Emergency support force intelligent dispatching method and device, electronic equipment and storage medium

Abstract

The invention discloses an emergency support force intelligent dispatching method and device, electronic equipment and a storage medium, and is used for solving the technical problems that a current emergency dispatching method is relatively low in efficiency, and an optimal scheme is difficult to quickly provide in complex and changeable disasters. The method comprises the following steps: determining a to-be-supported disaster-affected city, and obtaining emergency resource distribution data except the disaster-affected city in a preset jurisdiction range; according to the emergency resource distribution data, performing optimal dispatching integer planning based on different emergency support demands on the affected city, and generating an emergency support dispatching scheme corresponding to each emergency support demand; and respectively generating a dispatching order file of each emergency support dispatching scheme, wherein the dispatching order file is used for providing dispatching decision reference for the user. Therefore, the emergency support optimal dispatching scheme can be adaptively generated in the resource dispatching process, and a decision maker can quickly select the optimal emergency dispatching scheme in complex and changeable disasters.

Description

Technical Field

[0001] The present invention relates to the field of operations research and artificial intelligence technology, and in particular to an intelligent dispatching method, device, electronic equipment and storage medium for emergency support forces. Background Art

[0002] In recent years, with the frequent occurrence of extreme weather events, how to efficiently dispatch rescue resources has become an important research topic in disaster emergency management. When natural disasters such as typhoons occur, the rational allocation of rescue resources is directly related to the efficiency of emergency response and the effectiveness of post-disaster recovery.

[0003] However, most of the current methods are still at the stage of manual decision-making. Manual decision-making relies on experience and judgment, which is not only inefficient, but also difficult to quickly provide the best solution in complex and changing disaster situations. Especially when facing multi-objective optimization needs, there is a lack of systematic solutions that can balance efficiency and comprehensiveness.

[0004] On the other hand, the results of traditional scheduling methods are usually presented in the form of digital tables or static reports. The amount of information is huge and not intuitive enough, which easily increases the understanding burden of decision makers in high-pressure environments. For this reason, there is an urgent need for an intelligent algorithm that can analyze and generate multi-objective scheduling solutions in real time, while improving the intuitiveness of the expression of dispatch results. Summary of the invention

[0005] The present invention provides an intelligent dispatching method, device, electronic device and storage medium for emergency support forces, which are used to solve or partially solve the technical problems that the current emergency dispatching methods are inefficient and difficult to quickly provide optimal solutions in complex and changeable disaster situations.

[0006] The present invention provides an intelligent dispatching method for emergency support forces, the method comprising:

[0007] Determine the disaster-stricken cities to be supported, and obtain emergency resource distribution data other than the disaster-stricken cities within the preset jurisdiction;

[0008] According to the emergency resource distribution data, optimal dispatch integer programming is performed on the disaster-stricken cities based on different emergency support needs, and an emergency support dispatch plan corresponding to each emergency support need is generated;

[0009] A dispatch order file is generated for each of the emergency support dispatch plans, and the dispatch order file is used to provide a dispatch decision reference to the user.

[0010] Optionally, the emergency resource distribution data includes fire rescue detachments in all non-disaster-affected areas within the preset jurisdiction; performing optimal dispatch integer programming based on different emergency support needs for the disaster-affected city according to the emergency resource distribution data, and generating an emergency support dispatch plan corresponding to each emergency support need, including:

[0011] Taking all fire rescue detachments in non-disaster areas within the preset jurisdiction as planning objects and minimizing the driving time of the planning objects as the goal, an emergency dispatch objective function is constructed;

[0012] Constructing the shortest time support demand constraint, combining the emergency dispatch objective function with the shortest time support demand constraint to construct the shortest time scheduling demand model, solving the shortest time scheduling demand model through an integer programming solver, and obtaining the shortest time dispatch plan;

[0013] On the basis of the shortest time support demand constraint, the strongest force support demand constraint is constructed, the strongest force dispatch demand model is constructed by combining the emergency dispatch objective function with the strongest force support demand constraint, and the strongest force dispatch demand model is solved by an integer programming solver to obtain the strongest force dispatch plan;

[0014] On the basis of the shortest time support demand constraint, the least impact support demand constraint is constructed, and the least impact scheduling demand model is constructed by combining the emergency dispatch objective function and the least impact support demand constraint. The least impact scheduling demand model is solved by an integer programming solver to obtain the least impact dispatch plan.

[0015] Optionally, each of the fire rescue detachments is composed of a plurality of flood fighting and rescue teams and a plurality of water rescue professional teams; the shortest support demand constraint for building time includes:

[0016] Based on the number of flood fighting and rescue teams and the number of professional water rescue teams, a constraint on the number of emergency dispatches is established;

[0017] Obtain the reinforcement demand of the disaster-stricken city, and establish a constraint on the number of emergency support personnel based on the reinforcement demand, the dispatched quantity of the flood fighting and rescue team, and the dispatched quantity of the water rescue professional team;

[0018] Based on the dispatch volume of the flood fighting and rescue team and the dispatch volume of the water rescue professional team, a non-negative constraint on the emergency dispatch volume is constructed;

[0019] The emergency dispatch quantity constraint, the emergency support number constraint and the emergency dispatch quantity non-negative constraint are integrated to construct the shortest time support demand constraint.

[0020] Optionally, the fire rescue detachments in all unaffected areas include multiple first-level fire rescue detachments and multiple second-level fire rescue detachments, and the support force of the first-level fire rescue detachment is stronger than that of the second-level fire rescue detachment; the strongest support demand constraint is constructed on the basis of the shortest support demand constraint, including:

[0021] Based on the mutually exclusive reaction between the first-level fire rescue detachment and the second-level fire rescue detachment, a constraint of calling with the strongest force first is established;

[0022] The shortest support requirement constraint and the strongest force priority call constraint are integrated to construct the strongest force support requirement constraint.

[0023] Optionally, the fire rescue detachments in all non-disaster-affected areas include fire rescue detachments in adjacent cities and fire rescue detachments in non-adjacent cities; and the minimum impact support demand constraint is constructed based on the shortest time support demand constraint, including:

[0024] Based on the mutually exclusive reaction between the fire rescue detachment of the adjacent city and the fire rescue detachment of the non-adjacent city, a resource circle support priority call constraint is established;

[0025] The shortest time support requirement constraint and the resource circle support priority call constraint are integrated to construct the least impact support requirement constraint.

[0026] Optionally, the generating of the dispatch order files of the emergency support dispatch plans respectively includes:

[0027] Analyzing the emergency dispatch situation of the disaster-stricken cities based on each of the emergency support dispatch plans;

[0028] For each of the emergency support dispatch plans, based on the analysis results, the fire rescue detachments with the same number of emergency support dispatch personnel are merged;

[0029] Based on the merging result, a PDF generator is used to generate dispatch order files for each of the emergency support dispatch plans.

[0030] Optionally, the intelligent dispatching method for emergency support forces further includes:

[0031] Using a locally deployed large language model, or calling a cloud-based large language model API, respectively summarize the shortest time dispatch plan, the most forceful dispatch plan, and the least impact dispatch plan, and obtain a shortest time dispatch summary text, a most forceful dispatch summary text, and a least impact dispatch summary text;

[0032] Based on the keywords of the strongest detachment, the content of the summary text of the strongest dispatch is supplemented, and based on the keywords of the neighboring cities, the content of the summary text of the dispatch with the least impact is supplemented;

[0033] Get the task instruction text entered by the user;

[0034] Combine the task instruction text, the shortest time dispatch summary text, the strongest force dispatch summary text after replenishment, and the smallest impact dispatch summary text after replenishment to obtain an emergency support keyword text;

[0035] A locally deployed large language model is used, or a cloud-based large language model API is called to perform a secondary solution summary on the emergency support keyword text to obtain a final solution summary text.

[0036] The present invention also provides an intelligent dispatching device for emergency support forces, comprising:

[0037] A data acquisition unit, used to determine the disaster-stricken city to be supported, and to acquire emergency resource distribution data other than the disaster-stricken city within a preset jurisdiction;

[0038] An optimal dispatch integer programming unit, used to perform optimal dispatch integer programming based on different emergency support needs for the disaster-stricken city according to the emergency resource distribution data, and generate an emergency support dispatch plan corresponding to each emergency support need;

[0039] The dispatch order file generating unit is used to generate dispatch order files for each of the emergency support dispatch plans respectively, and the dispatch order files are used to provide dispatch decision references to users.

[0040] The present invention also provides an electronic device, the device comprising a processor and a memory:

[0041] The memory is used to store program code and transmit the program code to the processor;

[0042] The processor is used to execute the intelligent dispatching method of emergency support force as described in any of the above items according to the instructions in the program code.

[0043] The present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the intelligent dispatching method of emergency support force as described in any of the above items.

[0044] It can be seen from the above technical solutions that the present invention has the following advantages:

[0045] A method for intelligent dispatch of emergency support forces is provided. First, the disaster-stricken cities to be supported are determined, and the distribution data of emergency resources other than the disaster-stricken cities within the preset jurisdiction are obtained; then, according to the emergency resource distribution data, the disaster-stricken cities are subjected to optimal dispatch integer programming based on different emergency support needs, and emergency support dispatch plans corresponding to each emergency support need are generated; then, dispatch order files for each emergency support dispatch plan are generated respectively, and the dispatch order files are used to provide dispatch decision references to users. Therefore, through the optimal dispatch integer programming, the efficiency of emergency dispatch plan generation can be improved, and the optimal dispatch plan for emergency support can be adaptively generated during the resource dispatch process to meet the needs of different emergency support scenarios. At the same time, the dispatch order files generated based on different optimal dispatch plans provide decision makers with effective emergency dispatch decision references, so that decision makers can quickly select the optimal emergency dispatch plan in complex and changeable disaster situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 The present invention is a flowchart of the steps of an intelligent dispatching method for emergency support forces;

[0048] Figure 2 It is a schematic diagram of the overall framework of an intelligent dispatching method for emergency support forces;

[0049] Figure 3 It is a schematic diagram of the overall process of an intelligent dispatching method for emergency support forces;

[0050] Figure 4 The present invention is a structural block diagram of an intelligent dispatching device for emergency support forces. DETAILED DESCRIPTION

[0051] The embodiments of the present invention provide a method, device, electronic device and storage medium for intelligently dispatching emergency support forces, which are used to solve or partially solve the technical problems that the current emergency dispatch methods are inefficient and difficult to quickly provide optimal solutions in complex and changeable disaster situations.

[0052] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] As an example, in recent years, with the frequent occurrence of extreme weather events, how to efficiently dispatch rescue resources has become an important research topic in disaster emergency management. When natural disasters such as typhoons occur, the reasonable allocation of rescue resources is directly related to the efficiency of emergency response and the effectiveness of post-disaster recovery.

[0054] However, most of the current methods are still at the stage of manual decision-making. Manual decision-making relies on experience and judgment, which is not only inefficient, but also difficult to quickly provide the best solution in complex and changing disaster situations. Especially when facing multi-objective optimization needs, there is a lack of systematic solutions that can balance efficiency and comprehensiveness.

[0055] On the other hand, the results of traditional scheduling methods are usually presented in the form of digital tables or static reports. The amount of information is huge and not intuitive enough, which easily increases the understanding burden of decision makers in high-pressure environments. For this reason, there is an urgent need for an intelligent algorithm that can analyze and generate multi-objective scheduling solutions in real time, while improving the intuitiveness of the expression of dispatch results.

[0056] Therefore, one of the core inventive points of the embodiment of the present invention is: in view of the shortcomings of the current technology, a method for intelligent dispatch of emergency support forces and generation of program summary text combining integer programming and large language models is proposed. On the one hand, based on the characteristics of disasters and resource constraints, by combining integer programming with a large language model, the optimal dispatch plan for emergency support is adaptively generated in the process of resource dispatch, including the dispatch plan with the shortest time, the dispatch plan with the most force, and the dispatch plan with the least impact, to meet the needs of different emergency support scenarios. The specific dispatch content, such as the number of support personnel, the name of the support unit, and the assisted city, will automatically generate a dispatch order file through the system, and the dispatch process will be efficient, accurate and standardized. On the other hand, based on the dispatch results, a large language model is used to generate an intuitive and easy-to-understand summary text to provide support for decision makers to quickly understand the dispatch results.

[0057] Reference Figure 1 , shows a flowchart of a method for intelligently dispatching emergency support forces provided by an embodiment of the present invention, which may specifically include the following steps:

[0058] Step 101, determining the disaster-stricken city to be supported, and obtaining emergency resource distribution data other than the disaster-stricken city within a preset jurisdiction;

[0059] Steps 101 to 102 in the embodiment of the present invention are to achieve modeling based on emergency support scenarios. Specifically, an integer programming algorithm is designed for three different emergency support requirements, namely, "shortest time", "strongest force" and "least impact", in combination with their respective constraints.

[0060] Taking Guangdong Province as an example, first consider each city in the province as a resource distribution node. Each city is equipped with a certain number of flood fighting and rescue teams, namely fire rescue detachments. For ease of understanding, assume that one city corresponds to one fire rescue detachment as a team set. Each fire rescue detachment is composed of multiple flood fighting and rescue teams and multiple water rescue professional teams. Then for the disaster-stricken cities, the corresponding emergency resource distribution data can include all fire rescue detachments in the non-disaster-stricken areas within the preset jurisdiction (such as Guangdong Province).

[0061] For example, suppose that the water rescue professional team is set up with 50 people per team. This team focuses on water rescue tasks, such as rescuing trapped people and dealing with flood hazards. The flood fighting and rescue team is set up with 100 people per team. This team is mainly responsible for carrying out tasks such as strengthening embankments, draining and desilting.

[0062] In order to simplify the computational complexity and improve practical operability, the integer programming algorithm plans the number of teams rather than the specific number of people. This design method can not only effectively reduce the scale of scheduling calculations, but also ensure the overall coordination and professional division of labor of rescue forces.

[0063] Understandably, in reality, the disaster area usually covers multiple cities. The reinforcement needs of each affected city vary depending on the severity of the disaster. The mobilization of rescue forces is the responsibility of the fire rescue detachment in the unaffected area, which transports the team to the fire rescue detachment in the affected city. Due to the differences in geographical distribution between the detachments, there are different driving times between the detachments. The algorithm goal is to shorten the total dispatch time as much as possible while meeting the total dispatch demand, so as to achieve efficient resource utilization and rapid response.

[0064] It should be noted that, for ease of explanation, the embodiment of the present invention proposes a corresponding intelligent dispatch method for emergency support forces based on a single disaster-stricken city. When there are multiple disaster-stricken cities that need emergency support at the same time, emergency dispatch can be performed for each disaster-stricken city with reference to the technical solution provided in the embodiment of the present invention. It is understandable that the present invention is not limited to this.

[0065] In order to accurately describe the problem, the embodiment of the present invention introduces a series of variables. First, define the fire rescue detachment To the disaster-stricken cities The driving time is . And use and Indicates the number of flood fighting and rescue teams and water rescue professional teams owned by each fire rescue brigade. At the same time, the reinforcements needed by the disaster-stricken cities are The collection of disaster-affected cities, the collection of fire rescue detachments in Guangdong Province, and the collection of fire rescue detachments in the Greater Bay Area are respectively represented by In addition, the present invention also defines each disaster-affected city Collection of neighboring cities , in order to optimize the dispatch path within the region. Among the decision variables of the model, and Respectively from the Fire Rescue Brigade Transfer to disaster-stricken cities The number of flood fighting and rescue teams and the number of professional water rescue teams.

[0066] By integrating the above variables and the constraints under different emergency support scenario requirements, in the subsequent integer programming optimization calculation process, based on different emergency support demand emphases, the algorithm can generate three different dispatch plans: the shortest time dispatch plan, the most force-intensive dispatch plan, and the least impact dispatch plan. Among them, the shortest time dispatch plan aims to minimize the total driving time and gives priority to rapid response. The most force-intensive dispatch plan prioritizes the delivery of Greater Bay Area forces to the disaster-stricken cities. The least impact dispatch plan tries to avoid calling in support forces near the disaster area (bordering cities). The specific planning methods for these three emergency support dispatch plans are given in step 102.

[0067] Step 102, performing optimal dispatch integer programming based on different emergency support needs for the disaster-stricken cities according to the emergency resource distribution data, and generating an emergency support dispatch plan corresponding to each emergency support need;

[0068] In combination with the above-mentioned contents, in some embodiments, according to the emergency resource distribution data, optimal dispatch integer programming is performed on the disaster-stricken cities based on different emergency support needs, and the process of generating an emergency support dispatch plan corresponding to each emergency support need can be achieved by executing the following sub-steps S01 to S04:

[0069] Step S01: Taking all fire rescue detachments in non-disaster areas within the preset jurisdiction as planning objects, and minimizing the driving time of the planning objects as the goal, constructing an emergency dispatch objective function;

[0070] First, define the emergency dispatch objective function. The planning goal of each emergency support dispatch plan is to minimize the driving time, and the planning objects are all fire rescue detachments that include all unaffected areas. The expression of the emergency dispatch objective function is as follows:

[0071]

[0072] Step S02: construct the shortest time support demand constraint, combine the emergency dispatch objective function with the shortest time support demand constraint to construct the shortest time dispatch demand model, solve the shortest time dispatch demand model through the integer programming solver, and obtain the shortest time dispatch plan;

[0073] The shortest time dispatch plan is to gather the most rescue forces in the shortest time and quickly reach the designated disaster site. This plan is particularly suitable for situations where the typhoon has a small impact area but the situation is very urgent. In such cases, by adopting the shortest time dispatch plan, the rescue team can arrive at the scene as soon as possible to carry out rescue work.

[0074] Based on this, the shortest time plan only needs to consider the least driving time to quickly complete the dispatching needs. Therefore, the number of support personnel can be transferred from the fire rescue detachment of any unaffected city (unaffected area).

[0075] Specifically, the process of constructing the shortest support requirement constraint can be implemented by executing the following sub-steps S021 to S024:

[0076] Step S021: constructing emergency dispatch quantity constraints based on the number of flood fighting and rescue teams and the number of water rescue professional teams;

[0077] The emergency dispatch quantity constraint means that the number of dispatches of each fire rescue brigade cannot exceed the number of its existing teams. The expression of this constraint is as follows:

[0078]

[0079]

[0080] Step S022: Obtain the reinforcement demand of the disaster-stricken city, and construct a constraint on the number of emergency support personnel based on the reinforcement demand, the dispatched number of flood fighting and rescue teams, and the dispatched number of professional water rescue teams;

[0081] The emergency support number constraint means that the total number of rescue forces must meet the needs of each disaster-stricken city. The expression of this constraint is as follows:

[0082]

[0083] Step S023: constructing a non-negative constraint on the emergency dispatch quantity based on the dispatch quantity of the flood fighting and rescue teams and the dispatch quantity of the water rescue professional teams;

[0084] The non-negative constraint of emergency dispatch quantity means that the dispatch quantity of fire rescue teams is a non-negative integer. The expression of this constraint is as follows:

[0085]

[0086]

[0087] Step S024: Integrate the emergency dispatch quantity constraint, the emergency support number constraint and the emergency dispatch quantity non-negative constraint to construct the shortest time support demand constraint.

[0088] After constructing the shortest support demand constraint, the shortest scheduling demand model can be constructed by combining the emergency dispatch objective function with the shortest support demand constraint. Then, the shortest scheduling demand model can be solved by the integer programming solver to obtain the shortest dispatch plan.

[0089] Step S03: On the basis of the shortest time support demand constraint, the strongest force support demand constraint is constructed, and the strongest force dispatch demand model is constructed by combining the emergency dispatch objective function and the strongest force support demand constraint. The strongest force dispatch demand model is solved by an integer programming solver to obtain the strongest force dispatch plan;

[0090] The most powerful dispatching plan is to mobilize the strongest rescue forces among the fire rescue detachments and mobile teams in areas with strong support forces (such as the Greater Bay Area) to the designated disaster site. This plan is particularly suitable for large-scale disaster events. When experienced personnel are needed to command on site, choosing this plan can maximize the rescue effect.

[0091] Based on this, the dispatch plan with the most emphasis on strength needs to add a constraint to call the strongest detachment on the basis of considering the least driving time. Taking Guangdong Province as an example, support personnel can only be transferred from the detachments of the Greater Bay Area cities. Including Guangzhou Detachment, Shenzhen Detachment, Foshan Detachment, Dongguan Detachment, Huizhou Detachment. When the number of support personnel in the Greater Bay Area cities is not enough to meet the dispatch needs, personnel will be transferred from the detachments of other cities.

[0092] In this case, according to the strength of emergency support, all fire rescue detachments in non-disaster areas can be divided into multiple first-level fire rescue detachments and multiple second-level fire rescue detachments. Among them, the first-level fire rescue detachment represents a team with strong emergency support, such as the Greater Bay Area cities mentioned in the previous example. The second-level fire rescue detachment represents a team with average emergency support, such as other cities except the Greater Bay Area cities mentioned in the previous example. It can be seen that the support strength of the first-level fire rescue detachment is stronger than that of the second-level fire rescue detachment.

[0093] In terms of constraints, three of the constraints of the plan that emphasizes the most force are the same as those of the plan that emphasizes the shortest time, including the number of emergency dispatches, the number of emergency support personnel, and the non-negative emergency dispatch quantity. However, on the basis of these three constraints, the plan that emphasizes the most force sets an additional constraint on the strongest support demand to achieve priority dispatch of the Greater Bay Area detachment.

[0094] Specifically, on the basis of the shortest time support requirement constraint, the process of constructing the strongest force support requirement constraint can be achieved by executing the following sub-steps S031 to S032:

[0095] Step S031: Based on the mutually exclusive reaction between the first-level fire rescue detachment and the second-level fire rescue detachment, a constraint of calling with the strongest force first is constructed;

[0096] In actual dispatch, priority is given to calling the fire rescue detachment with strong support. The constraint of calling the strongest force first can actually be understood as a mutually exclusive constraint as shown in the following formula:

[0097]

[0098] Step S032: Integrate the shortest support requirement constraint and the strongest force priority call constraint to construct the strongest force support requirement constraint.

[0099] After constructing the strongest support demand constraint, we can combine the emergency dispatch objective function with the strongest support demand constraint to construct the strongest scheduling demand model, and then solve the strongest scheduling demand model through the integer programming solver to obtain the strongest dispatch plan.

[0100] Step S04: On the basis of the shortest support demand constraint, construct the least impact support demand constraint, and construct the least impact scheduling demand model by combining the emergency dispatch objective function and the least impact support demand constraint. Solve the least impact scheduling demand model through the integer programming solver to obtain the least impact dispatch plan.

[0101] The minimum impact dispatch plan means that while gathering rescue forces, other cities near the disaster-stricken city should be avoided as much as possible, and rescue forces from distant areas should be called in to ensure safety as much as possible. In this way, cities around the disaster can respond and deal with their own disasters on their own, thereby reducing the overall rescue pressure. This plan is particularly suitable for situations where the impact of typhoons is large and multiple areas are affected at the same time. In such cases, by adopting the minimum impact dispatch plan, the impact on surrounding areas can be minimized to ensure that all parties can effectively respond to their respective emergencies.

[0102] Based on this, the dispatch plan with the least impact needs to add a constraint that gives priority to calling detachments from non-neighboring cities (i.e. non-bordering cities) on the basis of considering the least driving time. That is, support personnel cannot be transferred from fire rescue detachments in cities adjacent to the disaster-stricken city. For example, assuming that "Yangjiang" was hit by a typhoon, since the neighboring cities of "Yangjiang" include "Maoming", "Yunfu" and "Jiangmen", then in actual dispatch, the dispatch plan with the least impact will not give priority to emergency support forces from "Maoming", "Yunfu" and "Jiangmen", but will give priority to support forces from cities farther away. When the number of support personnel from non-neighboring cities is not enough to meet the dispatch needs, personnel will be transferred from detachments in other cities.

[0103] In this case, all fire rescue detachments in non-disaster-stricken areas can be divided into bordering city fire rescue detachments and non-bordering city fire rescue detachments according to whether they are adjacent to the disaster-stricken city.

[0104] In terms of constraints, three of the constraints of the dispatch plan with the least impact are the same as those of the dispatch plan with the shortest time, including the emergency dispatch quantity constraint, the emergency support number constraint, and the emergency dispatch quantity non-negative constraint. However, on the basis of these three constraints, the dispatch plan with the least impact also sets the constraint of the least impact support demand to achieve the priority dispatch of non-adjacent city detachments.

[0105] Specifically, based on the shortest time support requirement constraint, the process of constructing the least impact support requirement constraint can be achieved by executing the following sub-steps S041 to S042:

[0106] Step S041: constructing resource circle support priority call constraints based on the mutually exclusive reaction between the fire rescue detachments of the adjacent cities and the fire rescue detachments of the non-adjacent cities;

[0107] In actual dispatch, the resource circle rescue detachment (i.e. the fire rescue detachment of non-bordering cities) is called first. The resource circle support priority calling constraint can actually be understood as a mutually exclusive constraint as shown in the following formula:

[0108]

[0109]

[0110] Step S042: Integrate the shortest time support requirement constraint and the resource circle support priority call constraint to construct the minimum impact support requirement constraint.

[0111] After constructing the support demand constraint with the least impact, the emergency dispatch objective function and the support demand constraint with the least impact can be combined to construct a scheduling demand model with the least impact. Then, the scheduling demand model with the least impact can be solved by the integer programming solver to obtain the dispatch plan with the least impact.

[0112] By completing the above steps, we can obtain the specific dispatch strategies (including starting point, end point, and number of dispatch teams) including the shortest time dispatch plan, the most powerful dispatch plan, and the dispatch plan with the least impact.

[0113] Step 103, respectively generate dispatch order files for the emergency support dispatch plans, wherein the dispatch order files are used to provide dispatch decision references to users.

[0114] This step mainly uses Python's third-party library - PDF generator Reportlab to generate three dispatch order PDF files based on the scheduling contents of the three schemes obtained in the previous integer programming step.

[0115] In some embodiments, the process of generating the dispatch order files of each emergency support dispatch plan can be implemented by executing the following sub-steps S11 to S13:

[0116] Step S11: analyzing the emergency dispatch situation of the disaster-stricken cities based on each emergency support dispatch plan;

[0117] Regarding the specific content of the dispatch plan in the dispatch order file, first traverse each disaster-stricken city that needs support (when there are multiple disaster-stricken cities at the same time), and analyze which fire rescue detachments will support each disaster-stricken city.

[0118] Step S12: for each emergency support dispatch plan, according to the analysis results, merge the fire rescue detachments with the same number of emergency support dispatch personnel;

[0119] For each emergency support dispatch plan for a disaster-stricken city, the fire rescue detachments of cities with the same number of support personnel are combined for text printing, making the content of the dispatch order file more concise and neat. For example: if City A, City B, and City C will dispatch 1 detachment of 50 people and 3 detachments of 100 people to the current disaster-stricken city, then the output will be combined into "City A, City B, and City C will each dispatch 1 detachment of 50 people and 3 detachments of 100 people to the disaster-stricken city."

[0120] Step S13: Based on the merging result, a PDF generator is used to generate dispatch order files for each emergency support dispatch plan.

[0121] In addition, in order to make the dispatch information more comprehensive, some replaceable information can be inserted into the dispatch order file, such as typhoon wind force, typhoon name, date, contact person, contact information, etc.

[0122] After obtaining the dispatch order files of three different emergency support dispatch plans, namely the shortest time dispatch plan, the most forceful dispatch plan, and the least impact dispatch plan, we can also use the large language model to summarize the characteristics of the three plans based on the dispatch results of the three different plans.

[0123] In some embodiments, the process of summarizing the characteristics of the three solutions based on the scheduling results of the three different solutions using the large language model can be implemented by executing the following sub-steps S21 to S25:

[0124] Step S21: using a locally deployed large language model, or calling a cloud-based large language model API, to summarize the dispatch plan with the shortest time, the dispatch plan with the most force, and the dispatch plan with the least impact, respectively, and obtain a dispatch summary text with the shortest time, a dispatch summary text with the most force, and a dispatch summary text with the least impact;

[0125] In this step, you need to use the locally deployed big language model or call the cloud-based big language model API (Application Programming Interface) to generate a summary text around the scheduling content of the current plan.

[0126] In the embodiment of the present invention, the format of the input prompt (prompt word or instruction) set by the large language model is as follows: task background + current scheduling details + summary of the characteristics of the current scheduling plan + task instruction text.

[0127] For example, the task background text may be "The typhoon hit some cities in Guangdong Province. Now the Guangdong Provincial Fire Department needs to dispatch support forces from fire brigades in various places to support these disaster-stricken cities."

[0128] The current dispatch details include the types and numbers of teams transferred from each non-disaster-affected city (or fire rescue detachment) to the disaster-affected city in the current dispatch plan, as well as the corresponding driving time.

[0129] The summary of the characteristics of the current dispatch plan includes a summary of the characteristics of three different emergency support dispatch plans:

[0130] The summary text of the "Shortest Dispatch Plan" is: "The Shortest Dispatch Plan refers to assembling the most rescue forces in the shortest time and quickly reaching the designated disaster site. This plan is particularly suitable for situations where the typhoon has a small impact range but the situation is very urgent, to ensure that the rescue team arrives at the scene as soon as possible to carry out rescue work."

[0131] The summary text of the "Most Powerful Dispatch Plan" is: "The Most Powerful Dispatch Plan refers to mobilizing the strongest rescue forces in the Greater Bay Area and the mobile team to the designated disaster site. This plan is particularly suitable for large-scale disaster events. When experienced personnel are needed to command on site, choosing this plan can maximize the rescue effect."

[0132] The summary text of the "Minimum Impact Dispatch Plan" is: "The minimum impact dispatch plan means that while assembling rescue forces, we will try our best to avoid involving other cities near the disaster, and instead call on rescue forces from farther away areas to ensure safety. In this way, cities around the disaster can respond and deal with their own disasters on their own, thereby reducing the overall rescue pressure. This plan is particularly suitable for use when the typhoon has a large impact and multiple areas are affected at the same time. It aims to minimize the impact on surrounding areas and ensure that all parties can effectively respond to their respective emergencies."

[0133] Step S22: supplementing the content of the summary text of the most powerful detachment based on the keywords of the most powerful detachment, and supplementing the content of the summary text of the least influential detachment based on the keywords of the neighboring cities;

[0134] In addition, after the summary of the characteristics, it is necessary to add the information of "Greater Bay Area cities" and "bordering cities" to the prompts of "the plan with the most emphasis on power" and "the plan with the least impact". That is, after the summary of the characteristics of the "plan with the most emphasis on power", add "the strongest detachments include: mobile detachment, Guangzhou detachment, Foshan detachment, Shenzhen detachment, Dongguan detachment, Huizhou detachment." After the summary of the characteristics of the "plan with the least impact", analyze the location of the disaster-stricken cities, obtain the information of bordering cities, and add the information of all bordering cities to the prompt.

[0135] Step S23: Obtaining the task instruction text input by the user;

[0136] For example, the task instruction text may be: "Please summarize the advantages of the current plan based on the above information, and focus on the characteristics of the current plan without diverging from other aspects. You may appropriately quote the current scheduling plan, but do not directly discuss the specific number of driving times."

[0137] Step S24: combining the task instruction text, the shortest time dispatch summary text, the most powerful dispatch summary text after replenishment, and the least influential dispatch summary text after replenishment to obtain an emergency support keyword text;

[0138] Finally, the information obtained through the above steps S21 to S24 is combined to obtain the emergency support keyword text Prompt text.

[0139] Step S25: Use the locally deployed large language model, or call the cloud-based large language model API to perform a secondary solution summary on the emergency support keyword text to obtain the final solution summary text.

[0140] Input the emergency support keyword text Prompt text into the large language model to obtain the summary text of the current plan.

[0141] In order to enable those skilled in the art to better understand the technical solution of the present invention, based on the contents of the intelligent dispatching solution for emergency support forces introduced in the previous embodiment, Figure 2 A corresponding overall framework schematic diagram is shown.

[0142] It should be noted that the method provided in the embodiment of the present invention provides three different emergency support dispatching schemes to provide users with dispatching decision references. In actual application scenarios, those skilled in the art can also choose whether to directly dispatch emergency support forces based on machine algorithm judgment according to actual conditions. For example, the applicability of these three schemes can be directly divided according to the type of disaster and the degree of disaster.

[0143] If it is determined that rescue operations need to be started quickly, the shortest time dispatch plan will be used. If flood control and the installation of preventive facilities are required, the maximum force dispatch plan will be used. If a disaster with a large affected area is required, such as a large-scale typhoon, the least impact dispatch plan will be used. It is understood that the present invention is not limited to this.

[0144] In an embodiment of the present invention, a method for intelligent dispatch of emergency support forces and generation of program summary text combining integer programming and large language model is proposed. On the one hand, based on disaster characteristics and resource constraints, by combining integer programming and large language model, the optimal dispatch plan for emergency support is adaptively generated in the resource dispatch process, including the dispatch plan with the shortest time, the dispatch plan with the most force, and the dispatch plan with the least impact, so as to meet the needs of different emergency support scenarios. The specific dispatch content, such as the number of support personnel, the name of the support unit, and the key information such as the assisted city, will automatically generate a dispatch order file through the system, and the dispatch process will be efficient, accurate and standardized. On the other hand, based on the dispatch results, an intuitive and easy-to-understand summary text is generated through a large language model to provide support for decision makers to quickly understand the dispatch results. The technical solution provided by the embodiment of the present invention, by combining the optimization algorithm with the natural language generation technology, can not only improve the efficiency and accuracy of fire rescue dispatch, but also significantly enhance the readability and user experience of the dispatch results, providing an innovative, efficient and intelligent solution for disaster emergency management, and has important reference significance for other fields that require efficient dispatch and information generation.

[0145] For better explanation, refer to Figure 3 , showing an overall flow diagram of an intelligent dispatching method for emergency support forces provided by an embodiment of the present invention. It should be pointed out that this embodiment only briefly describes the general process of intelligent dispatching of emergency support forces. The specific implementation process of each step can be understood by referring to the relevant content in the aforementioned embodiment. It will not be described here. It can be understood that the present invention does not limit this.

[0146] Step 301: determine the disaster-stricken city to be supported, and obtain emergency resource distribution data other than the disaster-stricken city within the preset jurisdiction;

[0147] Step 302: Based on the emergency resource distribution data, optimal dispatch integer programming is performed for the disaster-stricken cities based on different emergency support needs, and a dispatch plan with the shortest time, the most powerful force, and the least impact is generated;

[0148] Step 303: Generate dispatch order files for the shortest time dispatch plan, the most powerful dispatch plan, and the least impact dispatch plan respectively;

[0149] Step 304: Use a large language model to first summarize the dispatch plan with the shortest time, the dispatch plan with the most force, and the dispatch plan with the least impact, and then perform a secondary summary based on the summary results of each plan to obtain a final summary text.

[0150] Reference Figure 4, shows a structural block diagram of an intelligent dispatching device for emergency support forces provided by an embodiment of the present invention, which may specifically include:

[0151] The data acquisition unit 401 is used to determine the disaster-stricken city to be supported and obtain the emergency resource distribution data other than the disaster-stricken city within the preset jurisdiction;

[0152] The optimal dispatch integer programming unit 402 is used to perform optimal dispatch integer programming based on different emergency support needs for the disaster-stricken city according to the emergency resource distribution data, and generate an emergency support dispatch plan corresponding to each emergency support need;

[0153] The dispatch order file generating unit 403 is used to generate dispatch order files for each of the emergency support dispatch plans respectively, and the dispatch order files are used to provide dispatch decision references to users.

[0154] In an optional embodiment, the emergency resource distribution data includes fire rescue detachments in all unaffected areas within the preset jurisdiction; the optimal dispatch integer programming unit 402 includes:

[0155] An emergency dispatch objective function construction unit is used to construct an emergency dispatch objective function by taking the fire rescue detachments in all non-disaster-affected areas within the preset jurisdiction as planning objects and minimizing the driving time of the planning objects as the goal;

[0156] A shortest time dispatch plan generation unit is used to construct a shortest time support demand constraint, combine the emergency dispatch objective function with the shortest time support demand constraint to construct a shortest time dispatch demand model, solve the shortest time dispatch demand model through an integer programming solver, and obtain a shortest time dispatch plan;

[0157] A maximum force dispatch plan generating unit is used to construct the maximum force support demand constraint based on the shortest time support demand constraint, to construct the maximum force dispatch demand model by combining the emergency dispatch objective function with the maximum force support demand constraint, to solve the maximum force dispatch demand model by an integer programming solver, and to obtain the maximum force dispatch plan;

[0158] The least-impact dispatch plan generation unit is used to construct a least-impact support demand constraint based on the shortest time support demand constraint, and to construct a least-impact scheduling demand model by combining the emergency dispatch objective function with the least-impact support demand constraint. The least-impact scheduling demand model is solved by an integer programming solver to obtain a least-impact dispatch plan.

[0159] In an optional embodiment, each of the fire rescue detachments is composed of a plurality of flood fighting and rescue teams and a plurality of water rescue professional teams; the shortest time dispatch plan generating unit includes:

[0160] An emergency dispatch quantity constraint construction unit, used to construct an emergency dispatch quantity constraint based on the number of teams of the flood fighting and rescue team and the number of teams of the water rescue professional team;

[0161] An emergency support number constraint construction unit is used to obtain the reinforcement force demand of the disaster-stricken city, and construct an emergency support number constraint based on the reinforcement force demand, the dispatched quantity of the flood fighting and rescue team, and the dispatched quantity of the water rescue professional team;

[0162] An emergency dispatch quantity non-negative constraint construction unit, used to construct an emergency dispatch quantity non-negative constraint based on the dispatch quantity of the flood fighting and rescue team and the dispatch quantity of the water rescue professional team;

[0163] The shortest time support demand constraint integration unit is used to integrate the emergency dispatch quantity constraint, the emergency support number constraint and the emergency dispatch quantity non-negative constraint to construct the shortest time support demand constraint.

[0164] In an optional embodiment, the fire rescue detachments in all unaffected areas include a plurality of first-level fire rescue detachments and a plurality of second-level fire rescue detachments, and the support force of the first-level fire rescue detachment is stronger than that of the second-level fire rescue detachment; the force-maximizing dispatch plan generating unit includes:

[0165] A strongest force priority call constraint construction unit, used for constructing the strongest force priority call constraint based on the mutually exclusive reaction between the first-level fire rescue detachment and the second-level fire rescue detachment;

[0166] The strongest support requirement constraint integration unit is used to integrate the shortest time support requirement constraint and the strongest priority call constraint to construct the strongest support requirement constraint.

[0167] In an optional embodiment, the fire rescue detachments in all non-disaster-affected areas include fire rescue detachments in adjacent cities and fire rescue detachments in non-adjacent cities; the minimum impact dispatch plan generation unit includes:

[0168] A resource circle support priority call constraint construction unit, used to construct a resource circle support priority call constraint based on the mutually exclusive reaction between the adjacent city fire rescue detachment and the non-adjacent city fire rescue detachment;

[0169] The least-impact support requirement constraint integration unit is used to integrate the shortest-time support requirement constraint and the resource circle support priority call constraint to construct the least-impact support requirement constraint.

[0170] In an optional embodiment, the dispatch order file generating unit 403 includes:

[0171] An emergency dispatch situation analysis unit, used to analyze the emergency dispatch situation of the disaster-stricken city based on each of the emergency support dispatch plans;

[0172] A fire rescue detachment merging unit, for merging fire rescue detachments with the same number of emergency support dispatchers according to the analysis results for each of the emergency support dispatch plans;

[0173] The dispatch order file generation subunit is used to generate the dispatch order files of each of the emergency support dispatch plans respectively by using a PDF generator based on the merging result.

[0174] In an optional embodiment, the intelligent dispatching device for emergency support force further includes:

[0175] A scheme summarizing unit is used to use a locally deployed large language model, or call a cloud-based large language model API, to summarize the shortest time dispatch scheme, the most forceful dispatch scheme, and the least impact dispatch scheme, respectively, and obtain a shortest time dispatch summary text, a most forceful dispatch summary text, and a least impact dispatch summary text;

[0176] A content supplement unit, used to supplement the content of the summary text of the strongest detachment based on the keyword of the strongest detachment, and to supplement the content of the summary text of the least influential dispatch based on the keyword of the adjacent city;

[0177] A task instruction text acquisition unit, used to acquire the task instruction text input by the user;

[0178] An emergency support keyword text combination unit, used to combine the task instruction text, the shortest time dispatch summary text, the strongest force dispatch summary text after replenishment, and the smallest impact dispatch summary text after replenishment to obtain an emergency support keyword text;

[0179] The secondary solution summary unit is used to use the locally deployed large language model, or call the cloud-based large language model API to perform a secondary solution summary on the emergency support keyword text to obtain a final solution summary text.

[0180] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the aforementioned method embodiment.

[0181] An embodiment of the present invention further provides an electronic device, the device comprising a processor and a memory:

[0182] The memory is used to store the program code and transmit the program code to the processor;

[0183] The processor is used to execute the intelligent dispatching method of emergency support force of any embodiment of the present invention according to the instructions in the program code.

[0184] An embodiment of the present invention further provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the intelligent dispatching method for emergency support forces of any embodiment of the present invention.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0186] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0187] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0188] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0189] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0190] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligently dispatching emergency support forces, characterized in that: include: Determine the disaster-stricken cities to be supported, and obtain emergency resource distribution data other than the disaster-stricken cities within the preset jurisdiction; According to the emergency resource distribution data, optimal dispatch integer programming is performed on the disaster-stricken cities based on different emergency support needs, and an emergency support dispatch plan corresponding to each emergency support need is generated; A dispatch order file is generated for each of the emergency support dispatch plans, and the dispatch order file is used to provide a dispatch decision reference to the user.

2. The intelligent dispatching method for emergency support forces according to claim 1 is characterized in that: The emergency resource distribution data includes fire rescue detachments in all non-disaster-affected areas within the preset jurisdiction; the optimal dispatch integer programming based on different emergency support needs is performed on the disaster-affected city according to the emergency resource distribution data, and an emergency support dispatch plan corresponding to each emergency support need is generated, including: Taking all fire rescue detachments in non-disaster areas within the preset jurisdiction as planning objects and minimizing the driving time of the planning objects as the goal, an emergency dispatch objective function is constructed; Constructing the shortest time support demand constraint, combining the emergency dispatch objective function with the shortest time support demand constraint to construct the shortest time dispatch demand model, solving the shortest time dispatch demand model through an integer programming solver, and obtaining the shortest time dispatch plan; On the basis of the shortest time support demand constraint, the strongest force support demand constraint is constructed, the strongest force dispatch demand model is constructed by combining the emergency dispatch objective function with the strongest force support demand constraint, and the strongest force dispatch demand model is solved by an integer programming solver to obtain the strongest force dispatch plan; On the basis of the shortest time support demand constraint, the least impact support demand constraint is constructed, and the least impact scheduling demand model is constructed by combining the emergency dispatch objective function and the least impact support demand constraint. The least impact scheduling demand model is solved by an integer programming solver to obtain the least impact dispatch plan.

3. The intelligent dispatching method for emergency support forces according to claim 2 is characterized in that: Each fire rescue brigade is composed of multiple flood fighting and rescue teams and multiple water rescue professional teams; the shortest support demand constraints for the construction time include: Based on the number of flood fighting and rescue teams and the number of professional water rescue teams, a constraint on the number of emergency dispatches is established; Obtain the reinforcement demand of the disaster-stricken city, and establish a constraint on the number of emergency support personnel based on the reinforcement demand, the dispatched quantity of the flood fighting and rescue team, and the dispatched quantity of the professional water rescue team; Based on the dispatch volume of the flood fighting and rescue team and the dispatch volume of the water rescue professional team, a non-negative constraint on the emergency dispatch volume is constructed; The emergency dispatch quantity constraint, the emergency support number constraint and the emergency dispatch quantity non-negative constraint are integrated to construct the shortest time support demand constraint.

4. The intelligent dispatching method for emergency support forces according to claim 3 is characterized in that: The fire rescue detachments in all unaffected areas include a plurality of first-level fire rescue detachments and a plurality of second-level fire rescue detachments, and the support force of the first-level fire rescue detachment is stronger than that of the second-level fire rescue detachment; On the basis of the shortest support requirement constraint, the strongest support requirement constraint is constructed, including: Based on the mutually exclusive reaction between the first-level fire rescue detachment and the second-level fire rescue detachment, a constraint of calling with the strongest force first is established; The shortest support requirement constraint and the strongest force priority call constraint are integrated to construct the strongest force support requirement constraint.

5. The intelligent dispatching method for emergency support forces according to claim 3 is characterized in that: The fire rescue detachments in all unaffected areas include the fire rescue detachments in the adjacent cities and the fire rescue detachments in the non-adjacent cities; based on the shortest time support demand constraint, the minimum impact support demand constraint is constructed, including: Based on the mutually exclusive reaction between the fire rescue detachment of the adjacent city and the fire rescue detachment of the non-adjacent city, a resource circle support priority call constraint is established; The shortest time support requirement constraint and the resource circle support priority call constraint are integrated to construct the least impact support requirement constraint.

6. The intelligent dispatching method for emergency support forces according to any one of claims 2 to 5, characterized in that: The step of generating the dispatch order files for each of the emergency support dispatch plans comprises: Analyzing the emergency dispatch situation of the disaster-stricken cities based on each of the emergency support dispatch plans; For each of the emergency support dispatch plans, based on the analysis results, the fire rescue detachments with the same number of emergency support dispatch personnel are merged; Based on the merging result, a PDF generator is used to generate dispatch order files for each of the emergency support dispatch plans.

7. The intelligent dispatching method for emergency support forces according to claim 6 is characterized in that: Also includes: Using a locally deployed large language model, or calling a cloud-based large language model API, respectively summarize the shortest time dispatch plan, the most forceful dispatch plan, and the least impact dispatch plan, and obtain a shortest time dispatch summary text, a most forceful dispatch summary text, and a least impact dispatch summary text; Based on the keywords of the strongest detachment, the content of the summary text of the strongest dispatch is supplemented, and based on the keywords of the neighboring cities, the content of the summary text of the dispatch with the least impact is supplemented; Get the task instruction text entered by the user; Combine the task instruction text, the shortest time dispatch summary text, the strongest force dispatch summary text after replenishment, and the smallest impact dispatch summary text after replenishment to obtain an emergency support keyword text; A locally deployed large language model is used, or a cloud-based large language model API is called to perform a secondary solution summary on the emergency support keyword text to obtain a final solution summary text.

8. An intelligent dispatching device for emergency support forces, characterized in that: include: A data acquisition unit, used to determine the disaster-stricken city to be supported, and to acquire emergency resource distribution data other than the disaster-stricken city within a preset jurisdiction; An optimal dispatch integer programming unit, used to perform optimal dispatch integer programming based on different emergency support needs for the disaster-stricken city according to the emergency resource distribution data, and generate an emergency support dispatch plan corresponding to each emergency support need; The dispatch order file generating unit is used to generate dispatch order files for each of the emergency support dispatch plans respectively, and the dispatch order files are used to provide dispatch decision references to users.

9. An electronic device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the intelligent dispatching method of emergency support force according to any one of claims 1-7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the intelligent dispatching method for emergency support forces described in any one of claims 1-7.

Images