, the emergency plan corresponding to the current similarity score is the emergency plan matching the current disaster. When the selected result is not unique, the emergency plan corresponding to the highest similarity score is directly selected as the emergency plan matching the current disaster.
[0034] If no similarity score greater than a preset threshold is found, the emergency plan with the highest similarity score is directly selected as the emergency plan matching the current disaster. The preset threshold can be determined according to the actual situation, and is preferably set to 0.9.
[0035] After obtaining an emergency plan matching the current disaster, the process involves breaking down the current emergency plan into a task list containing at least two execution tasks. Specifically, this application utilizes Natural Language Processing (NLP) technology to decompose the current emergency plan into atomic task instructions (i.e., a task list).
[0036] In one possible implementation, when decomposing the current emergency plan, the process includes identifying each step, each entity, and the relationship between each entity in the current emergency plan based on the acquired current emergency plan; binding each entity to each step according to the relationship between each entity to generate a task description corresponding to each step, and identifying each task description to obtain a task identifier; and generating the task list based on the task identifier and each task description.
[0037] Specifically, key verb phrases are first identified from the current emergency plan as the identified steps. The relationships between these steps are then determined using the flowcharts configured in each plan. It should be noted that each plan contains a flowchart of the steps to be executed; therefore, the corresponding flowcharts can be used to identify the steps within the plan. Furthermore, the relationships between the steps can also be derived from the flowcharts.
[0038] Then, Named Entity Recognition (NER) and Relation Extraction (RE) are used to identify the entities in the current emergency plan and the relationships between them. The identified entities include at least one of the following: implementing entities, required resources, and objects. The implementing entities include responsible departments and executors; the required resources include materials and equipment; and the objects include buildings and floors.
[0039] Then, based on the identified relationships between the entities, an entity is bound to each step to generate a corresponding task description for each step. This task description generation for each step is done manually. For example, if the four tasks are the four steps of an emergency response plan, each step requires manually adding and configuring the responsible party, detailed description, and operational information.
[0040] Based on the above operations, task descriptions corresponding to each step can be obtained. Then, task identifiers are assigned to each task description according to the relationship between the steps, generating the task list based on the task identifiers and their corresponding task descriptions. Specifically, during task identifier assignment, each step is marked sequentially, such as Task 1, Task 2, etc. Each task identifier represents a task description corresponding to a different step. Arranging the tasks according to their task identifiers yields the task list corresponding to the current emergency plan.
[0041] For example, the generated task list can be seen as follows: Task 1 (Fire Extinguishing): Instructions - Activate the sprinkler system on the 3rd floor of Building 7; Responsible Entity - Property Engineering Department; Resources - Fire Protection System Authority.
[0042] Task 2 (Evacuation): Instructions - Issue a voice broadcast to guide personnel to evacuate through safety exits 2 and 4; Responsible entity - Security Center.
[0043] Task 3 (Environmental Monitoring): Instructions - Dispatch personnel to deploy atmospheric VOCs monitoring equipment 50 meters downwind; Responsible entity - Environmental protection team; Resources - Portable monitoring instrument.
[0044] Task 4 (Public Opinion Management): Instructions - Release the first incident report through the official WeChat account; Responsible entity - Brand Public Relations Department.
[0045] After breaking down the current emergency plan to generate an executable task list, the optimal execution path for the current emergency plan is generated by combining the resources within the current park.
[0046] In one possible implementation, generating the optimal execution path for the current emergency plan by combining resources within the current park includes: obtaining the current task list and all resources within the current park; allocating resources to each task in the task list according to preset constraints to obtain a currently executable task list; calculating the execution time of each task in the currently executable task list, and selecting the task execution path with the shortest execution time based on the constraints, which is the optimal execution path. The preset constraints include at least one of resource availability and temporal dependencies between tasks. It should be noted that while identifying each step in the current emergency plan, the temporal dependencies between each step are also identified. These temporal dependencies determine which steps will be prioritized for execution during subsequent disaster relief tasks.
[0047] In one possible implementation, the park is equipped with a resource monitoring system to collect real-time statistics on all resources within the park and their real-time status, thereby providing usable resources for disaster relief efforts in the event of a disaster. These resources include the location of fire trucks, personnel on-duty status, and material inventory.
[0048] Based on the resource monitoring system, the current resources and their status within the park are obtained. By using multi-agent reinforcement learning (MARL) or a greedy algorithm to allocate the resources required by each task in the task list, the current list of executable tasks can be obtained.
[0049] It should be noted that this application allocates resources to each task in the task list according to pre-set configuration rules. The pre-set rules are: with the accident point as the center, a radius of X kilometers is set, and resources within the radius are preferably allocated to each task. Specifically, resource allocation is achieved by matching resources within the radius with the resources required by each task in the current task list. It should be noted that each task may have multiple matching resources, and multiple tasks to be executed can be generated based on the resources matched to the current task. Furthermore, this application first generates a task execution order based on the temporal dependencies of each task in the aforementioned task list and the constraints. Based on this task execution order, the tasks to be executed corresponding to each task are arranged and combined to obtain multiple lists of tasks to be executed. Then, based on the resource status in the constraints, the execution time of each task list is predicted, and the task list with the shortest execution time is selected as the current optimal execution path. Those skilled in the art can understand this as the process of confirming the optimal path being the process of confirming the optimal allocation of resources for each task. That is, when executing the current task, based on the resources matched to the current task and the status of each resource, the resource that takes the least time to execute the current task is selected as the resource allocated to the current task.
[0050] The optimal execution path can be selected using the following formula: Minimize Σ (T_task_i) In the formula, T_task_i represents the time required to complete the i-th task.
[0051] Based on the above operations, the optimal task execution order for the current disaster relief efforts can be obtained. Executing these tasks according to this order will then achieve the goal of disaster relief in the current area. Furthermore, the given optimal execution path significantly improves the efficiency and accuracy of the area's emergency response.
[0052] In one possible implementation, after the task execution is completed, a process of optimizing the task execution results is also included. Specifically, the execution results after task execution are obtained, and based on the execution results, the task list decomposition and resource allocation scheme in this application are continuously optimized according to a pre-set update strategy model, so that a better execution scheme can be adopted for disaster relief in subsequent similar events.
[0053] In one possible implementation, this application continuously receives feedback on task execution (such as "sprinklers activated," "evacuation routes blocked") after the task is completed, compares this feedback with the expected results, and continuously optimizes the current emergency plan based on the results. This allows the plan to dynamically adapt to actual changes and provides a more flexible basis for disaster relief. Specifically, this application feeds a feedback reward signal R (success / failure) at the end of each task execution to a reinforcement learning model. The reinforcement learning model continuously updates the task execution strategy (i.e., the optimal execution path in this application) based on the feedback reward signal. This can be expressed by the following formula: V(s) ← V(s) + α [R + γV(s') - V(s)] In the formula, R represents the feedback stimulus signal. R+γV(s′): called the TD Target, representing the estimated target value based on the current reward R and the next state value V(s). R+γV(s`)−V(s): called the TD Error, measuring the deviation between the current state value estimate V(s) and the TD Target, updated by scaling with α (learning rate). This part is also the loss function form of the Critic network in methods such as PPO. α: learning rate, used to scale the TD error; the larger the learning rate, the larger the magnitude of V(S) update.
[0054] Therefore, this application provides a digital emergency plan execution optimization method. This method generates the optimal disaster relief task execution sequence of the current emergency plan when a disaster occurs in a park. The method includes acquiring the current emergency plan and all resources within the park; decomposing the current emergency plan to generate a task list containing at least two execution tasks; and generating the optimal execution path of the current emergency plan based on the task list and various resources. This optimal execution path is the current optimal disaster relief task execution sequence. When a disaster occurs in a park, this application decomposes the currently matching emergency plan into an executable task list, and then allocates resources based on existing resources within the park. This ensures that each task requires the most suitable resources at the moment of execution, and based on the allocated resources, generates the task execution path with the shortest execution time for each task in the task list, thereby improving disaster relief response time and efficiency.
[0055] <Device Embodiment> Figure 2 A schematic block diagram of a digital plan execution optimization apparatus according to an embodiment of this application is shown. Figure 2 As shown, the device 100 is used to generate the optimal disaster relief task execution sequence of the current emergency plan when a disaster occurs in a smart park. The device 100 includes: a data acquisition module 110, a task list generation module 120, and an optimal execution path generation module 130. The data acquisition module 110 is used to acquire the current emergency plan and various resources of the current park; the task list generation module 120 is used to decompose the current emergency plan to generate a task list containing at least two execution tasks; and the optimal execution path generation module 130 is used to generate the optimal execution path of the emergency plan based on the task list and the resources. This optimal execution path is the current optimal disaster relief task execution sequence.
[0056] <Equipment Example> Figure 3 A schematic block diagram of a digital scheme execution optimization device according to an embodiment of this application is shown. Figure 3 As shown, the digital contingency plan execution optimization device 200 includes a processor 210 and a memory 220 for storing executable instructions of the processor 210. The processor 210 is configured to implement any of the aforementioned digital contingency plan execution optimization methods when executing the executable instructions.
[0057] It should be noted here that the number of processors 210 can be one or more. Furthermore, the digital plan execution optimization device 200 in this embodiment may also include an input device 230 and an output device 240. The processors 210, memory 220, input device 230, and output device 240 can be connected via a bus or other means, which are not specifically limited here.
[0058] The memory 220, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the digital plan execution optimization method in this application embodiment. The processor 210 executes various functional applications and data processing of the digital plan execution optimization device 200 by running the software program or module stored in the memory 220.
[0059] Input device 230 can be used to receive input digital numbers or signals. These signals may include key signals related to user settings and function control of the device / terminal / server. Output device 240 may include a display device such as a screen.
[0060] <Storage Medium Examples> According to a fourth aspect of this application, a non-volatile computer-readable storage medium is also provided, on which computer program instructions are stored, which, when executed by processor 210, implement any of the aforementioned digital scheme execution optimization methods.
[0061] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technological improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.
Claims
1. A method for optimizing the execution of digital contingency plans, characterized in that, Used to generate the optimal order of disaster relief task execution in the event of a disaster in the park, including: Obtain current emergency plans and resources within the park; The current emergency plan is broken down to generate a task list containing at least two execution tasks; Based on the task list and the resources mentioned above, the optimal execution path of the emergency plan is generated, and the optimal execution path is the current optimal order of disaster relief tasks.
2. The method for optimizing the execution of a digital contingency plan according to claim 1, characterized in that, When breaking down the current emergency plan to generate a task list containing at least two execution tasks, the following are included: Based on the acquired current emergency plan, identify each step, each entity, and the relationships between them in the current emergency plan; Based on the relationships between the entities, each entity is bound to each step to generate a task description corresponding to each step, and each task description is identified to obtain a task identifier. The task list is generated based on the task identifier and the description of each task.
3. The method for optimizing the execution of a digital contingency plan according to claim 2, characterized in that, The entity includes at least one of the following: an execution entity, a resource, and an object.
4. The method for optimizing the execution of a digital contingency plan according to claim 1, characterized in that, When generating the optimal execution path for the emergency plan based on the task list and the resources mentioned above, the following steps are included: Obtain the task list and all resources within the current park; Based on preset constraints, resources are allocated to each task in the task list to obtain the currently executable task list; The optimal execution path is the task execution path whose execution time is the shortest when the execution time of each task in the currently executable task list is calculated and combined with the constraints.
5. The method for optimizing the execution of a digital contingency plan according to claim 4, characterized in that, The constraints include at least one of resource availability and temporal dependencies between tasks.
6. The method for optimizing the execution of a digital contingency plan according to claim 1, characterized in that, Before generating the optimal order of disaster relief tasks in the current emergency plan, the process also includes confirming the emergency plan corresponding to the current disaster.
7. The method for optimizing the execution of a digital contingency plan according to claim 2, characterized in that, When confirming an emergency response plan that matches the current disaster, this includes: Obtain real-time alarm information within the park; Calculate the similarity between the real-time alarm information and each emergency plan in the contingency plan database; Based on the similarity, an emergency plan matching the current disaster is identified.
8. A digital contingency plan execution optimization device, characterized in that, Used to generate the optimal order of disaster relief tasks in the event of a disaster in a smart park, including: The data acquisition module is used to acquire the current emergency plan and various resources in the current park; The task list generation module is used to break down the current emergency plan and generate a task list containing at least two execution tasks. The optimal execution path generation module is used to generate the optimal execution path of the emergency plan based on the task list and each of the resources. The optimal execution path is the current optimal order of execution of disaster relief tasks.
9. A digital contingency plan execution optimization device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 7 when executing the executable instructions.
10. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.