Digital plan construction method for emergency command decision

Abstract

The invention belongs to the technical field of emergency command decisions, and particularly relates to an emergency command decision-oriented digital plan construction method, which comprises the following steps of: 1, establishing a plan database; 2, setting an entity and parameter identification and extraction method, and setting a method for judging an emergency scene and key parameters according to emergency response conditions after the entity and the parameters are obtained; 3, generating an emergency command flow; 4, constructing a digital plan model; 5, making an emergency command decision to form a plan; 6, implementing an emergency command decision; and 7, evaluating and optimizing the emergency command decision. According to the method, on one hand, the influence weight of the risk corresponding to the processing task is calculated through the entropy method, on the other hand, the loss classification model is trained through the historical record data, and the influence of the sequential emergency task and the related risk is combined in the training process.

Description

Technical Field

[0001] The invention belongs to the technical field of emergency command decision-making, and specifically relates to a digital plan construction method for emergency command decision-making. Background Art

[0002] In the field of emergency management, emergency command decision-making is a complex and urgent task. Traditional manual paper plans have many problems, such as difficulty in updating and untimely information transmission. With the development of digital technology, it has become a trend to use digital technology to build and manage emergency plans. However, there are still problems in the existing technology on how to quickly form the best emergency command decision plan based on the acquired accident description. Generally speaking, the emergency command decision plan mainly includes various emergency tasks based on decision-making, but for different accidents and different emergency scenarios, the emergency tasks implemented are diverse. Some existing technologies directly use deep learning models combined with historical record data to study the relationship between plans and results, thereby realizing the selection and optimization of plans. However, emergency tasks are not all directly used to prevent risks, and historical record data rarely records various risks. Usually, only the final accident loss is focused on. In addition, the historical record data of accidents that actually occurred and were recorded is limited. Therefore, based on the limited historical record data, it is difficult for the existing technology to accurately judge whether the emergency plan formed in an orderly manner by multiple emergency handling tasks is the one with the least loss and risk. At present, there is still a lack of digital plan construction methods that can truly effectively and accurately automatically provide a list of the best plans. Summary of the invention

[0003] The purpose of the present invention is to provide a digital plan construction method for emergency command decision-making, which is used to solve the technical problem that it is difficult to accurately judge whether an emergency plan is the one with the least loss and risk based on limited historical record data in the prior art, and thus there is a lack of an effective and accurate digital plan construction method that automatically provides an optimal plan list.

[0004] The method for constructing a digital plan for emergency command decision-making includes the following steps.

[0005] Step 1: Establish a plan database; the plan database includes basic plan information, emergency response conditions, emergency resource information, emergency organization information, risk prevention information, disposal task information and emergency support information.

[0006] Step 2: Set the entity and parameter recognition and extraction method, and set the method for determining the emergency scenario and key parameters according to the emergency response conditions after obtaining the entities and parameters.

[0007] Step 3: Generate emergency command process; the emergency command process includes task allocation, information transmission, and resource scheduling.

[0008] Step 4: Build a digital plan model; combine the plan database, emergency scenarios and emergency command process to build a digital plan model.

[0009] Step 5: Make emergency command decisions to form plans; emergency command strategies, including resource scheduling strategies, decision support strategies, and risk prevention strategies; the plan is optimized and supported by evaluation algorithms, relevant risks are determined based on the accident type, and the impact weights of various risks are determined according to the emergency scenario, so as to obtain accurate risk assessment results; at the same time, considering the accident losses, emergency tasks and corresponding prevented risks are time-sequenced, risks containing impact weights and emergency tasks are combined, and loss classification models are constructed and trained, so as to realize the prediction and evaluation of accident losses and risks caused by the plan, and the plan is selected and optimized based on the evaluation obtained.

[0010] Step 6: Implement emergency command decisions.

[0011] Step 7: Evaluate and optimize emergency command decisions.

[0012] Preferably, in step 5, the relationship between the task and the risk has been described by the risk prevention information. After the emergency plan is formed, the description of the corresponding emergency scenario is obtained: S=<T,R,P> , where S represents the emergency scenario and T represents each emergency task t i The set of T=(t 1 ,t 2 ,…,t n ), T also includes network establishment tasks and resource scheduling tasks, and R represents various risks r i The set of R = (r 1 ,r 2 ,…,r m ), P represents the set of parameters related to emergency tasks, and P = (p 1 ,p 2 ,…,p o ), n≥o>m; the entropy method is used to determine the impact weights of various risks according to the emergency scenarios, so as to obtain accurate evaluation results.

[0013] Preferably, the entropy method includes: for a number of samples obtained from historical data, the j-th risk of the i-th sample is x ij , i=1,2,…,l,j=1,2,…,m; the risk index is:

[0014]

[0015] Then calculate the risk indicator element yij in the standardized matrix:

[0016]

[0017] y ij ∈[0,1] This completes the standardization of risk indicators; calculate the information entropy corresponding to the j-th risk:

[0018]

[0019] Further calculation is performed to obtain the impact weight w of the j-th risk j :

[0020]

[0021] Evaluate each emergency command decision-making plan by calculating the risks and their impact weights.

[0022] Preferably, the historical record data is a record of the emergency command decisions that have been implemented and their results, including emergency command decisions including emergency scenarios, task allocation, information transmission, and resource scheduling, and the corresponding accident loss classification. After extracting emergency tasks and parameters related to emergency tasks from the historical record data through the recognition model, the identified emergency tasks are time-series according to the time parameters, and then the corresponding risks are time-series combined with the impact weights to obtain two time-series feature sequences T', R', R'={r' 1 ,r' 2 ,…,r' n}, T'={t 1 ,t 2 ,…,t n}, and there is Then, the task feature sequence and the risk feature sequence are fused to obtain a new merged sequence feature C', which forms the input of the loss classification model. The BILSTM model is used to construct the loss classification model, and the classification results of the accident losses are used as the output. The recognition results of the historical record data are used as the training set to train the loss classification model.

[0023] Preferably, accident losses are divided into personal losses and property losses. When grading, the personal losses and property losses in the historical record data are first graded according to the accident level definition to obtain the grading results, and then the two grading results are combined to determine the final grading result. In the process of combining, the highest level in the grading results is taken as the final result.

[0024] Preferably, in step 2, the BERT-BILSTM-CRF model is used to extract entities and parameters, and the entity information is normalized; the pre-trained BERT model is used as the first layer to extract the semantic representation in the text information and obtain the parameter content, and the second layer of the BERT-BILSTM-CRF model is the BILSTM layer, and the text sequence information is processed from both the forward and reverse directions through the BILSTM layer to strengthen the position information and direction information of the text sequence; the last third layer uses the CRF layer as a decoder to obtain the global optimal entity label sequence to achieve entity extraction, and then the obtained entity is normatively described according to the set entity description specification; the emergency response conditions in the plan database are judged, and the accident is judged as the corresponding emergency scenario after the corresponding emergency response conditions are met, and the relevant key parameters are determined from the extracted parameters according to the emergency scenario.

[0025] Preferably, in step 3, the emergency command process is to set the contents of each plan including task allocation, information transmission and resource scheduling and the corresponding construction order. The emergency command process is generated and displayed using a graphical interface, and information transmission and resource scheduling are set according to the results of task allocation; in step 4, the digital plan model is constructed using an object-oriented programming method to achieve the association and interaction between the plan database, emergency scenarios and emergency command process; and the order in which the digital plan model is constructed is based on the emergency command process.

[0026] Preferably, in step 5, based on the digital plan model constructed in the previous step, the specific content of the plan is further determined in the order of risk prevention strategy, decision support strategy, and resource scheduling strategy to form an emergency command decision that can actually be executed; it should be noted that the execution order of the specific emergency tasks is opposite to the decision order.

[0027] The present invention has the following advantages: the present invention constructs the relationship between the disposal task and the risk in the accident, and realizes the time-series construction of different disposal tasks for the accident through the digital plan model, so as to form an emergency command decision plan including task allocation, information transmission, resource scheduling and other contents. In this method, on the one hand, the entropy method is used to calculate the impact weight of the risk corresponding to the disposal task, so as to take the corresponding plan. The risk involved and the impact of the risk, on the other hand, the BERT-BILSTM-CRF model is used to recognize and calculate the disposal task and loss classification in the historical record data, thereby realizing the training of the loss classification model, and combining the influence of the time-series emergency tasks and related risks during the training process, so that the impact of the plan on the accident loss can be more accurately evaluated, that is, the accident loss and the risk that may be caused by the plan are predicted, thereby optimizing the plan and reducing the loss caused by the emergency command decision by selecting the plan with the lowest accident loss level and the lowest risk.

[0028] At the same time, the present invention uses digital technology to build and manage emergency plans, which can improve the efficiency and accuracy of emergency command decisions, reduce the risk of human errors and delays, and improve emergency response capabilities and decision-making and command effects. The plan database can be stored and managed using a relational database or a distributed database. The emergency command process can also be generated and displayed using a graphical interface to facilitate user understanding and operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The present invention is a flowchart of a method for constructing a digital emergency plan for emergency command decision-making.

[0030] Figure 2 This is a diagram showing the structure of the plan database in the present invention.

[0031] Figure 3 Schematic diagram of the BERT-BILSTM-CRF model in the present invention. DETAILED DESCRIPTION

[0032] The specific implementation modes of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, so as to help those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention.

[0033] like Figure 1-Figure 3 As shown, the present invention provides a digital plan construction method for emergency command decision-making, comprising the following steps:

[0034] Step 1: Establish a plan database, which includes basic plan information, emergency response conditions, emergency resource information, emergency organization information, risk prevention information, disposal task information and emergency support information.

[0035] Among them, the basic information of the plan is the basic information required to build a plan for emergency command decision-making, such as the plan number, etc.; the emergency response conditions refer to the specific conditions used to judge the accident when judging the emergency scenario, such as the entities that must be included in the event, whether the threshold range of the collected parameters in the accident is met, etc.; the emergency resource information is the relevant information used for emergency resources, such as the resource name, quantity and storage location, etc.; the emergency organization information is the relevant information of the emergency organization that performs the emergency task, such as the organization name, organization members, executable tasks, etc.; the risk prevention information refers to the relevant information of the risks that need to be prevented when responding to different types of accidents, such as the risk name, involved accidents, and the corresponding disposal tasks for risk prevention, etc.; the disposal task refers to the relevant information of the disposal task used in the emergency disposal, such as the task name, task content, and resources required for the task, etc.; the emergency guarantee information refers to the relevant information required to be guaranteed before the emergency disposal, such as the confirmation of emergency resources and the confirmation of the members of the emergency organization, etc.

[0036] Step 2: Set the entity and parameter identification and extraction method, and set the method for determining the emergency scenario and key parameters according to the emergency response conditions after obtaining the entity and parameters;

[0037] This step uses the BERT-BILSTM-CRF model to extract entities and parameters, and normalize entity information. This method uses the pre-trained BERT (Bidirectional Encoder Representation) model as the first layer to extract semantic representations from text information, and can also obtain parameter content. The second layer of this method is the BILSTM (Bidirectional Long Short-Term Memory Network) layer, which processes text sequence information from both forward and reverse directions through the BILSTM layer to strengthen the position and direction information of the text sequence. The final third layer uses the CRF layer as a decoder to obtain the global optimal entity label sequence to achieve entity extraction, and then the obtained entities are described normatively according to the set entity description specification.

[0038] For the extracted entities and parameters, further judgments are made based on the emergency response conditions in the plan database. When the corresponding emergency response conditions are met, the accident is judged as the corresponding emergency scenario, and the relevant key parameters are determined from the extracted parameters according to the emergency scenario. In practical applications, this solution uses sensors, monitoring equipment, etc. to obtain real-time data of emergency scenarios, analyzes and determines key parameters based on the data, and then combines the description text of the reported accident to extract entities and parameters through the BERT-BILSTM-CRF model.

[0039] Step 3: Generate an emergency command process. The emergency command process is used to determine the emergency command plan process based on the emergency scenario and plan database. The command plan process includes task allocation, information transmission, resource scheduling, etc.

[0040] Task allocation is to retrieve relevant handling task information and emergency organization information according to the emergency scenario, and assign the corresponding handling tasks to the corresponding emergency organization according to the rules, while also including necessary information and parameters such as location and time.

[0041] Information transmission is based on the setting of corresponding communication networks and network establishment tasks according to the emergency scenarios. The communication network terminology specifies the communication relationship between different emergency organizations, and the network establishment tasks are assigned to the emergency organizations that establish the communication networks.

[0042] Resource scheduling is to retrieve emergency resource information based on emergency tasks, determine the emergency resources required for each emergency task and generate corresponding resource scheduling tasks. Resource scheduling tasks are assigned to emergency organizations that perform resource scheduling. Resource scheduling tasks include necessary information and parameters such as location and time provided by the post-assignment disposal tasks.

[0043] The emergency command process is to set the contents of the various plans for task allocation, information transmission and resource scheduling and the corresponding construction order. This order takes into account that when building the model, information transmission and resource scheduling should be set according to the results of task allocation. The emergency command process in this step is generated and displayed using a graphical interface to facilitate user understanding and operation.

[0044] Step 4: Build a digital emergency plan model by combining the emergency plan database, emergency scenarios, and emergency command processes to build a digital emergency plan model;

[0045] In this step, the digital emergency plan model is constructed using object-oriented programming methods to achieve the association and interaction between the emergency plan database, emergency scenarios, and emergency command processes. The order of model construction of the digital emergency plan model is in the order of task allocation, information transmission, and resource scheduling. The disposal tasks, emergency organizations, and emergency resources used are obtained from the disposal task information, emergency organization information, and emergency resource information, respectively.

[0046] Step 5: Make emergency command decisions and form plans. According to the digital plan model and actual needs, formulate emergency command strategies, including risk prevention strategies, decision support strategies, resource scheduling strategies, etc.

[0047] When setting the risk prevention strategy, it is necessary to retrieve the risk prevention information, which is used to describe the relationship between the disposal task and the various risks in the corresponding emergency scenario. The various risks can be used as evaluation indicators for subsequent emergency command decisions. The risk prevention strategy is used to determine the strategy for implementing task allocation for the corresponding disposal task from the disposal task information according to the various risks in the corresponding emergency scenario.

[0048] The decision support strategy is used to determine how to build an information transmission network, so as to establish a command center based on the information transmission network to exchange information with the corresponding emergency organizations and implement the decision support strategy.

[0049] Resource scheduling policy is used to determine how to allocate resources.

[0050] In this step, based on the digital plan model constructed in the previous step, the specific content of the plan is further determined in the order of risk prevention strategy, decision support strategy, and resource scheduling strategy to form an emergency command decision that can actually be executed. It should be noted that the execution order of the specific emergency tasks is opposite to the decision order, because when executing the specific tasks, it is necessary to complete resource scheduling first, then realize the construction of information transmission, and finally let the emergency organization complete the relevant emergency tasks according to the results of information transmission.

[0051] The emergency command decision-making plan obtained in this step should be optimized and supported by the evaluation algorithm. Since the relationship between tasks and risks has been described through risk prevention information, after the plan is formed, the description of the corresponding emergency scenario can be obtained: S =<T,R,P> , where S represents the emergency scenario and T represents each emergency task t i The set of T=(t 1 ,t 2 ,…,t n ), T also includes network establishment tasks and resource scheduling tasks, and R represents various risks r i The set of R = (r 1 ,r 2 ,…,r m ), P represents the set of parameters related to emergency tasks, and P = (p 1 ,p 2 ,…,p o ), n≥o>m. In this scheme, one disposal task corresponds to preventing one or zero risks, and there are two or more disposal tasks for preventing the same risk; each risk r i The set R is an evaluation index for evaluating emergency command decision-making plans. However, due to the different impacts of various risks in different emergency scenarios, it is also necessary to determine the impact weights of various risks according to the emergency scenarios in order to obtain accurate risk assessment results.

[0052] This step uses the entropy method to determine the impact weight. Therefore, for several samples obtained from historical data (corresponding to emergency scenario S), the j-th risk of the i-th sample is x ij , i = 1, 2, ..., l, j = 1, 2, ..., m. Since the risk is a negatively correlated indicator for the evaluation of emergency command decision-making plans, the risk index is:

[0053]

[0054] Then calculate the risk indicator element yij in the standardized matrix:

[0055]

[0056] y ij ∈[0,1] thus completing the standardization of risk indicators.

[0057] Calculate the information entropy corresponding to the j-th risk:

[0058]

[0059] Further calculation is performed to obtain the impact weight w of the j-th risk j :

[0060]

[0061] By calculating the risks and their impact weights, we can better evaluate the plans of various emergency command decisions, and then select the best one from the feasible plans to achieve the optimization of the emergency command decision plans.

[0062] In addition to risks, the evaluation indicators should also consider accident losses. Although different types of accidents face different risks, the losses caused by accidents are similar. Generally speaking, accident losses can be divided into personnel losses and property losses. When grading, the personnel losses and property losses in the historical records are first graded according to the accident level definition to obtain the grading results, and then the two grading results are combined to determine the final grading result. In the process of combining, the highest grade in the grading results is used as the final result. In order to predict the loss level generated according to the plan, a loss classification model is constructed. Since the input of the loss classification model is a time-series feature, the loss classification model is trained using the BILSTM model.

[0063] Before training, the emergency command decision-making data and the historical records of the corresponding accident loss classification, which record the emergency scenarios, task allocation, information transmission, resource scheduling, etc., are input into the trained recognition model. The recognition model can use the BERT-BILSTM-CRF model trained in step 2. After the recognition model extracts the emergency tasks and parameters related to the emergency tasks, it time-series the identified emergency tasks according to the time parameters, and then time-series the corresponding risks combined with the impact weights to obtain two time-series feature sequences T', R', R'={r' 1 ,r' 2 ,…,r' n}, T'={t 1 ,t 2 ,…,tn}, and there is Then, the task feature sequence and the risk feature sequence are fused to obtain a new combined sequence feature C', which forms the input of the loss classification model. The BILSTM model is used to construct the loss classification model, with the combined sequence feature C' as input and the classification result of the accident loss as output. The recognition result of the historical record data is used as the training set to train the loss classification model. In this way, the loss classification model can simultaneously take the emergency tasks and related risks in the plan into consideration, and the impact weight of the risk in the corresponding accident is also taken into account. Therefore, the loss classification model obtained by this training can better determine the impact of different plans on accident losses. For several plans with the same classification results, they are arranged in order from small to large risks for the commander to choose.

[0064] Step 6: Implement emergency command decisions.

[0065] After an accident occurs, this step first identifies the corresponding emergency scenario according to the above steps, then generates several emergency command decision plans through the digital plan model, optimizes the generated plans through the evaluation algorithm, predicts the accident classification results and sorts them according to the risk size, and sorts the plans with the lowest classification results and provides them to the decision maker. The decision maker selects the plan according to the emergency scenario and implements the emergency command decision based on the plan. This step uses large-screen display, voice commands, and other methods to implement emergency command decisions.

[0066] Step 7: Evaluate and optimize emergency command decisions to improve decision-making efficiency and accuracy.

[0067] After the actual implementation of the emergency command decision, the relevant plans and the accident losses after implementation are recorded to form new historical record data, which are added to the training set for further updating the loss classification model, thereby realizing the evaluation and optimization of the emergency command decision plan.

[0068] The present invention is described above by way of example in conjunction with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the inventive concept and technical solution of the present invention, or the inventive concept and technical solution are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A digital plan construction method for emergency command decision-making, characterized by: The following steps are included: Step 1: Establish a plan database; the plan database includes basic plan information, emergency response conditions, emergency resource information, emergency organization information, risk prevention information, disposal task information and emergency support information; Step 2: Set the entity and parameter identification and extraction method, and set the method for determining the emergency scenario and key parameters according to the emergency response conditions after obtaining the entity and parameters; Step 3: Generate emergency command process; the emergency command process includes task allocation, information transmission, and resource scheduling; Step 4: Build a digital emergency plan model; combine the emergency plan database, emergency scenarios and emergency command process to build a digital emergency plan model; Step 5: Make emergency command decisions to form plans; emergency command strategies, including resource scheduling strategies, decision support strategies, and risk prevention strategies; the plan is optimized and supported by evaluation algorithms, and the relevant risks are determined based on the accident type, and the impact weights of various risks are determined according to the emergency scenario, so as to obtain accurate risk assessment results; at the same time, considering the accident losses, the emergency tasks and the corresponding prevented risks are time-sequenced, the risks with impact weights and emergency tasks are combined, and the loss classification model is constructed and trained, so as to realize the prediction and evaluation of the accident losses and risks caused by the plan, and the plan is selected and optimized based on the evaluation obtained; Step 6: Implement emergency command decisions; Step 7: Evaluate and optimize emergency command decisions.

2. A digital plan construction method for emergency command decision-making according to claim 1, characterized in that: In step 5, the relationship between tasks and risks has been described through risk prevention information. After the emergency plan is formed, the description of the corresponding emergency scenario is obtained: S =<T,R,P> , where S represents the emergency scenario and T represents each emergency task t i The set of T = (t1, t2, ..., t n ), T also includes network establishment tasks and resource scheduling tasks, and R represents various risks r i The set of R=(r1,r2,…,r m ), P represents the set of parameters related to the emergency task, and P = (p1, p2, ..., p o ), n≥o>m; the entropy method is used to determine the impact weights of various risks according to the emergency scenarios, so as to obtain accurate evaluation results.

3. A digital plan construction method for emergency command decision-making according to claim 2, characterized in that: The entropy value method includes: for a number of samples obtained from historical data, the j-th risk of the i-th sample is x ij , i=1,2,…,l,j=1,2,…,m; the risk index is: Then calculate the risk indicator element yij in the standardized matrix: y ij ∈[0,1] This completes the standardization of risk indicators; calculate the information entropy corresponding to the j-th risk: Further calculation is performed to obtain the impact weight w of the j-th risk j : Evaluate each emergency command decision-making plan by calculating the risks and their impact weights.

4. A digital plan construction method for emergency command decision-making according to claim 3, characterized in that: The historical record data is a record of the emergency command decisions that have been implemented and their results, including emergency command decisions and corresponding accident loss classification, including emergency scenarios, task allocation, information transmission, and resource scheduling. After extracting emergency tasks and parameters related to emergency tasks from the historical record data through the recognition model, the identified emergency tasks are time-series according to the time parameters, and then the corresponding risks are time-series combined with the impact weights to obtain two time-series feature sequences T', R', R'={r'1,r'2,…,r' n }, T'={t1,t2,…,t n }, and there is Then, the task feature sequence and the risk feature sequence are fused to obtain a new merged sequence feature C', which forms the input of the loss classification model. The BILSTM model is used to construct the loss classification model, and the classification results of the accident losses are used as the output. The recognition results of the historical record data are used as the training set to train the loss classification model.

5. A digital plan construction method for emergency command decision-making according to claim 4, characterized in that: Accident losses are divided into personal losses and property losses. When classifying, the personal losses and property losses in the historical record data are first classified according to the accident level definition to obtain the classification results, and then the two classification results are combined to determine the final classification result. In the process of combining, the highest level in the classification results is taken as the final result.

6. A digital plan construction method for emergency command decision-making according to any one of claims 1 to 5, characterized in that: In the step 2, the BERT-BILSTM-CRF model is used to extract entities and parameters, and the entity information is normalized; the pre-trained BERT model is used as the first layer to extract the semantic representation in the text information and obtain the parameter content, and the second layer of the BERT-BILSTM-CRF model is the BILSTM layer, which processes the text sequence information from both the forward and reverse directions at the same time, and strengthens the position information and direction information of the text sequence; the last third layer uses the CRF layer as a decoder to obtain the global optimal entity label sequence to achieve entity extraction, and then the obtained entity is normatively described according to the set entity description specification; the emergency response conditions in the plan database are judged, and the accident is judged as the corresponding emergency scenario after the corresponding emergency response conditions are met, and the relevant key parameters are determined from the extracted parameters according to the emergency scenario.

7. A digital plan construction method for emergency command decision-making according to any one of claims 1 to 5, characterized in that: In step 3, the emergency command process is to set the contents of the plans for task allocation, information transmission and resource scheduling and the corresponding construction order. The emergency command process is generated and displayed using a graphical interface, and information transmission and resource scheduling are set according to the results of task allocation; in step 4, the digital plan model is constructed using an object-oriented programming method to achieve the association and interaction between the plan database, emergency scenarios and emergency command processes; and the order in which the digital plan model is constructed is based on the emergency command process.

8. A digital plan construction method for emergency command decision-making according to any one of claims 1 to 5, characterized in that: In step 5, based on the digital plan model constructed in the previous step, the specific content of the plan is further determined in the order of risk prevention strategy, decision support strategy, and resource scheduling strategy to form an emergency command decision that can actually be executed; it should be noted that the execution order of the specific emergency tasks is opposite to the decision order.

Images