Emergency resource allocation decision support method based on space-time dynamic deduction

Abstract

The invention discloses an emergency resource allocation decision support method based on spatio-temporal dynamic deduction, and relates to the technical field of emergency management, comprising the following steps: pre-constructing a multi-source spatio-temporal data acquisition module, collecting emergency resource basic information, emergency real-time spatio-temporal data, historical disposal case data, geographic information data, environment dynamic data and resource movement track data; and carrying out data preprocessing and time-space fusion. According to the method, fundamental transformation of emergency management from passive response to active pre-judgment is realized through deep integration of space-time dynamic deduction and a multi-dimensional resource configuration technology, various kinds of scattered information are converted into an organic linkage overall situation through construction of an event resource geographic three-dimensional association system, and the emergency management efficiency is improved. The emergency command personnel can comprehensively and intuitively master the full view of the event and the resource distribution, the control capability of the complex situation is greatly improved, and the systematic advantage is more outstanding especially when the emergency event with large-range and multi-chain diffusion is coped with.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of emergency management, in particular to an emergency resource allocation decision support method based on space-time dynamic deduction. BACKGROUND

[0002] At present, various types of emergencies present complex and changeable characteristics, not only with fast spreading speed and wide influence range, but also with significant space-time dynamic changes, which significantly increase the difficulty of emergency disposal. Traditional emergency resource allocation methods mostly rely on static data and experience-based decision making, which is difficult to accurately capture the space-time evolution law of the event, resulting in the difficulty of "time lag" and "space mismatch" in resource scheduling. For example, in flood disasters, due to the inability to predict the spread path and speed of the flood in advance, rescue materials may be transported to the flooded area, causing resource waste; in the scenario of epidemic spread, the static resource allocation mode is difficult to match the dynamic demand changes in different periods and different regions, affecting the timeliness and effectiveness of emergency response.

[0003] The development of big data analysis technology provides the possibility for integrating multi-source emergency data, but the existing methods do not sufficiently mine the space-time correlation between data, and fail to form a dynamic perception and prediction ability for the whole cycle of the emergency. This makes emergency decision-making often stay at the passive response level, and it is difficult to achieve forward-looking layout of resource allocation. Especially in the face of complex emergencies, such as secondary disasters caused by earthquakes, floods, etc. or new types of emergencies, the existing technology lacks effective data fusion mechanism and deduction model, and cannot provide scientific and accurate decision support for emergency command, so there is an urgent need for an emergency resource allocation and decision support method that integrates big data analysis and space-time dynamic deduction technology to break through the limitations of traditional modes.

[0004] In view of the problems in the related art, no effective solution has been proposed so far. SUMMARY

[0005] In view of the problems in the related art, the present application proposes an emergency resource allocation decision support method based on space-time dynamic deduction to overcome the above technical problems existing in the prior art.

[0006] The technical solution of the present application is as follows:

[0007] An emergency resource allocation decision support method based on space-time dynamic deduction, comprising the following steps:

[0008] Pre-constructing a multi-source space-time data acquisition module to collect emergency resource basic information, real-time space-time data of emergencies, historical disposal case data, geographic information data, environmental dynamic data and resource movement trajectory data;

[0009] Data preprocessing and spatio-temporal fusion are performed, including cleaning, conversion and standardization processing of the collected multi-source heterogeneous data, construction of an event resource geographic three-dimensional data model through a spatio-temporal correlation algorithm, and formation of a structured spatio-temporal database;

[0010] Resource monitoring and spatio-temporal situation assessment are performed, including visualization monitoring of emergency resource distribution, state and evolution of the emergency event based on real-time data and in combination with a geographic information system (GIS), and dynamic generation of a resource supply-demand spatio-temporal matching degree report through a spatio-temporal situation assessment model;

[0011] Spatio-temporal dynamic deduction and demand prediction are performed, including construction of an emergency event spatio-temporal evolution model, deduction of an event diffusion track in combination with historical cases and real-time data, and prediction of emergency resource demand at different time nodes and in different regions by using a spatio-temporal sequence prediction algorithm;

[0012] Dynamic configuration is performed, including construction of a resource optimization configuration model under spatio-temporal constraints based on spatio-temporal deduction results and resource monitoring data, and generation of a dynamic scheduling scheme covering three-dimensional dimensions of time, space and resource types;

[0013] Intelligent decision support is performed, including fusion of spatio-temporal deduction conclusions, resource configuration schemes and historical cases, generation of targeted decision suggestions through spatio-temporal correlation rule reasoning, and pushing of the decision suggestions to an emergency command platform.

[0014] The spatio-temporal data includes longitude and latitude of an emergency event occurrence site, diffusion speed and direction, influence range data changing with time, emergency resource storage location, GPS data of a moving track, time series data of road network traffic state, spatial attribute data of population density and infrastructure distribution in a geographic partition, air and soil pollutant concentration data in a special event such as chemical leakage, and backup data source information of real-time guarantee data.

[0015] The data preprocessing and spatio-temporal fusion include using a spatio-temporal alignment algorithm to map multi-source data collected asynchronously to a unified spatio-temporal coordinate system, and establishing an associated index of an event timestamp, a geographic coordinate and a resource ID through a spatio-temporal indexing technology; and calibrating specific definition methods of abnormal values of different types of emergency data.

[0016] The spatio-temporal situation assessment includes real-time labeling of resource location, quantity and state on a GIS map, and displaying a spatio-temporal distribution of an emergency event influence range and a resource supply-demand gap in the form of a dynamic heat map, and an evaluation index is represented as:

[0017] Spatio-temporal matching degree = (regional available resource x response timeliness) / (regional demand prediction value x spatial distance weight).

[0018] The spatio-temporal dynamic deduction and demand prediction comprises: constructing a spatio-temporal sequence prediction model by using an improved LSTM neural network, input parameters comprising an event type, an initial influence range, a geographical feature parameter, a historical similar event diffusion coefficient and real-time monitoring data, and outputting an event diffusion trajectory prediction in the future 1-6 hours and a resource demand curve of the corresponding area.

[0019] The dynamic configuration optimization target comprises: minimizing the time difference between the calibration resource arrival time and the event key node, minimizing the spatial overlap rate between the calibration resource transportation path and the event diffusion area, and maximizing the spatio-temporal coordination efficiency of the calibration cross-regional resource scheduling, and solving the optimal configuration scheme by using a spatio-temporal genetic algorithm.

[0020] The intelligent decision support comprises: converting the event spatio-temporal characteristics, resource configuration scheme and disposal effect association relationship in the historical cases into decision rules by constructing a spatio-temporal association rule base, and generating a decision scheme comprising a priority rescue area, a resource scheduling time window and a path obstacle avoidance suggestion in combination with the real-time deduction result.

[0021] The beneficial effects of the present application are:

[0022] 1. The present application realizes the fundamental change from passive response to active prediction of emergency management by deeply integrating spatio-temporal dynamic deduction and multi-dimensional resource configuration technology. The core advantage lies in accurately capturing the spatio-temporal evolution law of the sudden event, relying on the dynamic deduction model to prospectively grasp the event diffusion trajectory and resource demand change, so that the resource scheduling can be arranged in advance and accurately, effectively avoiding the problems of resource lag and mismatch in traditional static configuration. At the same time, by constructing an event resource geographical three-dimensional association system, various scattered information is converted into an organic and integrated whole situation, so that the emergency command personnel can comprehensively and intuitively master the whole picture of the event and the resource distribution, greatly improving the control ability of complex situation, especially in dealing with sudden events with large range and multi-chain diffusion, the systematic advantage is more prominent.

[0023] 2. The present application breaks through the traditional mode driven by experience, and forms a scientific, dynamic and operable intelligent decision system. By integrating the spatio-temporal deduction conclusion, the optimized configuration scheme and the historical experience, the generated decision suggestion contains not only the macro rescue priority and resource scheduling direction, but also the specific time window, path planning and other operation details, and can be adjusted in real time according to the event dynamics, fully adapting to the uncertainty of emergency scene. This multi-level and adaptive decision support mechanism not only provides scientific basis for high-level command at the strategic level, but also provides clear action guidelines for grassroots execution personnel, significantly improving the overall efficiency and decision scientificity of emergency response, and effectively enhancing the comprehensive ability of the city to deal with various complex sudden events. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description only aim to some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0025] Figure 1 FIG. 1 is a flow diagram of a method for emergency resource allocation decision support based on space-time dynamic deduction according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art belong to the scope of protection of the present application.

[0027] According to an embodiment of the present application, a method for emergency resource allocation decision support based on space-time dynamic deduction is provided.

[0028] As shown in FIG. 1, the method for emergency resource allocation decision support based on space-time dynamic deduction according to an embodiment of the present application comprises the following steps: Figure 1

[0029] Step S1, a multi-source space-time data acquisition module is constructed in advance to acquire the following data:

[0030] Emergency resource basic information: resource type, quantity, storage location, belonging unit, available state, transportation vehicle parameter, resource quality and function information;

[0031] Real-time space-time data of emergency events: event type, occurrence time, initial location, influence range, diffusion speed and direction, influence range change data with time, air and soil pollutant concentration data in special events such as chemical leakage;

[0032] Historical disposal case data: space-time characteristics of historical events, such as earthquake epicenter migration, fire spread path, resource allocation scheme, disposal effect evaluation;

[0033] Geographical information data: road network topology structure, key node location such as bridge / tunnel, terrain elevation data, administrative boundary;

[0034] Environmental dynamic data: real-time meteorological data, hydrological data, traffic congestion index time series;

[0035] ​Resource movement trajectory data: real-time GPS trajectory, driving speed, and estimated arrival time of rescue vehicles and material transport vehicles;

[0036] Backup data source information: when the main data source fails or transmission is delayed, the related information of the backup data source that can be quickly switched to ensure real-time data.

[0037] Step S2, data preprocessing and spatio-temporal fusion, including the following steps:

[0038] Remove outliers, and clearly define the specific method for defining outliers of different types of emergency data, such as defining resource quantity outliers as values exceeding the regular reserve of the type of resource ± 50%, and defining meteorological data outliers by referring to the historical same period data fluctuation range; Fill in the missing values, which can be completed based on the spatio-temporal interpolation algorithm to complete the monitoring data of global and local areas;

[0039] Convert unstructured data into structured spatio-temporal coordinates. To solve the conversion accuracy problem, multiple algorithms are used to convert and calculate errors. When the error exceeds the preset threshold, manual correction is performed. Text event descriptions are analyzed into quantified diffusion vectors, representing the direction angle and speed value;

[0040] Through the spatio-temporal alignment algorithm, map multiple source data to a unified time-latitude-longitude coordinate system using dynamic time warping (DTW) based time alignment or coordinate projection based space alignment; Build an event resource geographic three-dimensional correlation model, establish data correlation through spatio-temporal indexing technology, and realize quick query of event status and resource distribution in a certain spatio-temporal range.

[0041] Step S3, resource monitoring and spatio-temporal situation assessment, including the following steps:

[0042] Based on the GIS platform, display in the form of dynamic layers, including: current impact range of the event, resource storage point and real-time location, transport vehicle trajectory, and trapped personnel distribution heat map;

[0043] Spatio-temporal situation assessment, build a resource supply-demand spatio-temporal matching degree model, the calculation formula is:

[0044] Spatio-temporal matching degree = (total amount of available resources in the region × transportation timeliness coefficient × resource quality function correction coefficient) / (predicted demand value in the region × spatial distance weight);

[0045] Wherein, the transportation timeliness coefficient = 1-(estimated transportation time / emergency response threshold time), the spatial distance weight = 1 / (straight line distance from the center of the region to the resource point + road correction coefficient), and the resource quality function correction coefficient is set according to the degree to which the resource meets the special needs of the event, ranging from 0.5 to 1.5;

[0046] When the spatio-temporal matching degree of a certain area is lower than the preset threshold, an early warning is automatically triggered, and the area is marked as a resource shortage area.

[0047] Step S4, spatio-temporal dynamic deduction and demand prediction, including the following steps:

[0048] Select a corresponding deduction algorithm according to the event type, which includes a fire spreading model based on wind speed and combustible material distribution, a flood submergence range model based on terrain and hydrological data, and a model fusion mechanism for complex events such as fire caused by earthquakes and floods. The deduction results of each single event model are integrated; input real-time monitoring data and environmental parameters to deduce the spatio-temporal changes of the event influence range within 1-6 hours in the future;

[0049] Demand prediction is performed based on the spatio-temporal evolution results of the event deduction, combined with the "event intensity and resource demand" correlation law of historical cases, using a spatio-temporal sequence prediction algorithm to predict the emergency resource demand at different time nodes and different geographical partitions in the future. The output is a "resource type, time, and region" three-dimensional demand matrix; when a new type of emergency event occurs, the prediction results are corrected based on expert experience and similar event characteristics.

[0050] Step S5, dynamic configuration, including the following steps:

[0051] A resource optimization configuration model under spatio-temporal constraints is constructed, and the objective function is:

[0052] min(Σ resource transportation time difference) + min(Σ spatial redundancy rate) + max(Σ demand coverage rate)

[0053] Wherein, transportation time difference = resource actual arrival time - event deduction demand peak time; spatial redundancy rate = (allocated resource quantity - actual demand) / actual demand (to avoid over-allocation); demand coverage rate = satisfied demand quantity / total demand quantity;

[0054] The calibration constraints include road capacity, which can limit the number of transport vehicles based on real-time traffic data, total resource quantity limit, and key nodes such as hospitals and shelters that are given priority protection;

[0055] Spatio-temporal genetic algorithm is used for solution, and efficiency optimization is performed on the algorithm, such as parallel computing technology and simplifying the calculation complexity of low-impact areas; the chromosome coding in the algorithm contains four-dimensional information of "resource ID-target area-departure time-transportation path", and the configuration scheme is optimized through crossover and mutation operations to generate a dynamic scheduling plan covering the next 6 hours.

[0056] Step S6, intelligent decision-making, including the following steps:

[0057] A spatio-temporal rule library is constructed, and rules are mined from historical cases, such as "when a flood is expected to submerge an area of more than 5 km 2 and the wind speed is greater than 3 levels, preferentially deploy rescue vehicles with amphibious capability" "in mountainous terrain, the resource transportation time needs to increase the terrain correction value of 30% on the basis time"; and

[0058] A decision scheme is generated, combining the event spatio-temporal deduction result, the resource dynamic configuration scheme and the rule library, to generate a multi-level decision suggestion, as follows:

[0059] Spatio-temporal priority suggestion: mark the "high-priority rescue area" within the next 2 hours, which can be based on the densely populated area deduced by the event and the area with the largest resource gap;

[0060] Resource scheduling time window: recommend the optimal departure time of each resource, for example, "the ambulance needs to depart 20 minutes before t0 to avoid the bridge congestion expected to occur at t0+40 minutes";

[0061] Path obstacle avoidance suggestion: based on the dangerous area deduced by the event, for example, "the B road segment will be submerged within the next 1 hour, and it is recommended to bypass the C tunnel for the transportation vehicle";

[0062] In addition, a mechanism is provided for quickly adjusting the decision suggestion according to real-time sudden factors, such as sudden traffic control and temporary damage of resources. By comparing the real-time monitoring data with the preset threshold, when the adjustment condition is triggered, a new decision suggestion is automatically recalculated and generated. Different decision contents are provided for different levels of emergency command personnel. The high-level command personnel push the macro decision direction and overall situation analysis, and the grassroots execution personnel push the specific operation steps and detailed requirements. The decision suggestion is pushed to the emergency command platform to support visual display, such as marking the recommended path and time node on the GIS map.

[0063] Specifically, with the aid of the above scheme, the specific implementation process of the present application is illustrated by taking urban flood disaster as an example:

[0064] Real-time water level data is collected in advance by water level sensors deployed at various places in the city, which contains latitude and longitude and time stamp; 6-hour precipitation prediction is obtained through meteorological satellite, and the corresponding backup meteorological data source information is recorded; the geographical range of the waterlogged road section is identified through the video monitoring of the urban management department; the storage location, quantity, quality and function information of flood control materials such as water pumps, sandbags and inflatable boats are obtained through the database of the emergency management bureau; real-time road conditions and bridge height limit information are obtained through the API of the traffic department, and the backup acquisition channel when the traffic data transmission delay is clear.

[0065] Convert water level data into "time, latitude, longitude, water level value" three-dimensional data; perform image recognition on the waterlogged road segment video, convert it into polygon coordinates of the waterlogged area using multi-algorithm fusion, and calculate the error, when the error exceeds 5%, manually correct; through the spatio-temporal alignment algorithm, align the precipitation prediction data, such as time granularity 1 hour, with the water level monitoring data, such as time granularity 5 minutes, to the unified time axis; build a correlation model to automatically associate the flood control material inventory and the nearest transportation route when querying the water level at time t in area x.

[0066] On the GIS platform, display the current water depth distribution in a blue gradient heat map, and mark the dangerous areas where the water depth exceeds 50 cm with a red polygon; mark the flood control material storage points with green dots, and the size of the dots represents the quantity; calculate the spatio-temporal matching degree of each street, and correct it considering the power of the water pump and other quality factors, such as "East City Street current waterlogged area 5 km 2 , needs 20 water pumps, the nearby warehouse only has 8 high-efficiency water pumps, the quality function correction coefficient is 1.2, the transportation time efficiency coefficient is 0.7, the spatio-temporal matching degree = (8*0.7*1.2) / (20*1.2) = 0.28 < 0.6, trigger resource shortage warning".

[0067] Spatio-temporal dynamic deduction and demand prediction: use hydrodynamic model to input current water level, precipitation prediction, and terrain data to deduce the future 6-hour waterlogged area expansion trajectory: t1 = 1 hour later, the waterlogged area in East City Street will expand to the northwest to a residential area; t2 = 2 hours later, a bridge may be submerged. According to the rule in historical data that "water depth increases by 10 cm, water pump quantity increases by 5%", predict that East City Street needs 25 water pumps at t1 and 10 inflatable boats for resident relocation at t2.

[0068] Dynamic configuration: based on the deduction results, build an optimization model with the goal of "ensuring that water pumps are in place by t1 and inflatable boats are in place by t2, and the transportation route avoids future submerged areas". Use the optimized spatio-temporal genetic algorithm to solve it, use parallel computing to improve efficiency, and generate the scheme: deploy 15 water pumps from the West City warehouse, transport them via the unaffected outer ring highway, and ensure that they depart within 40 minutes of the current time to avoid congestion that may occur in 1 hour; deploy 10 inflatable boats from the emergency dock, travel via the standby waterway, and ensure that they arrive by t2.

[0069] Decision support: based on the rule base, generate recommendations, push macro information such as "East City residential area is a high-priority rescue area for the next 2 hours" to high-level commanders, and push specific operation steps such as "ambulance needs to depart 20 minutes before t0" to grassroots execution personnel; when a sudden traffic control is detected on a transportation road, automatically recalculate and generate a new detour recommendation; mark the recommended path and time nodes on the GIS map to assist commanders in decision-making.

[0070] In summary, by means of the above technical solutions of the present application, the following effects can be achieved:

[0071] 1、The present application realizes the fundamental change of emergency management from passive response to active prediction by deeply fusing spatio-temporal dynamic deduction and multi-dimensional resource allocation technology. Its core advantage lies in accurately capturing the spatio-temporal evolution law of the emergency, relying on the dynamic deduction model to predict the event diffusion track and resource demand changes, so that resource scheduling can be arranged in advance and precise force can be exerted, effectively avoiding the common problems of resource lag and mismatch in traditional static configuration. At the same time, by constructing an event resource geographic three-dimensional correlation system, various scattered information is transformed into an organic whole situation, so that emergency command personnel can comprehensively and intuitively master the whole picture of the event and the distribution of resources, greatly improving the control ability of complex situation, especially when dealing with large-scale, multi-chain diffusion of emergencies, the systematic advantage is more prominent.

[0072] 2、The present application breaks through the traditional mode driven by experience and forms a scientific, dynamic and operable intelligent decision system. By integrating spatio-temporal deduction conclusions, optimizing configuration schemes and historical experience, the generated decision suggestions not only contain macro rescue priority and resource scheduling direction, but also cover specific time window, path planning and other operation details, and can be adjusted in real time according to the event dynamics, fully adapting to the uncertainty of emergency scene. This multi-level and adaptive decision support mechanism not only provides scientific basis for high-level command at the strategic level, but also provides clear action guidelines for grassroots execution personnel, significantly improving the overall efficiency and decision scientificity of emergency response, and effectively enhancing the comprehensive ability of the city to deal with various complex emergencies.

[0073] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Those skilled in the art will easily think of other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and examples are only considered as exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

[0074] It should be understood that the present disclosure is not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A decision support method for emergency resource allocation based on spatiotemporal dynamic simulation, characterized in that, Includes the following steps: A multi-source spatiotemporal data acquisition module is pre-built to collect basic information on emergency resources, real-time spatiotemporal data of emergencies, historical case data of handling incidents, geographic information data, environmental dynamic data, and resource movement trajectory data; Data preprocessing and spatiotemporal fusion are carried out, including cleaning, transforming and standardizing the collected multi-source heterogeneous data, constructing a three-dimensional geographic data model of event resources through spatiotemporal correlation algorithms, and forming a structured spatiotemporal database. Conduct resource monitoring and spatiotemporal situation assessment, including visual monitoring of emergency resource distribution, status and emergency event evolution based on real-time data and combined with Geographic Information System (GIS), and dynamically generate resource supply and demand spatiotemporal matching reports through spatiotemporal situation assessment models; Conduct spatiotemporal dynamic simulation and demand forecasting, including building a spatiotemporal evolution model of emergencies, combining historical cases and real-time data to simulate the event spread trajectory, and using spatiotemporal sequence prediction algorithms to predict emergency resource demands at different time points and in different regions; Dynamic configuration is performed, including constructing a resource optimization configuration model under spatiotemporal constraints based on spatiotemporal simulation results and resource monitoring data, and generating a dynamic scheduling scheme covering three dimensions of time, space, and resource types. Intelligent decision support is provided, including integrating spatiotemporal simulation conclusions, resource allocation plans, and historical cases, generating targeted decision suggestions through spatiotemporal correlation rules, and pushing them to the emergency command platform.

2. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 1, characterized in that, The spatiotemporal data includes: latitude and longitude of the location of the emergency, data on the speed and direction of spread, and data on the range of impact over time; GPS data on the location of emergency resources and their movement trajectories; time series data on the traffic status of the road network; spatial attribute data on population density and infrastructure distribution within a geographical region; data on the concentration of air and soil pollutants in special events such as chemical leaks; and backup data source information to ensure the real-time nature of the data.

3. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 2, characterized in that, The data preprocessing and spatiotemporal fusion process includes using a spatiotemporal alignment algorithm to map asynchronously collected multi-source data to a unified spatiotemporal coordinate system, establishing an associated index of event timestamps, geographic coordinates, and resource IDs through spatiotemporal indexing technology, and defining specific methods for identifying outliers in different types of emergency data.

4. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 1, characterized in that, The aforementioned spatiotemporal situation assessment includes real-time labeling of resource location, quantity, and status using GIS maps, and displaying the spatiotemporal distribution of the impact range of the emergency and the resource supply-demand gap in the form of a dynamic heat map. The assessment indicators are expressed as follows: Spatiotemporal matching degree = (Available resources in the region × Response timeliness) / (Forecasted demand in the region × Spatial distance weight).

5. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 1, characterized in that, The process of performing spatiotemporal dynamic simulation and demand forecasting includes: constructing a spatiotemporal sequence prediction model using an improved LSTM neural network, with input parameters including event type, initial impact range, geographical feature parameters, historical diffusion coefficients of similar events, and real-time monitoring data, and outputting a predicted event diffusion trajectory for the next 1-6 hours and a corresponding resource demand curve for the region.

6. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 1, characterized in that, The optimization objectives of the dynamic configuration include: minimizing the time difference between the arrival time of the calibrated resources and the key nodes of the event, minimizing the spatial overlap between the calibrated resource transportation path and the event diffusion area, maximizing the spatiotemporal coordination efficiency of cross-regional resource scheduling, and using a spatiotemporal genetic algorithm to solve for the optimal configuration scheme.

7. The emergency resource allocation decision support method based on spatiotemporal dynamic simulation according to claim 1, characterized in that, The intelligent decision support includes: by constructing a spatiotemporal correlation rule base, transforming the spatiotemporal characteristics of events, resource allocation schemes and the correlation of disposal effects in historical cases into decision rules, and combining real-time simulation results to generate decision schemes that include priority rescue areas, resource scheduling time windows and path avoidance suggestions.

Images