Dynamic emergency resource scheduling method and system based on multi-dimensional data analysis and medium

Abstract

The invention discloses a dynamic emergency resource scheduling method based on multi-dimensional data analysis, and belongs to the technical field of emergency management and resource scheduling, and the method comprises the following steps: S1, collecting and fusing multi-dimensional data, and constructing a multi-source heterogeneous database; s2, disaster chain network construction: constructing a disaster chain evolution network model by applying a complex network theory based on the data in the step S1, calculating node degree, betweenness centrality and vulnerability, and identifying key risk nodes and evolution paths; s3, performing dynamic risk assessment, quantifying a current risk level and predicting a future risk situation; s4, performing an emergency resource optimization scheduling decision, and preferentially blocking a high-risk chained propagation path according to a risk assessment result in the step S3; and S5, scheduling and implementing feedback regulation and control. According to the method, multi-dimensional dynamic information such as real-time weather, infrastructure states and social disaster-bearing bodies is effectively fused, dynamic simulation and risk assessment are actively carried out on a disaster chain type evolution law, disaster spreading is effectively controlled, and overall loss is reduced.

Description

Technical Field

[0001] This invention belongs to the field of emergency management and resource scheduling technology, specifically involving a dynamic emergency resource scheduling method, system and storage medium based on multidimensional data analysis, which is applicable to the rapid response and optimal resource allocation of urban critical infrastructure under extreme weather conditions. Background Technology

[0002] With the intensification of global climate change, extreme weather events (such as torrential rains, high temperatures, and cold waves) are becoming more frequent and intense, posing a serious threat to critical infrastructure such as urban water and gas supply. These disasters often exhibit chain-like evolution characteristics, forming complex disaster chains. For example, torrential rains can cause urban flooding, which leads to pipeline ruptures, which in turn cause water supply disruptions, creating a domino effect that further leads to secondary disasters such as social disorder and economic losses. Traditional emergency management systems exhibit numerous limitations when facing these dynamic disaster scenarios caused by intensified global climate change and extreme weather events, including: reliance on historical statistics or single-source data, failing to effectively integrate real-time meteorological data, infrastructure status, and social disaster-bearing entities, resulting in a single data dimension; a lack of dynamic simulation and risk assessment of the chain-like evolution of disasters, making it impossible to predict secondary and derivative disasters, leading to passive and delayed resource allocation; and security barriers in cross-departmental data sharing, with dispatch decisions not fully considering multi-party collaboration and privacy protection, affecting overall rescue efficiency. Resource allocation schemes are often based on experience and fail to accurately match critical nodes and vulnerable paths in the disaster chain, making it difficult to achieve the goal of "disrupting the chain to mitigate disasters." Therefore, there is an urgent need for a new method that can dynamically perceive risks, accurately predict the situation, and optimize resource allocation. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a dynamic emergency resource scheduling method, system and storage medium based on multidimensional data analysis. By constructing a disaster chain network, dynamically assessing risks and optimizing decisions, it achieves efficient and accurate scheduling of emergency resources and solves the technical problems in the background art.

[0004] The objective of this invention is achieved as follows: a dynamic emergency resource scheduling method based on multidimensional data analysis, comprising the following steps: S1, multidimensional data acquisition and fusion, real-time acquisition of extreme weather disaster-causing factor data, urban critical infrastructure status data, social disaster-bearing body data, and historical disaster case data to construct a multi-source heterogeneous database; S2, disaster chain network construction, based on the data from step S1, using complex network theory to construct a disaster chain evolution network model, calculating node degree, betweenness centrality, and vulnerability, and identifying key risk nodes and evolution paths; S3, dynamic risk assessment, fusing real-time monitoring data with the disaster chain model, using Bayesian networks or dynamic trust assessment algorithms to quantify the current risk level and predict future risk trends; S4, emergency resource optimization scheduling decision-making, based on the risk assessment results of step S3, combined with resource inventory, geographic information, and scheduling constraints, using a multi-objective optimization algorithm to generate resource allocation schemes, prioritizing the blocking of high-risk chain propagation paths; S5, scheduling implementation feedback control, issuing scheduling instructions through the emergency command platform, and dynamically adjusting resource deployment based on real-time feedback data, forming a closed loop of assessment, decision-making, execution, and feedback.

[0005] By collecting and integrating multi-dimensional data from meteorology, infrastructure, and society in real time into a unified database, a disaster chain model is constructed using complex network theory to identify key risk points and evolution paths. Then, dynamic risk assessment and prediction are performed by combining real-time data with the disaster chain model. Based on the risk assessment results, resource reserves and geographical constraints are comprehensively considered, and an optimal resource scheduling plan is generated through a multi-objective optimization algorithm. Finally, instructions are issued through an emergency platform, and adjustments are made dynamically based on on-site feedback, forming a closed-loop management system. Its core principles are "data-driven decision-making" and "predictive intervention." By connecting isolated disaster information into a dynamically evolving "disaster chain network," the system can anticipate disaster development rather than simply responding to disasters that have already occurred. The multi-objective optimization algorithm seeks the optimal solution to block the most critical disaster path under multiple constraints (such as time, cost, and resource availability), realizing a shift from "experience-based scheduling" to "model-optimized scheduling."

[0006] By effectively integrating multi-dimensional dynamic information such as real-time meteorology, infrastructure status, and social disaster-bearing entities, the data dimensions are multi-dimensional, enabling a fundamental shift in emergency resource scheduling from passive response to proactive prediction and intervention. This allows for precise allocation of resources to key links in the disaster chain, proactive dynamic simulation and risk assessment of the disaster chain's evolution, early prediction of secondary and derivative disasters, and advance resource scheduling. It also fully considers multi-party collaboration and privacy protection, enhances rescue efficiency through cross-departmental data sharing, and precisely matches resource allocation plans with key nodes and vulnerable paths in the disaster chain. This achieves the goal of breaking the chain and mitigating disaster, improving scheduling efficiency and resource utilization, effectively controlling the spread of disasters, and reducing overall losses.

[0007] Furthermore, the multidimensional data collected and fused in step S1 includes meteorological data, infrastructure data, and social data. The meteorological data includes the intensity and spatiotemporal distribution data of extreme rainfall, high temperatures, low temperatures, snowfall, and cold waves. The infrastructure data includes the probability of water supply network rupture, the corrosion status of gas supply facilities, and power outage records. The social data includes population density, traffic flow, and the distribution of medical resources. In implementation, during the data collection phase, meteorological data (such as rainfall intensity), infrastructure data (such as the probability of network rupture), and social data (such as population density) are explicitly collected. This adopts the disaster systems theory, which states that disasters are the result of the combined effects of disaster-causing factors (meteorology), disaster-bearing bodies (infrastructure), and the social environment. For example, the risk level of a rainfall event of the same intensity will be much higher in densely populated areas with aging pipelines than in other areas, leading to different dispatch decisions. Only by comprehensively covering data in all three dimensions can a complete risk assessment be conducted, avoiding decision-making biases caused by missing data dimensions.

[0008] Furthermore, the disaster chain network construction described in step S2 includes: extracting disaster event chains using event tree analysis, defining nodes as disaster events and edges as evolutionary relationships; visualizing the network topology using Gephi or Pajek software, and calculating edge betweenness numbers, average path length, and connectivity. The disaster evolution sequence is extracted in a structured manner using event tree analysis, and visualized and quantitatively calculated using professional complex network analysis tools such as Gephi. Centrality indicators from complex network theory are used to identify key nodes and vulnerable edges in the network. For example, an evolutionary path with a high "edge betweenness number" indicates that it is a necessary path for multiple potential disaster chains; blocking it can effectively prevent the spread of disasters. This achieves objective and quantitative identification of key links in the disaster chain, replacing the traditional subjective judgment relying on expert experience, improving the scientific rigor and accuracy of risk analysis, providing a reliable basis for "using resources wisely," and enabling limited emergency resources to generate the greatest disaster reduction benefits.

[0009] Furthermore, the dynamic risk assessment in step S3 employs a dynamic trust assessment model based on trusted reports, integrating device status, environmental context, and behavioral logs. This "dynamic trust assessment model" comprehensively considers the device's operational status (reliability), its environment (context), and its historical behavioral logs to calculate its current credibility or risk value. Instead of treating the data from monitored devices as absolutely accurate, it treats it as an "object" requiring continuous credibility assessment, effectively addressing real-world issues such as device failures and data anomalies. Introducing the dynamic trust assessment model into the emergency data fusion process enhances the system's tolerance and anti-interference capabilities against data uncertainty and anomalies. It improves the robustness and reliability of risk assessment results; even if some monitored device data is abnormal, the system can identify and correct it through the model, avoiding erroneous scheduling decisions.

[0010] Furthermore, the optimized scheduling decision in step S4 includes: applying DS evidence theory to fuse multi-departmental data and generate a resource demand heatmap; prioritizing resource allocation to the most vulnerable disaster chain edges with the goal of chain disruption and disaster mitigation. The optimized scheduling decision also includes combining privacy computing with federated learning or TEE to achieve cross-domain data, making cross-domain data usable but invisible, thus ensuring data security. By applying the DS evidence theory information fusion algorithm, data from different departments such as meteorology, municipal administration, and transportation (which may be incomplete or even conflicting) are effectively fused to generate a comprehensive "resource demand heatmap," handling the fusion of "uncertain" information. With "chain disruption and disaster mitigation" as its fundamental goal, it directs resources to prioritize high-vulnerability areas marked on the heatmap that are most likely to trigger chain reactions. The decision objective is clear (chain disruption) and the method is scientific (handling uncertain information), enabling the resource scheduling scheme to not only meet current needs but also focus on preventing the escalation of disasters and achieving strategic disaster mitigation.

[0011] Furthermore, the feedback control described in step S5 includes: covering multiple channels such as mobile phone text messages and social media through the emergency early warning information dissemination platform; and recording the entire dispatch process based on blockchain evidence for auditing and traceability. Dispatch objects include fire-fighting robots, explosion-proof fire-fighting reconnaissance and extinguishing robots, and reconnaissance and delivery drones. A reliable auditing and tracing system is established using the distributed ledger and immutability of blockchain. Multi-channel dissemination utilizes the redundancy of information dissemination to ensure reliability. Dispatch instructions are issued through multiple early warning platforms such as mobile phone text messages and social media to ensure information reach. Simultaneously, blockchain technology is used to immutably record key operations such as dispatch instructions and execution feedback. This enhances the authority of emergency command and the traceability of instructions, facilitates post-event review and responsibility determination, and improves the standardization and credibility of the entire emergency management system. The dispatch system is integrated with specific intelligent and unmanned emergency equipment to achieve automated and efficient linkage from decision-making to execution. This improves the efficiency and safety of rescue operations, especially in high-risk disaster sites where personnel cannot easily access them; unmanned equipment can replace human labor to perform tasks, reducing the risk of casualties.

[0012] A dynamic emergency resource scheduling system based on multidimensional data analysis is used to implement the above-mentioned method, comprising: a data acquisition module for real-time acquisition of multidimensional data, including meteorological sensors and infrastructure monitoring equipment; a disaster chain analysis module for calculating node vulnerability based on a complex network model; a scheduling decision module for executing a multi-objective optimization algorithm to generate resource allocation schemes; and a feedback control module for connecting to an emergency command platform through a communication interface to realize command issuance and dynamic adjustment.

[0013] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method.

[0014] The beneficial effects of this invention are as follows: By effectively integrating multi-dimensional dynamic information such as real-time meteorology, infrastructure status, and social disaster-bearing entities, the data dimensions are multi-dimensional, fundamentally transforming emergency resource scheduling from passive response to proactive prediction and intervention. It accurately locates key links in the disaster chain for resource allocation, proactively simulates and assesses the dynamic evolution of the disaster chain, predicts secondary and derivative disasters in advance, and advances resource scheduling. It fully considers multi-party collaboration and privacy protection, improves rescue efficiency through cross-departmental data sharing, and accurately matches resource allocation schemes with key nodes and vulnerable paths in the disaster chain, achieving the goal of breaking the chain and mitigating disasters. It improves scheduling efficiency and resource utilization, effectively controls the spread of disasters, and reduces overall losses. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method steps of the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention; Figure 3 This is a schematic diagram of the event tree analysis process of the present invention. Detailed Implementation

[0016] The present invention will now be described in further detail with reference to the accompanying drawings. It should be noted that this is only for the purpose of more clearly illustrating and explaining the present invention. Example 1

[0017] like Figure 1-3 As shown, this embodiment discloses a dynamic emergency resource scheduling method based on multidimensional data analysis, including the following steps: S1, multidimensional data acquisition and fusion: real-time acquisition of extreme weather disaster-causing factor data, urban critical infrastructure status data, social disaster-bearing body data, and historical disaster case data to construct a multi-source heterogeneous database; S2, disaster chain network construction: based on the data in step S1, a disaster chain evolution network model is constructed using complex network theory, calculating node degree, betweenness centrality, and vulnerability, and identifying key risk nodes and evolution paths; S3, dynamic risk assessment: integrating real-time monitoring data and the disaster chain model, using Bayesian networks or dynamic trust assessment algorithms to quantify the current risk level and predict future risk trends; S4, emergency resource optimization scheduling decision-making: based on the risk assessment results in step S3, combined with resource inventory, geographic information, and scheduling constraints, a multi-objective optimization algorithm is used to generate a resource allocation scheme, prioritizing the blocking of high-risk chain propagation paths; S5, scheduling implementation feedback control: issuing scheduling instructions through the emergency command platform, and dynamically adjusting resource deployment based on real-time feedback data to form a closed loop of assessment, decision-making, execution, and feedback.

[0018] By collecting and integrating multi-dimensional data from meteorology, infrastructure, and society in real time into a unified database, a disaster chain model is constructed using complex network theory to identify key risk points and evolution paths. Then, dynamic risk assessment and prediction are performed by combining real-time data with the disaster chain model. Based on the risk assessment results, resource reserves and geographical constraints are comprehensively considered, and an optimal resource scheduling plan is generated through a multi-objective optimization algorithm. Finally, instructions are issued through an emergency platform, and adjustments are made dynamically based on on-site feedback, forming a closed-loop management system. Its core principles are "data-driven decision-making" and "predictive intervention." By connecting isolated disaster information into a dynamically evolving "disaster chain network," the system can anticipate disaster development rather than simply responding to disasters that have already occurred. The multi-objective optimization algorithm seeks the optimal solution to block the most critical disaster path under multiple constraints (such as time, cost, and resource availability), realizing a shift from "experience-based scheduling" to "model-optimized scheduling."

[0019] By effectively integrating multi-dimensional dynamic information such as real-time meteorology, infrastructure status, and social disaster-bearing entities, the data dimensions are multi-dimensional, enabling a fundamental shift in emergency resource scheduling from passive response to proactive prediction and intervention. This allows for precise allocation of resources to key links in the disaster chain, proactive dynamic simulation and risk assessment of the disaster chain's evolution, early prediction of secondary and derivative disasters, and advance resource scheduling. It also fully considers multi-party collaboration and privacy protection, enhances rescue efficiency through cross-departmental data sharing, and precisely matches resource allocation plans with key nodes and vulnerable paths in the disaster chain. This achieves the goal of breaking the chain and mitigating disaster, improving scheduling efficiency and resource utilization, effectively controlling the spread of disasters, and reducing overall losses. Example 2

[0020] like Figure 1-3 As shown, this embodiment discloses a dynamic emergency resource scheduling method based on multidimensional data analysis, including the following steps: S1, multidimensional data acquisition and fusion: real-time acquisition of extreme weather disaster-causing factor data, urban critical infrastructure status data, social disaster-bearing body data, and historical disaster case data to construct a multi-source heterogeneous database; S2, disaster chain network construction: based on the data in step S1, a disaster chain evolution network model is constructed using complex network theory, calculating node degree, betweenness centrality, and vulnerability, and identifying key risk nodes and evolution paths; S3, dynamic risk assessment: integrating real-time monitoring data and the disaster chain model, using Bayesian networks or dynamic trust assessment algorithms to quantify the current risk level and predict future risk trends; S4, emergency resource optimization scheduling decision-making: based on the risk assessment results in step S3, combined with resource inventory, geographic information, and scheduling constraints, a multi-objective optimization algorithm is used to generate a resource allocation scheme, prioritizing the blocking of high-risk chain propagation paths; S5, scheduling implementation feedback control: issuing scheduling instructions through the emergency command platform, and dynamically adjusting resource deployment based on real-time feedback data to form a closed loop of assessment, decision-making, execution, and feedback.

[0021] By collecting and integrating multi-dimensional data from meteorology, infrastructure, and society in real time into a unified database, a disaster chain model is constructed using complex network theory to identify key risk points and evolution paths. Then, dynamic risk assessment and prediction are performed by combining real-time data with the disaster chain model. Based on the risk assessment results, resource reserves and geographical constraints are comprehensively considered, and an optimal resource scheduling plan is generated through a multi-objective optimization algorithm. Finally, instructions are issued through an emergency platform, and adjustments are made dynamically based on on-site feedback, forming a closed-loop management system. Its core principles are "data-driven decision-making" and "predictive intervention." By connecting isolated disaster information into a dynamically evolving "disaster chain network," the system can anticipate disaster development rather than simply responding to disasters that have already occurred. The multi-objective optimization algorithm seeks the optimal solution to block the most critical disaster path under multiple constraints (such as time, cost, and resource availability), realizing a shift from "experience-based scheduling" to "model-optimized scheduling."

[0022] By effectively integrating multi-dimensional dynamic information such as real-time meteorology, infrastructure status, and social disaster-bearing entities, the data dimensions are multi-dimensional, enabling a fundamental shift in emergency resource scheduling from passive response to proactive prediction and intervention. This allows for precise allocation of resources to key links in the disaster chain, proactive dynamic simulation and risk assessment of the disaster chain's evolution, early prediction of secondary and derivative disasters, and advance resource scheduling. It also fully considers multi-party collaboration and privacy protection, enhances rescue efficiency through cross-departmental data sharing, and precisely matches resource allocation plans with key nodes and vulnerable paths in the disaster chain. This achieves the goal of breaking the chain and mitigating disaster, improving scheduling efficiency and resource utilization, effectively controlling the spread of disasters, and reducing overall losses.

[0023] For better results, the multidimensional data collected and fused in step S1 includes meteorological data, infrastructure data, and social data. The meteorological data includes the intensity and spatiotemporal distribution of extreme rainfall, high temperatures, low temperatures, snowfall, and cold waves. The infrastructure data includes the probability of water supply network rupture, the corrosion status of gas supply facilities, and power outage records. The social data includes population density, traffic flow, and the distribution of medical resources. During implementation, in the data collection phase, meteorological data (e.g., rainfall intensity), infrastructure data (e.g., probability of network rupture), and social data (e.g., population density) are explicitly collected. This adopts the disaster systems theory, which states that disasters are the result of the combined effects of disaster-causing factors (meteorology), disaster-bearing bodies (infrastructure), and the social environment. For example, the risk level of a rainstorm of the same intensity will be much higher in densely populated areas with aging pipelines than in other areas, leading to different dispatch decisions. Only by comprehensively covering data in all three dimensions can a complete risk assessment be conducted, avoiding decision-making biases caused by missing data dimensions.

[0024] To achieve better results, the disaster chain network construction in step S2 includes: extracting disaster event chains using event tree analysis, defining nodes as disaster events and edges as evolutionary relationships; visualizing the network topology using Gephi or Pajek software, and calculating edge betweenness numbers, average path length, and connectivity. The event tree analysis method is used to structurally extract disaster evolution sequences, and professional complex network analysis tools such as Gephi are used for visualization and quantitative calculation. Centrality indicators from complex network theory are used to identify key nodes and vulnerable edges in the network. For example, an evolutionary path with a high "edge betweenness number" indicates that it is a necessary path for multiple potential disaster chains; blocking it can effectively prevent the spread of disasters. This achieves objective and quantitative identification of key links in the disaster chain, replacing the traditional subjective judgment relying on expert experience, improving the scientific rigor and accuracy of risk analysis, providing a reliable basis for "using resources wisely," and enabling limited emergency resources to generate the greatest disaster reduction benefits.

[0025] For better results, the dynamic risk assessment in step S3 employs a dynamic trust assessment model based on trusted reports, integrating device status, environmental context, and behavioral logs. This "dynamic trust assessment model" comprehensively considers the device's operational status (reliability), its environment (context), and its historical behavioral logs to calculate its current credibility or risk value. Instead of treating the data from monitored devices as absolutely accurate, it treats it as an "object" requiring continuous credibility assessment, effectively addressing real-world issues such as device failures and data anomalies. Introducing the dynamic trust assessment model into the emergency data fusion process enhances the system's tolerance and resilience to data uncertainty and anomalies. It improves the robustness and reliability of risk assessment results; even if some monitored device data is abnormal, the system can identify and correct it through the model, avoiding erroneous scheduling decisions.

[0026] To achieve better results, the optimized scheduling decision in step S4 includes: applying DS evidence theory to fuse multi-department data and generate a resource demand heatmap; prioritizing resource allocation to the most vulnerable disaster chain edges with the goal of chain disruption and disaster mitigation. The optimized scheduling decision also includes combining privacy computing with federated learning or TEE to achieve cross-domain data, making cross-domain data usable but invisible, ensuring data security. By applying the DS evidence theory information fusion algorithm, data from different departments such as meteorology, municipal administration, and transportation (which may be incomplete or even conflicting) is effectively fused to generate a comprehensive "resource demand heatmap," handling the fusion of "uncertain" information. With "chain disruption and disaster mitigation" as its fundamental goal, it directs resources to prioritize high-vulnerability areas marked on the heatmap that are most likely to trigger chain reactions. The decision objective is clear (chain disruption) and the method is scientific (handling uncertain information), enabling the resource scheduling plan to not only meet current needs but also focus on preventing the escalation of disasters, achieving strategic disaster mitigation.

[0027] For better results, the feedback control described in step S5 includes: covering multiple channels such as mobile phone text messages and social media through the emergency early warning information dissemination platform; and recording the entire dispatch process based on blockchain evidence for auditing and traceability. Dispatch objects include fire-fighting robots, explosion-proof fire-fighting reconnaissance and extinguishing robots, and reconnaissance and delivery drones. A reliable audit and traceability system is established using the distributed ledger and immutability of blockchain. Multi-channel dissemination utilizes the redundancy of information dissemination to ensure reliability. Dispatch instructions are issued through multiple early warning platforms such as mobile phone text messages and social media to ensure information reach. Simultaneously, blockchain technology is used to immutably record key operations such as dispatch instructions and execution feedback. This enhances the authority of emergency command and the traceability of instructions, facilitates post-event review and responsibility determination, and improves the standardization and credibility of the entire emergency management system. The dispatch system is integrated with specific intelligent and unmanned emergency equipment to achieve automated and efficient linkage from decision-making to execution. This improves the efficiency and safety of rescue operations, especially in high-risk disaster sites where personnel cannot easily access them; unmanned equipment can replace human labor to perform tasks, reducing the risk of casualties.

[0028] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method.

[0029] A dynamic emergency resource scheduling system based on multidimensional data analysis is used to implement the above-mentioned method, comprising: a data acquisition module for real-time acquisition of multidimensional data, including meteorological sensors and infrastructure monitoring equipment; a disaster chain analysis module for calculating node vulnerability based on a complex network model; a scheduling decision module for executing a multi-objective optimization algorithm to generate resource allocation schemes; and a feedback control module for connecting to an emergency command platform through a communication interface to realize command issuance and dynamic adjustment.

[0030] Take a city experiencing extreme rainstorms as an example: In the S1 phase, the system collects rainfall radar data from the meteorological bureau, pipeline pressure data from the municipal department, and road condition data from the transportation department in real time.

[0031] In the S2 phase, a disaster chain model was constructed for "rainstorm, waterlogging, traffic disruption, flooding of critical facilities, and power / gas outages". The calculation revealed that the "flooded substation" node had the highest betweenness centrality and was the key risk point.

[0032] In the S3 phase, the dynamic trust assessment model integrates real-time rainfall intensity and water level data around the substation to predict that the station has an extremely high risk of flooding within the next 2 hours.

[0033] In the S4 phase, the scheduling decision module prioritizes allocating resources such as waterproof barriers and emergency power generation vehicles to the vicinity of the substation, and uses federated learning to generate the optimal transportation route without obtaining the original data from the transportation department.

[0034] In the S5 phase, instructions are issued to the on-site rescue team through the platform, drones transmit on-site images in real time, the system adjusts resource deployment points based on feedback, and the entire process is recorded through blockchain.

[0035] This invention uses a dynamic risk assessment model to predict risk situations, transforming passive response into proactive defense. By identifying key risk nodes through a disaster chain network, resource allocation becomes more targeted, effectively blocking the spread of disaster chains. The combination of multi-objective optimization algorithms and DS evidence theory improves decision-making speed and resource utilization efficiency, effectively enhancing the intelligence level of emergency resource scheduling and rescue efficiency.

[0036] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic emergency resource scheduling method based on multidimensional data analysis, characterized in that, Includes the following steps: S1. Multi-dimensional data collection and fusion: Real-time collection of extreme weather disaster-causing factor data, urban critical infrastructure status data, social disaster-bearing body data and historical disaster case data to construct a multi-source heterogeneous database; S2. Disaster chain network construction: Based on the data from step S1, a disaster chain evolution network model is constructed using complex network theory. The node degree, betweenness centrality, and vulnerability are calculated to identify key risk nodes and evolution paths. S3, Dynamic Risk Assessment, integrates real-time monitoring data with disaster chain models, and quantifies the current risk level and predicts future risk trends through Bayesian networks or dynamic trust assessment algorithms; S4. Make emergency resource optimization and scheduling decisions. Based on the risk assessment results of step S3, combined with resource inventory, geographic information and scheduling constraints, use a multi-objective optimization algorithm to generate a resource allocation plan, and prioritize blocking high-risk chain propagation paths. S5. Dispatch and implementation feedback control: Dispatch instructions are issued through the emergency command platform, and resource deployment is dynamically adjusted based on real-time feedback data, forming a closed loop of assessment, decision-making, execution and feedback.

2. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The multidimensional data collected and fused in step S1 includes meteorological data, infrastructure data, and social data; The meteorological data includes the intensity and spatiotemporal distribution data of extreme rainfall, high temperature, low temperature, snowfall, and cold waves; The infrastructure data includes the probability of water supply network rupture, the corrosion status of gas supply facilities, and power outage records; The social data includes population density, traffic flow, and distribution of medical resources.

3. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The disaster chain network construction described in step S2 includes: The event tree analysis method is used to extract the disaster event chain, defining nodes as disaster events and edges as evolutionary relationships; Visualize the network topology using Gephi or Pajek software, and calculate edge betweenness, average path length, and connectivity.

4. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The dynamic risk assessment in step S3 adopts a dynamic trust assessment model based on trusted reports, which integrates device status, environmental context and behavior logs.

5. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The optimization scheduling decision in step S4 includes: By applying the DS evidence theory and integrating data from multiple sectors, a resource demand heatmap is generated. With the goal of mitigating disasters by breaking supply chains, resources are prioritized for allocation to the most vulnerable edges of disaster chains.

6. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 5, characterized in that, Optimizing scheduling decisions also includes combining privacy-preserving computation with federated learning or TEE to enable cross-domain data, making cross-domain data available but not visible, thus ensuring data security.

7. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The feedback control described in step S5 includes: covering multiple channels such as mobile phone text messages and social media through the emergency early warning information dissemination platform; and recording the entire scheduling process based on blockchain evidence for auditing and traceability.

8. The dynamic emergency resource scheduling method based on multidimensional data analysis according to claim 1, characterized in that, The dispatching objects in step S5 include fire-fighting robots, explosion-proof fire-fighting reconnaissance and extinguishing robots, and reconnaissance and delivery drones.

9. A dynamic emergency resource scheduling system based on multidimensional data analysis, used to implement the method as described in any one of claims 1-8, characterized in that, include: Data acquisition module: used to collect multidimensional data in real time, including meteorological sensors and infrastructure monitoring equipment; Disaster Chain Analysis Module: Used to calculate node vulnerability based on complex network models; Scheduling decision module: Used to execute multi-objective optimization algorithms and generate resource allocation schemes; Feedback control module: Used to connect to the emergency command platform via a communication interface to enable command issuance and dynamic adjustment.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1-8.

Images