Resource scheduling method and device under natural disaster, medium and equipment

Abstract

The invention discloses a resource scheduling method and device under natural disasters, a medium and equipment, and relates to the field of medical insurance, and the method comprises the steps: obtaining effective feature data corresponding to the natural disasters currently occurring in a to-be-scheduled region, inputting the effective feature data into a report prediction model, and obtaining an insurance report prediction value corresponding to the natural disasters; performing rescue resource estimation based on the insurance case report prediction value, obtaining a natural disaster occurrence site from the effective feature data corresponding to the natural disaster, performing resource deployment based on the natural disaster occurrence site and resource estimation information, and generating resource configuration information; after an insurance case reporting request is received, resource scheduling planning is carried out according to the insurance case reporting request and the resource configuration information, and a rescue scheduling strategy is generated. According to the method, a proper amount of rescue resources are prepared in advance, the strain capacity of an insurance company under natural disasters is improved, optimal path planning is performed based on the resource configuration information and the rescue site, and the rescue rationality of an insurance mechanism is improved.

Description

Technical Field

[0001] The present application relates to the field of medical insurance technology, and in particular to a method, device, medium, and equipment for resource scheduling during natural disasters. Background Art

[0002] Medical rescue agencies and insurance companies, as crucial components of emergency rescue dispatch, play an indispensable role in disaster response. However, when faced with sudden, large-scale natural disasters, insurance companies' local rescue resources often struggle to meet the surging demand. This not only limits the rescue capabilities of local insurance agencies but also impacts the overall efficiency of rescue efforts. Therefore, scientific and rational resource dispatch is crucial.

[0003] Currently, insurance companies' rescue services primarily rely on the actual number of insurance claims received to allocate resources. During major disasters, especially in the early stages of a rescue effort, the number of actual claims received is relatively low. Allocation of resources based on these initial claims cannot meet the needs of the entire rescue period. Rescheduling resources based on these needs during the rescue period can severely impact rescue progress. Preparing a large amount of resources in the early stages of a rescue effort can lead to waste. Therefore, existing insurance companies' resource allocation methods fail to scientifically and rationally allocate rescue supplies. Summary of the Invention

[0004] In view of this, the present application provides a method, device, medium and equipment for resource scheduling under natural disasters, which solves the problem that the resource scheduling methods of existing insurance companies cannot perform resource scheduling scientifically and reasonably.

[0005] According to one aspect of the present application, a method for resource scheduling under natural disasters is provided, the method comprising:

[0006] Obtain valid feature data corresponding to the natural disaster currently occurring in the area to be dispatched, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster;

[0007] Based on the insurance report prediction value, rescue resources are estimated to obtain resource estimation information, the location of the natural disaster is obtained from the valid feature data corresponding to the natural disaster, resources are deployed based on the natural disaster location and the resource estimation information, and resource allocation information is generated;

[0008] After receiving the insurance report request, resource scheduling planning is performed according to the insurance report request and the resource configuration information to generate a rescue scheduling strategy.

[0009] Optionally, obtaining valid feature data corresponding to the natural disaster currently occurring in the area to be scheduled includes:

[0010] Acquiring a variety of real-time related information corresponding to a natural disaster within a preset period of time after the natural disaster occurs in the area to be dispatched;

[0011] fusing the multiple real-time related information to obtain fused information, and extracting initial feature data from the fused information, wherein the initial feature data includes the time of occurrence of the natural disaster, the location of the natural disaster, the type of natural disaster, the level of the natural disaster, and the extent of damage;

[0012] The initial feature data is preprocessed to obtain effective feature data.

[0013] Optionally, performing resource scheduling planning according to the insurance report request and the resource configuration information to generate a rescue scheduling strategy includes:

[0014] Analyze the insurance report request and obtain rescue addresses and resource requirements corresponding to multiple rescue tasks from the analysis results;

[0015] Based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, a simulated annealing algorithm is used to plan a rescue path to obtain a rescue scheduling path, and a rescue scheduling strategy is generated based on the rescue scheduling path.

[0016] Optionally, the rescue path planning is performed using a simulated annealing algorithm based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue task to obtain a rescue dispatch path, including:

[0017] Based on the shortest rescue time and its corresponding weight and the shortest rescue path and its corresponding weight, an objective function is constructed, and a solution space is constructed according to a preset data structure, wherein the solution space includes resource allocation information, the execution order of rescue tasks, and the driving route;

[0018] Setting initial parameters in a simulated annealing algorithm, generating an initial solution based on the solution space, the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, and using the initial solution as the current solution;

[0019] Calculate a first shortest rescue time and a first shortest rescue path corresponding to the initial solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, and obtain a first evaluation value corresponding to the current solution;

[0020] modifying the current solution according to a first preset rule to obtain a modified solution, calculating a second shortest rescue time and a second shortest rescue path corresponding to the modified solution, substituting the second shortest rescue time and the second shortest rescue path into the objective function to obtain a second evaluation value corresponding to the modified solution;

[0021] Compare the second evaluation value with the first evaluation value, and obtain a new current solution and a new modified solution based on the comparison result. Compare the second evaluation value corresponding to the new modified solution with the first evaluation value corresponding to the new current solution until the iteration condition is met. Take the modified solution at this time as the optimal solution, and take the optimal solution as the rescue scheduling path.

[0022] Optionally, obtaining a new current solution and a new modified solution according to the comparison result includes:

[0023] When the second evaluation value is less than the first evaluation value, taking the modified solution as the new current solution, and modifying the new current solution according to the first preset rule to obtain a new modified solution;

[0024] When the second evaluation value is greater than the first evaluation value, the current solution remains unchanged, and the current solution is modified according to a second preset rule to obtain a new modified solution.

[0025] Optionally, after generating the rescue dispatch strategy based on the rescue dispatch path, the method further includes:

[0026] Obtaining rescue progress, resource consumption, and real-time traffic information during the rescue process, and analyzing the rescue scheduling strategy based on the rescue progress, resource consumption, and real-time traffic information;

[0027] When it is determined based on the analysis results that the rescue dispatch strategy is consistent with the current rescue status, the rescue dispatch path is fine-tuned according to the resource consumption and real-time traffic information, and the rescue dispatch strategy is modified based on the adjusted content;

[0028] When it is determined that the rescue scheduling strategy does not conform to the current rescue status, the latest rescue address, the latest material demand and the latest resource configuration information under the current rescue status are obtained, and according to the latest rescue address, the latest resource demand, the latest resource configuration information and the real-time road condition information, the rescue path is re-planned using a simulated annealing algorithm to obtain a new rescue scheduling path, and a new rescue scheduling strategy is generated based on the new rescue scheduling path.

[0029] Optionally, the trained case reporting prediction model is obtained by the following method, including:

[0030] Obtain various historical related information corresponding to historical natural disasters;

[0031] fusing the plurality of historical related information, extracting historical initial feature data from the fused historical related information, and preprocessing the historical initial feature data to obtain historical valid feature data;

[0032] The number of insurance reports corresponding to the historical natural disasters is obtained, the historical valid feature data and the number of insurance reports are used as training data, an initial report prediction model is trained based on the training data, and a trained report prediction model is obtained.

[0033] According to another aspect of the present application, a resource scheduling device under natural disasters is provided, the device comprising:

[0034] An insurance report prediction module is used to obtain valid feature data corresponding to the natural disaster currently occurring in the dispatch area, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster;

[0035] a resource allocation module, configured to estimate rescue resources based on the insurance report prediction value, obtain resource estimation information, obtain the location of the natural disaster from the valid characteristic data corresponding to the natural disaster, deploy resources based on the natural disaster location and the resource estimation information, and generate resource allocation information;

[0036] The scheduling strategy generation module is used to receive an insurance report request, perform resource scheduling planning according to the insurance report request and the resource configuration information, and generate a rescue scheduling strategy.

[0037] Optionally, the insurance claim prediction module is further configured to:

[0038] Acquiring a variety of real-time related information corresponding to a natural disaster within a preset period of time after the natural disaster occurs in the area to be dispatched;

[0039] fusing the multiple real-time related information to obtain fused information, and extracting initial feature data from the fused information, wherein the initial feature data includes the time of occurrence of the natural disaster, the location of the natural disaster, the type of natural disaster, the level of the natural disaster, and the extent of damage;

[0040] The initial feature data is preprocessed to obtain effective feature data.

[0041] Optionally, the scheduling strategy generation module is further configured to:

[0042] Analyze the insurance report request and obtain rescue addresses and resource requirements corresponding to multiple rescue tasks from the analysis results;

[0043] Based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, a simulated annealing algorithm is used to plan a rescue path to obtain a rescue scheduling path, and a rescue scheduling strategy is generated based on the rescue scheduling path.

[0044] Optionally, the scheduling strategy generation module is further configured to:

[0045] Based on the shortest rescue time and its corresponding weight and the shortest rescue path and its corresponding weight, an objective function is constructed, and a solution space is constructed according to a preset data structure, wherein the solution space includes resource allocation information, the execution order of rescue tasks, and the driving route;

[0046] Setting initial parameters in a simulated annealing algorithm, generating an initial solution based on the solution space, the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, and using the initial solution as the current solution;

[0047] Calculate a first shortest rescue time and a first shortest rescue path corresponding to the initial solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, and obtain a first evaluation value corresponding to the current solution;

[0048] modifying the current solution according to a first preset rule to obtain a modified solution, calculating a second shortest rescue time and a second shortest rescue path corresponding to the modified solution, substituting the second shortest rescue time and the second shortest rescue path into the objective function to obtain a second evaluation value corresponding to the modified solution;

[0049] Compare the second evaluation value with the first evaluation value, and obtain a new current solution and a new modified solution based on the comparison result. Compare the second evaluation value corresponding to the new modified solution with the first evaluation value corresponding to the new current solution until the iteration condition is met. Take the modified solution at this time as the optimal solution, and take the optimal solution as the rescue scheduling path.

[0050] Optionally, the scheduling strategy generation module is further configured to:

[0051] When the second evaluation value is less than the first evaluation value, taking the modified solution as the new current solution, and modifying the new current solution according to the first preset rule to obtain a new modified solution;

[0052] When the second evaluation value is greater than the first evaluation value, the current solution remains unchanged, and the current solution is modified according to a second preset rule to obtain a new modified solution.

[0053] Optionally, the scheduling strategy generation module is further configured to:

[0054] Obtaining rescue progress, resource consumption, and real-time traffic information during the rescue process, and analyzing the rescue scheduling strategy based on the rescue progress, resource consumption, and real-time traffic information;

[0055] When it is determined based on the analysis results that the rescue dispatch strategy is consistent with the current rescue status, the rescue dispatch path is fine-tuned according to the resource consumption and real-time traffic information, and the rescue dispatch strategy is modified based on the adjusted content;

[0056] When it is determined that the rescue scheduling strategy does not conform to the current rescue status, the latest rescue address, the latest material demand and the latest resource configuration information under the current rescue status are obtained, and according to the latest rescue address, the latest resource demand, the latest resource configuration information and the real-time road condition information, the rescue path is re-planned using a simulated annealing algorithm to obtain a new rescue scheduling path, and a new rescue scheduling strategy is generated based on the new rescue scheduling path.

[0057] Optionally, the resource scheduling device under natural disasters further includes:

[0058] The model training module is used to obtain a variety of historical related information corresponding to historical natural disasters; fuse the multiple historical related information, extract historical initial feature data from the fused historical related information, pre-process the historical initial feature data, and obtain historical valid feature data; obtain the number of insurance reports corresponding to the historical natural disasters, use the historical valid feature data and the number of insurance reports as training data, and train the initial report prediction model based on the training data to obtain a trained report prediction model.

[0059] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the method for resource scheduling under natural disasters described above is implemented.

[0060] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned resource scheduling method under natural disasters when executing the program.

[0061] By means of the above technical scheme, the present application provides a method, device, medium and equipment for resource scheduling under natural disasters. When performing rescue scheduling, insurance report prediction is performed based on the effective characteristic data corresponding to the natural disasters currently occurring in the scheduling area, and the insurance report prediction is used as the basis for evaluating the required resources, and resources are deployed according to the evaluation results to generate resource allocation information to ensure that sufficient resources can be scheduled to respond to emergencies. After receiving the insurance report request, resource scheduling planning is performed according to the insurance report request and resource allocation information, and a rescue scheduling strategy is generated to ensure the effectiveness and rationality of the agency's rescue plan. This method can not only prepare sufficient rescue resources in advance, improve the response speed of insurance institutions in the face of natural disasters, but also accurately evaluate resource needs to avoid resource waste, and further enhance the insurance company's ability to respond to natural disasters. In addition, through resource allocation information and optimal path planning of the rescue location, the most effective and reasonable rescue scheduling strategy is generated, which improves the timeliness and rationality of the insurance institution's rescue.

[0062] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0064] Figure 1 A schematic diagram of a process for resource scheduling under natural disasters provided by an embodiment of the present application is shown;

[0065] Figure 2 Another schematic diagram of a process for resource scheduling under natural disasters provided by an embodiment of the present application is shown;

[0066] Figure 3 A schematic diagram of the structure of a resource scheduling device under natural disasters provided by an embodiment of the present application is shown;

[0067] Figure 4 A schematic diagram of the device structure of a computer device provided in an embodiment of the present application is shown.

[0068] in,

[0069] Figure 3 In: 302 - insurance claim prediction module; 304 - resource allocation module; 306 - scheduling strategy generation module. DETAILED DESCRIPTION

[0070] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0071] In this embodiment, a resource scheduling method under natural disasters is provided. Figure 1 As shown, the method includes:

[0072] Step 102: Obtain valid feature data corresponding to the natural disaster currently occurring in the dispatched area, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster;

[0073] Step 104: Estimate rescue resources based on the insurance report prediction value to obtain resource estimation information, obtain the location of the natural disaster from the valid feature data corresponding to the natural disaster, deploy resources based on the natural disaster location and resource estimation information, and generate resource allocation information;

[0074] Step 106: After receiving the insurance report request, perform resource scheduling planning based on the insurance report request and resource configuration information to generate a rescue scheduling strategy.

[0075] The embodiment of the present application is used to generate a rescue scheduling strategy after a natural disaster occurs.

[0076] First, relevant information about the natural disaster that has just occurred in the dispatch area is obtained from multiple channels, such as the time, location, level, degree of damage, etc., including text information, image information and voice information. Data preprocessing is performed on these relevant information, such as data cleaning, outlier processing, feature encoding, normalization, etc., to form effective feature data corresponding to the natural disaster. The effective feature data is input into a pre-trained report prediction model, and the model is used to predict the number of insurance reports that may be received for this natural disaster.

[0077] After receiving insurance claim forecasts, insurance institutions can use this information to pre-evaluate available human and material resources, develop contingency plans, prepare sufficient resources, determine the specific locations of schedulable resources, and generate resource allocation information based on resource deployment information. Upon receiving an insurance claim request, the rescue location is determined based on the claim request. The optimal rescue path is planned based on the resource allocation information and the rescue location. For example, a path planning model is used to find the optimal rescue path. A rescue dispatch strategy is generated based on the optimal rescue path and distributed to the rescue center that deploys resources, enabling the rescue center to implement the rescue mission according to the strategy. Resource allocation information is updated in real time during resource scheduling and rescue mission progress, facilitating the generation and implementation of subsequent rescue mission strategies.

[0078] The present application provides a resource scheduling method under natural disasters. Compared with the existing technology, when conducting rescue scheduling, insurance report prediction is performed based on the effective characteristic data corresponding to the natural disasters currently occurring in the scheduling area, and the insurance report prediction is used as the basis for evaluating the required resources. Resources are deployed according to the evaluation results, and resource allocation information is generated to ensure that sufficient resources can be dispatched when responding to emergencies. After receiving the insurance report request, resource scheduling planning is performed according to the insurance report request and resource allocation information, and a rescue scheduling strategy is generated to ensure the effectiveness and rationality of the agency's rescue plan. This method can not only prepare sufficient rescue resources in advance, but also improve the response speed of insurance institutions in the face of natural disasters, and can also accurately evaluate resource needs to avoid resource waste, further improving the insurance company's ability to respond to natural disasters. In addition, through resource allocation information and optimal path planning of the rescue location, the most effective and reasonable rescue scheduling strategy is generated, which improves the timeliness and rationality of the insurance institution's rescue.

[0079] In one embodiment, obtaining valid feature data corresponding to a natural disaster currently occurring in the area to be dispatched includes:

[0080] Obtain various real-time related information corresponding to the natural disaster within a preset period of time after the natural disaster occurs in the dispatch area;

[0081] A variety of real-time related information is integrated to obtain integrated information, and initial feature data is extracted from the integrated information, wherein the initial feature data includes the time of natural disaster occurrence, the location of natural disaster occurrence, the type of natural disaster, the level of natural disaster and the extent of damage;

[0082] The initial feature data is preprocessed to obtain effective feature data.

[0083] Specifically, relevant information about natural disasters is obtained from various channels, including news media and emergency rescue command centers. This information includes text, images, and voice information from news media and emergency rescue command centers, including the time, location, severity, and extent of damage, as well as on-site images of the natural disaster. This information is then fused, for example, by chronological order or by location. Initial feature data is then extracted from this fused information, including the specific time and location of the natural disaster, the type of natural disaster, the severity, and extent of damage. This initial feature data is then preprocessed, including data cleaning, outlier processing, feature encoding, and normalization, to generate valid feature data for the natural disaster. This processed valid feature data is then fed into a pre-trained report volume prediction model (based on an LSTM network) to predict the likely volume of insurance reports received for this natural disaster.

[0084] Furthermore, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another resource scheduling method under natural disasters is provided, such as Figure 2 As shown, the method includes:

[0085] Step 202: Obtain valid feature data corresponding to the natural disaster currently occurring in the area to be dispatched, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster;

[0086] Step 204: Estimate rescue resources based on the insurance report prediction value to obtain resource estimation information. Obtain the location of the natural disaster from the valid feature data corresponding to the natural disaster. Deploy resources based on the natural disaster location and resource estimation information to generate resource allocation information.

[0087] Step 206: After receiving the insurance report request, analyze the insurance report request and obtain the rescue addresses and resource requirements corresponding to the multiple rescue tasks from the analysis results;

[0088] Step 208: Based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue mission, a simulated annealing algorithm is used to plan the rescue path to obtain a rescue scheduling path, and a rescue scheduling strategy is generated based on the rescue scheduling path.

[0089] In this embodiment, the effective feature data of the natural disaster that has just occurred in the area to be scheduled is input into a pre-trained report volume prediction model. The model is used to predict the number of insurance reports that may be received for this natural disaster. Resources are scheduled and deployed based on the number of insurance reports, and resource configuration information is generated. After receiving the insurance report request, the insurance report request is analyzed to determine the rescue address and resource requirements corresponding to each rescue task. Based on the rescue address, resource requirements and resource configuration information, a simulated annealing algorithm is used to plan the rescue path to obtain the rescue scheduling path.

[0090] In this embodiment, the simulated annealing algorithm is used for rescue path planning. The specific method for obtaining the rescue dispatch path is as follows:

[0091] Based on the shortest rescue time and its corresponding weight and the shortest rescue path and its corresponding weight, an objective function is constructed, and a solution space is constructed according to a preset data structure, where the solution space includes resource allocation information, the execution order of rescue tasks, and the driving route;

[0092] Set the initial parameters in the simulated annealing algorithm, generate an initial solution based on the solution space, resource configuration information, the rescue address and resource requirements corresponding to each rescue mission, and use the initial solution as the current solution;

[0093] Calculate the first shortest rescue time and the first shortest rescue path corresponding to the initial solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, and obtain the first evaluation value corresponding to the current solution;

[0094] Modify the current solution according to the first preset rule to obtain a modified solution, calculate the second shortest rescue time and the second shortest rescue path corresponding to the modified solution, substitute the second shortest rescue time and the second shortest rescue path into the objective function, and obtain a second evaluation value corresponding to the modified solution;

[0095] Compare the second evaluation value with the first evaluation value. When the second evaluation value is less than the first evaluation value, take the modified solution as the new current solution, and modify the new current solution according to the first preset rule to obtain a new modified solution. When the second evaluation value is greater than the first evaluation value, the current solution remains unchanged, and modify the current solution according to the second preset rule to obtain a new modified solution. Compare the second evaluation value corresponding to the new modified solution with the first evaluation value corresponding to the new current solution until the iteration condition is met. Take the modified solution at this time as the optimal solution, and take the optimal solution as the rescue dispatch path.

[0096] Specifically, the specific method for implementing path planning based on the simulated annealing algorithm is as follows:

[0097] (1) Define the problem and goal

[0098] Problem definition: Determine resource requirements and rescue locations, such as the resources that need to be dispatched, such as rescue vehicles, rescue personnel, and rescue supplies, as well as the location and requirements of the disaster site.

[0099] Determine the goal: Define the optimization goal, and use the shortest rescue time and shortest rescue path as the weighted optimization goal.

[0100] (2) Constructing the solution space

[0101] Representation solution: Use appropriate data structures to represent the rescue dispatch path, including the allocation of rescue resources, rescue order, path selection, etc.

[0102] Initialization solution: Randomly generate an initial solution based on the rescue address, resource requirements, resource configuration information and the data structure of the solution space.

[0103] (3) Define the evaluation function

[0104] Design an evaluation function: Based on the optimization goal, design an evaluation function to evaluate the quality of the current solution. For example, the shortest rescue time and the shortest rescue path weighted as the optimization objective function. That is:

[0105]

[0106] Among them, x tis the shortest rescue time, x r The shortest rescue path.

[0107] (3) Initialize the simulated annealing algorithm, set the initial value parameters and iterative termination conditions, including the initial temperature T0: a higher initial temperature allows the algorithm to explore more solution space in the initial stage; the temperature decay rate α: the ratio of the temperature drop after each iteration, usually between 0.8 and 0.99; the stop temperature T min : When the temperature is lower than this value, the algorithm stops. Stop times: When the number of iterations reaches the stop times, the algorithm stops.

[0108] (4) Perform algorithm iteration to find the optimal solution that meets the conditions. Take the initial solution as the current solution, calculate the first shortest rescue time and the first shortest rescue path corresponding to the current solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, obtain the first evaluation value corresponding to the current solution, modify the current solution according to the first preset rule, obtain a modified solution, calculate the second shortest rescue time and the second shortest rescue path corresponding to the modified solution, substitute the second shortest rescue time and the second shortest rescue path into the objective function, obtain the second evaluation value corresponding to the modified solution, and calculate the difference between the second evaluation value and the first evaluation value.

[0109] Rescue time calculation: Calculate the total rescue time based on vehicle speed, path length, disaster site processing time, etc.

[0110] Path length calculation: Calculate the total length of all vehicle paths.

[0111] The difference between the second evaluation value and the first evaluation value is calculated. When the difference is less than zero, the modified solution is accepted as the current solution. When the difference is greater than or equal to zero, the current solution remains unchanged and is modified according to the second preset rule to obtain a new modified solution to avoid falling into a local optimum. The current temperature is equal to the initial temperature multiplied by the temperature decay rate. The current temperature or the number of iterations is used to determine whether the stopping condition has been met. If the stopping condition has not been met, the difference between the new second evaluation value and the new first evaluation value is further determined based on the new current solution and the new modified solution. Based on the difference, a new current solution and a new modified solution are further determined. The above iterative process is repeated until the stopping condition is met. The modified solution at this time is regarded as the optimal solution, and the optimal solution is used as the rescue scheduling path. The actual rescue scheduling path is obtained from the optimal solution, including the allocation of rescue resources, rescue sequence, specific path, etc.

[0112] The simulated annealing algorithm is used to find the optimal dispatch path in the insurance report rescue process to improve rescue efficiency and effectiveness.

[0113] In one embodiment, after generating the rescue dispatch strategy based on the rescue dispatch path, the method further includes:

[0114] Obtain rescue progress, resource consumption, and real-time traffic information during the rescue process, and analyze the rescue dispatch strategy based on the rescue progress, resource consumption, and real-time traffic information;

[0115] Based on the analysis results, when it is determined that the rescue dispatch strategy is in line with the current rescue status, the rescue dispatch path is fine-tuned according to resource consumption and real-time road conditions, and the rescue dispatch strategy is modified based on the adjusted content;

[0116] When it is determined that the rescue dispatch strategy does not conform to the current rescue status, the latest rescue address, the latest material demand and the latest resource configuration information under the current rescue status are obtained. According to the latest rescue address, the latest resource demand, the latest resource configuration information and the real-time road condition information, the simulated annealing algorithm is used to re-plan the rescue path to obtain a new rescue dispatch path, and a new rescue dispatch strategy is generated based on the new rescue dispatch path.

[0117] Specifically, rescue operations are carried out according to the rescue dispatch strategy generated from the optimal rescue dispatch path, collecting real-time data during the rescue process, such as rescue progress, resource consumption, and changes in road conditions. Based on this real-time feedback, the rescue dispatch path is dynamically adjusted to address changes in the actual situation. For example, when both resource consumption and road condition changes are relatively small, such as when a traffic jam is short and resource consumption is rapid, the dispatch path is fine-tuned. When both resource consumption and road condition changes are significant, such as when a traffic jam is severe, the simulated annealing algorithm can be rerun, if necessary, based on the latest road condition information, to obtain a new optimal solution and generate a new rescue dispatch strategy.

[0118] In one embodiment, the trained case reporting prediction model is obtained using the following method, including:

[0119] Obtain various historical related information corresponding to historical natural disasters;

[0120] Fusing multiple historical related information, extracting historical initial feature data from the fused historical related information, and preprocessing the historical initial feature data to obtain historical effective feature data;

[0121] The number of insurance reports corresponding to historical natural disasters is obtained, and the historical valid feature data and the number of insurance reports are used as training data. The initial report prediction model is trained based on the training data to obtain a trained report prediction model.

[0122] Specifically, we collect various historical information and historical case reporting data corresponding to natural disasters, including the time, type, and number of reports. We fuse these various historical information and extract initial historical feature data from the fused information. We preprocess this initial feature data to obtain valid historical feature data. We preprocess missing values, outliers, and duplicates to ensure data accuracy and consistency. We also extract time-related features, such as date, week, month, and quarter. These features help capture periodic patterns in time series. We use the valid historical feature data and historical case reporting data corresponding to natural disasters as training data.

[0123] The training data was divided into a training set, a test set, and a validation set in a ratio of 7:2:1. The LSTM model was initialized using the PyTorch framework, and the initial case report prediction model was iteratively trained using the training set to obtain the optimal case report prediction model. The performance of the optimal case report prediction model, including prediction accuracy and stability, was evaluated on the test set. Once the evaluation passed, the optimal case report prediction model was validated on the validation set. After validation, the final prediction model was output as the trained case report prediction model. If the model failed the test set or the validation set, the model was retrained until it passed.

[0124] Through the above steps, we can use the LSTM network to build an insurance claim prediction model, providing an evaluation basis for insurance agency rescue dispatch.

[0125] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a resource scheduling device under natural disasters, such as Figure 3 As shown, the device includes:

[0126] The insurance report prediction module 302 is used to obtain valid feature data corresponding to the natural disaster currently occurring in the dispatch area, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster;

[0127] Resource allocation module 304 is used to estimate rescue resources based on the insurance report prediction value, obtain resource estimation information, obtain the location of the natural disaster from the valid feature data corresponding to the natural disaster, deploy resources based on the natural disaster location and resource estimation information, and generate resource allocation information;

[0128] The dispatching strategy generating module 306 is used to perform resource dispatching planning according to the insurance report request and resource configuration information after receiving the insurance report request, and generate a rescue dispatching strategy.

[0129] The present application provides a resource scheduling device under natural disasters. Compared with the existing technology, when performing rescue scheduling, insurance report prediction is performed based on the effective characteristic data corresponding to the natural disasters currently occurring in the scheduling area, and the insurance report prediction is used as a basis for evaluating the required resources. Resources are deployed according to the evaluation results, and resource allocation information is generated to ensure that sufficient resources can be scheduled to respond to emergencies. After receiving the insurance report request, resource scheduling planning is performed according to the insurance report request and resource allocation information, and a rescue scheduling strategy is generated to ensure the effectiveness and rationality of the agency's rescue plan. This method can not only prepare sufficient rescue resources in advance, improve the response speed of insurance institutions in the face of natural disasters, but also accurately evaluate resource needs to avoid resource waste, and further improve the insurance company's ability to respond to natural disasters. In addition, through resource allocation information and optimal path planning of the rescue location, the most effective and reasonable rescue scheduling strategy is generated, which improves the timeliness and rationality of the insurance agency's rescue.

[0130] In one embodiment, the insurance claim prediction module is further configured to:

[0131] Obtain various real-time related information corresponding to the natural disaster within a preset period of time after the natural disaster occurs in the dispatch area;

[0132] A variety of real-time related information is integrated to obtain integrated information, and initial feature data is extracted from the integrated information, wherein the initial feature data includes the time of natural disaster occurrence, the location of natural disaster occurrence, the type of natural disaster, the level of natural disaster and the extent of damage;

[0133] The initial feature data is preprocessed to obtain effective feature data.

[0134] In one embodiment, the scheduling strategy generation module is further configured to:

[0135] Analyze insurance report requests and obtain the rescue addresses and resource requirements corresponding to multiple rescue tasks from the analysis results;

[0136] Based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue mission, the simulated annealing algorithm is used to plan the rescue path, obtain the rescue scheduling path, and generate the rescue scheduling strategy based on the rescue scheduling path.

[0137] In one embodiment, the scheduling strategy generation module is further configured to:

[0138] Based on the shortest rescue time and its corresponding weight and the shortest rescue path and its corresponding weight, an objective function is constructed, and a solution space is constructed according to a preset data structure, where the solution space includes resource allocation information, the execution order of rescue tasks, and the driving route;

[0139] Set the initial parameters in the simulated annealing algorithm, generate an initial solution based on the solution space, resource configuration information, the rescue address and resource requirements corresponding to each rescue mission, and use the initial solution as the current solution;

[0140] Calculate the first shortest rescue time and the first shortest rescue path corresponding to the initial solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, and obtain the first evaluation value corresponding to the current solution;

[0141] Modify the current solution according to the first preset rule to obtain a modified solution, calculate the second shortest rescue time and the second shortest rescue path corresponding to the modified solution, substitute the second shortest rescue time and the second shortest rescue path into the objective function, and obtain a second evaluation value corresponding to the modified solution;

[0142] Compare the second evaluation value with the first evaluation value, and according to the comparison result, obtain a new current solution and a new modified solution, and compare the second evaluation value corresponding to the new modified solution with the first evaluation value corresponding to the new current solution until the iteration condition is met, and take the modified solution at this time as the optimal solution, and take the optimal solution as the rescue scheduling path.

[0143] In one embodiment, the scheduling strategy generation module is further configured to:

[0144] When the second evaluation value is less than the first evaluation value, the modified solution is used as the new current solution, and the new current solution is modified according to the first preset rule to obtain a new modified solution;

[0145] When the second evaluation value is greater than the first evaluation value, the current solution remains unchanged, and the current solution is modified according to the second preset rule to obtain a new modified solution.

[0146] In one embodiment, the scheduling strategy generation module is further configured to:

[0147] Obtain rescue progress, resource consumption, and real-time traffic information during the rescue process, and analyze the rescue dispatch strategy based on the rescue progress, resource consumption, and real-time traffic information;

[0148] Based on the analysis results, when it is determined that the rescue dispatch strategy is in line with the current rescue status, the rescue dispatch path is fine-tuned according to resource consumption and real-time road conditions, and the rescue dispatch strategy is modified based on the adjusted content;

[0149] When it is determined that the rescue dispatch strategy does not conform to the current rescue status, the latest rescue address, the latest material demand and the latest resource configuration information under the current rescue status are obtained. According to the latest rescue address, the latest resource demand, the latest resource configuration information and the real-time road condition information, the simulated annealing algorithm is used to re-plan the rescue path to obtain a new rescue dispatch path, and a new rescue dispatch strategy is generated based on the new rescue dispatch path.

[0150] In one embodiment, the resource scheduling device under natural disasters further includes:

[0151] The model training module is used to obtain a variety of historical related information corresponding to historical natural disasters; fuse the multiple historical related information, extract historical initial feature data from the fused historical related information, pre-process the historical initial feature data, and obtain historical effective feature data; obtain the number of insurance reports corresponding to historical natural disasters, use the historical effective feature data and the number of insurance reports as training data, train the initial report prediction model based on the training data, and obtain a trained report prediction model.

[0152] It should be noted that for other corresponding descriptions of the functional units involved in the resource scheduling device under natural disasters provided in the embodiment of the present application, please refer to Figures 1 to 2 The corresponding description in the method will not be repeated here.

[0153] The present application also provides a computer device, which can be a personal computer, a server, a network device, etc. Figure 4 As shown, the computer device includes a bus, a processor, a memory, and a communication interface, and may also include an input / output interface and a display device. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps of each method embodiment are implemented.

[0154] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0155] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0156] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0157] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0158] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0159] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A resource scheduling method under natural disasters, characterized in that: The method comprises: Obtain valid feature data corresponding to the natural disaster currently occurring in the area to be dispatched, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster; Based on the insurance report prediction value, rescue resources are estimated to obtain resource estimation information, the location of the natural disaster is obtained from the valid feature data corresponding to the natural disaster, resources are deployed based on the natural disaster location and the resource estimation information, and resource allocation information is generated; After receiving the insurance report request, resource scheduling planning is performed according to the insurance report request and the resource configuration information to generate a rescue scheduling strategy.

2. The resource scheduling method under natural disasters according to claim 1, characterized in that: The obtaining of valid characteristic data corresponding to the natural disaster currently occurring in the area to be dispatched includes: Acquiring a variety of real-time related information corresponding to a natural disaster within a preset period of time after the natural disaster occurs in the area to be dispatched; fusing the multiple real-time related information to obtain fused information, and extracting initial feature data from the fused information, wherein the initial feature data includes the time of occurrence of the natural disaster, the location of the natural disaster, the type of natural disaster, the level of the natural disaster, and the extent of damage; The initial feature data is preprocessed to obtain effective feature data.

3. The resource scheduling method under natural disasters according to claim 1, characterized in that: The performing resource scheduling planning according to the insurance report request and the resource configuration information to generate a rescue scheduling strategy includes: Analyze the insurance report request and obtain rescue addresses and resource requirements corresponding to multiple rescue tasks from the analysis results; Based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, a simulated annealing algorithm is used to plan a rescue path to obtain a rescue scheduling path, and a rescue scheduling strategy is generated based on the rescue scheduling path.

4. The resource scheduling method under natural disasters according to claim 3, characterized in that: The rescue path planning is performed based on the resource configuration information, the rescue address and resource requirements corresponding to each rescue task using a simulated annealing algorithm to obtain a rescue dispatch path, including: Based on the shortest rescue time and its corresponding weight and the shortest rescue path and its corresponding weight, an objective function is constructed, and a solution space is constructed according to a preset data structure, wherein the solution space includes resource allocation information, the execution order of rescue tasks, and the driving route; Setting initial parameters in a simulated annealing algorithm, generating an initial solution based on the solution space, the resource configuration information, the rescue address and resource requirements corresponding to each rescue task, and using the initial solution as the current solution; Calculate a first shortest rescue time and a first shortest rescue path corresponding to the initial solution, substitute the first shortest rescue time and the first shortest rescue path into the objective function, and obtain a first evaluation value corresponding to the current solution; modifying the current solution according to a first preset rule to obtain a modified solution, calculating a second shortest rescue time and a second shortest rescue path corresponding to the modified solution, substituting the second shortest rescue time and the second shortest rescue path into the objective function to obtain a second evaluation value corresponding to the modified solution; Compare the second evaluation value with the first evaluation value, and obtain a new current solution and a new modified solution based on the comparison result. Compare the second evaluation value corresponding to the new modified solution with the first evaluation value corresponding to the new current solution until the iteration condition is met. Take the modified solution at this time as the optimal solution, and take the optimal solution as the rescue scheduling path.

5. The resource scheduling method under natural disasters according to claim 4, characterized in that: The step of obtaining a new current solution and a new modified solution based on the comparison results includes: When the second evaluation value is less than the first evaluation value, taking the modified solution as the new current solution, and modifying the new current solution according to the first preset rule to obtain a new modified solution; When the second evaluation value is greater than the first evaluation value, the current solution remains unchanged, and the current solution is modified according to a second preset rule to obtain a new modified solution.

6. The resource scheduling method under natural disasters according to claim 3, characterized in that: After generating the rescue dispatch strategy based on the rescue dispatch path, the method further includes: Obtaining rescue progress, resource consumption, and real-time traffic information during the rescue process, and analyzing the rescue scheduling strategy based on the rescue progress, resource consumption, and real-time traffic information; When it is determined based on the analysis results that the rescue dispatch strategy is consistent with the current rescue status, the rescue dispatch path is fine-tuned according to the resource consumption and real-time traffic information, and the rescue dispatch strategy is modified based on the adjusted content; When it is determined that the rescue scheduling strategy does not conform to the current rescue status, the latest rescue address, the latest material demand and the latest resource configuration information under the current rescue status are obtained, and according to the latest rescue address, the latest resource demand, the latest resource configuration information and the real-time road condition information, the rescue path is re-planned using a simulated annealing algorithm to obtain a new rescue scheduling path, and a new rescue scheduling strategy is generated based on the new rescue scheduling path.

7. The resource scheduling method under natural disasters according to claim 2, characterized in that: The trained case reporting prediction model is obtained using the following method, including: Obtain various historical related information corresponding to historical natural disasters; fusing the plurality of historical related information, extracting historical initial feature data from the fused historical related information, and preprocessing the historical initial feature data to obtain historical valid feature data; The number of insurance reports corresponding to the historical natural disasters is obtained, the historical valid feature data and the number of insurance reports are used as training data, an initial report prediction model is trained based on the training data, and a trained report prediction model is obtained.

8. A resource scheduling device under natural disasters, characterized in that: The device comprises: An insurance report prediction module is used to obtain valid feature data corresponding to the natural disaster currently occurring in the dispatch area, input the valid feature data into the trained report prediction model, and obtain the insurance report prediction value corresponding to the natural disaster; a resource allocation module, configured to estimate rescue resources based on the insurance report prediction value, obtain resource estimation information, obtain the location of the natural disaster from the valid characteristic data corresponding to the natural disaster, deploy resources based on the natural disaster location and the resource estimation information, and generate resource allocation information; The scheduling strategy generation module is used to receive an insurance report request, perform resource scheduling planning according to the insurance report request and the resource configuration information, and generate a rescue scheduling strategy.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Images