Earthquake post-disaster emergency rescue material delivery method

Abstract

This invention relates to the field of emergency logistics technology, specifically disclosing a method for the delivery of emergency relief supplies after an earthquake. The method includes data collection and preprocessing, post-earthquake road network capacity assessment, emergency facility site selection and material pre-positioning, assessment of the urgency of needs at disaster sites, multi-objective material delivery route planning, dynamic adjustment and replanning, and result visualization and decision support. This scheme extracts influencing factors from multiple levels, updates the posterior probability of each road segment using Bayesian network theory combined with real-time observation data, and achieves dynamic perception of road network capacity. Based on p-robust optimization theory, it constructs a collaborative site selection and allocation model for emergency facilities, providing feasible and efficient site selection and allocation schemes under various uncertain scenarios, significantly improving the resilience of the emergency collaborative response network. An optimization model is established, and a hybrid large neighborhood search-simulated annealing algorithm is designed. Through improved destruction operators and greedy insertion heuristics, the global search performance and solution efficiency of the algorithm are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of emergency logistics technology, specifically to a method for distributing emergency relief supplies after an earthquake. Background Technology

[0002] Earthquake disasters, characterized by their suddenness, destructiveness, and numerous secondary disasters, often cause massive casualties and property losses in a short period. Effective deployment of emergency relief supplies is crucial for reducing earthquake mortality rates and mitigating disaster losses. However, post-earthquake emergency relief supply deployment faces multiple challenges: First, earthquake damage to the road network structure leads to a sharp decline or even complete closure of some road sections, making traditional material distribution route planning methods based on static road networks ill-suited to the dynamically changing post-earthquake traffic environment. Second, the condition of the injured worsens over time, and the arrival time of relief supplies directly impacts the mortality rate. Furthermore, the severity of the disaster and the urgency of material needs vary significantly across different affected areas, and existing methods do not adequately consider fairness among affected areas. Finally, emergency relief involves multi-dimensional decision-making issues such as rapid site selection and pre-positioning of temporary emergency facilities and the coordinated allocation of various types of materials. Traditional single-layer optimization models struggle to simultaneously consider the comprehensive optimization of site selection, allocation, and routes. Summary of the Invention

[0003] To address the above issues and overcome the shortcomings of existing technologies, this invention provides a method for the deployment of emergency relief supplies after an earthquake. Addressing the problem that traditional post-earthquake road network capacity assessment relies on static data and lacks dynamic updates, leading to inaccurate route planning, this solution extracts influencing factors at multiple levels. It updates the posterior probability of each road segment using Bayesian network theory combined with real-time observation data, achieving dynamic perception of road network capacity and providing accurate foundational data for subsequent route planning. Addressing the issue of high uncertainty in the number of injured after an earthquake and the poor robustness of traditional deterministic models, this solution constructs a collaborative site selection and allocation model for emergency facilities based on p-robust optimization theory. This model provides feasible and efficient site selection and allocation schemes under various uncertain scenarios, significantly improving the resilience of the emergency collaborative response network. Addressing the problem that traditional rescue time minimization models fail to directly optimize the number of deaths, this solution establishes an optimization model with the objective of minimizing the cumulative number of deaths. It establishes a functional relationship between the arrival time of rescue vehicles and the cumulative mortality rate at the disaster site, and designs a hybrid large neighborhood search-simulated annealing algorithm. Through improved destruction operators and greedy insertion heuristics, the global search performance and solution efficiency of the algorithm are significantly improved.

[0004] The technical solution adopted in this invention is as follows: This invention provides a method for distributing emergency relief supplies after an earthquake, which includes the following steps:

[0005] Step S1: Data acquisition and preprocessing. Collect basic geographic information data, road network structure data, and historical earthquake disaster data of the earthquake-stricken area. Obtain real-time aftershock monitoring data. Clean and register the real-time monitoring data and normalize it into a unified format.

[0006] Step S2: Post-earthquake road network traffic capacity assessment. Based on Bayesian network, construct a dynamic assessment model of road network traffic capacity after earthquake, calculate the traffic probability of each road segment after earthquake, and update the traffic probability according to real-time monitoring data.

[0007] Step S3: Emergency facility site selection and material pre-positioning. Based on the p-robust optimization model, a collaborative site selection and allocation model for emergency facilities is constructed to determine the construction location of temporary emergency facilities and the material pre-positioning quantity of each facility.

[0008] Step S4: Assess the urgency of needs at disaster-stricken areas, construct a model for assessing the urgency of needs at disaster-stricken areas, and calculate the weight of material needs for each disaster-stricken area.

[0009] Step S5: Multi-objective material distribution route planning, constructing a multi-objective optimization model, and planning material distribution routes;

[0010] Step S6: Dynamic adjustment and replanning. Based on real-time monitoring data, dynamically adjust the material distribution route and trigger the replanning mechanism.

[0011] Step S7: Results visualization and decision support. Generate an earthquake emergency relief material deployment plan, presenting the optimal route, key road sections, and emergency facility layout in a visual format, and providing decision-making suggestions.

[0012] Furthermore, in step S2, the post-earthquake road network traffic capacity assessment specifically includes the following steps:

[0013] Step S21: Extraction of seismic damage influencing factors for each road segment. Extract influencing factors for each road segment and construct a seismic damage feature vector for each road segment.

[0014] Step S22: Bayesian network structure construction. Using road segments as parent nodes, network edges are constructed based on the connectivity between road segments. The connectivity of OD pairs (start-end pairs) is used as child nodes to establish a Bayesian network reflecting the road network topology. The formula used is as follows: ; ;

[0015] In the formula, Represents the road network topology. Represents the set of road segment nodes. This represents the set of connectivity relationships between road segments. Indicates road segment nodes, Indicates the total number of road segments;

[0016] Step S23: Calculation of the passability probability of the roadbed section. The average earthquake damage index is calculated based on the fuzzy comprehensive evaluation method, and then the passability probability of the roadbed section is calculated. The formula used is as follows: ; ;

[0017] In the formula, Indicates the road segment index. Indicates the first This section of road, Indicates road segment The average earthquake damage index. Indicates road segment The probability of passage of the roadbed portion. This represents the bridge seismic damage factor index, with values ​​ranging from 1 to 7. Indicates road segment No. Quantitative values ​​of individual bridge seismic damage factors;

[0018] Step S24: Calculate the probability of passage of the bridge section. Calculate the average seismic damage index of the bridge based on the empirical formula for bridge seismic damage. Determine the seismic damage level of the bridge by referring to a table based on the average seismic damage index, and then obtain the probability of passage of the bridge section. The formula used is as follows: ;

[0019] In the formula, Indicates road segment The average earthquake damage index of bridges. Indicates factor classification index, Indicates the first The first type of bridge seismic damage factor Statistical coefficients for classification A variable to indicate whether a bridge belongs to this category;

[0020] Step S25: Calculate the prior traffic probability of each road segment. Based on historical earthquake disaster data and the fuzzy comprehensive evaluation method, calculate the prior traffic probability of each road segment: For road segments without bridges, the prior traffic probability is the roadbed portion traffic probability; for road segments containing bridges, the traffic probability is the product of the roadbed portion traffic probability and the bridge portion traffic probability, using the following formula: ;

[0021] In the formula, This represents the prior probability of passage for a road segment. Indicates road segment The probability of passage for the bridge section;

[0022] Step S26: Acquisition of observation data and update of posterior probability. Real-time traffic status observation data of road segments are acquired through UAV reconnaissance, satellite remote sensing, and IoT sensors. The posterior probability of road segments is updated based on Bayes' theorem, using the following formula: ;

[0023] In the formula, This represents real-time traffic status observation data. This indicates the data observed in real-time traffic conditions. road section under the condition The posterior probability of passage. This represents the likelihood function, i.e., the road segment. Data observed during passage The probability, Represents the marginal probability of the observed data;

[0024] Step S27: Calculate the path passability probability. For any path composed of multiple road segments connected in series, its passability probability is the product of the posterior passability probabilities of each road segment. The formula used is as follows: ;

[0025] In the formula, Indicates a path. Representing a path The probability of passage, Representing a path The set of included road segments;

[0026] Step S28: Calculate the connectivity probability of OD pairs. For any OD pair from the start point to the end point, if there are multiple parallel paths, the connectivity probability of the OD pair is the probability that at least one path is passable. The formula used is as follows: ;

[0027] In the formula, This represents the connectivity probability of an OD pair. This represents the set of all paths contained in the OD pair.

[0028] Furthermore, in step S3, the selection of emergency facility locations and pre-positioning of supplies specifically includes the following steps:

[0029] Step S31: Classify emergency facilities into temporary emergency facilities and fixed emergency facilities. Temporary emergency facilities are material transfer stations that are quickly built after the earthquake, while fixed emergency facilities are existing material storage warehouses.

[0030] Step S32: Scenario Construction and Parameter Setting. Design multi-level earthquake disaster scenarios and the probability of occurrence for different scenarios. The formulas used are as follows: ; ;

[0031] In the formula, Represents a set of scenarios. Indicates the total number of scenarios. To describe a scenario, Indicates a scenario index. Indicates the first The probability of each scenario occurring;

[0032] Step S33: Construction of p-robust optimization model. With resilience and cost as objective functions, a collaborative site selection and allocation model for emergency facilities is constructed. The formula used is as follows: ;

[0033] In the formula, Represents the objective function value. This indicates the assembly point for temporary emergency candidate facilities. Indicates facility point index, Indicates at candidate point The cost of constructing temporary emergency facilities, Whether at the candidate point The 0-1 decision variables for constructing temporary emergency facilities Indicates the index of the disaster-affected point. Indicates a collection of disaster-stricken areas. Indicates the index of material types. Represents a set of material types. Indicates from facility point to the disaster site The rescue time Indicates the situation From the facility point To the disaster-stricken areas Delivery Quantity of this type of material;

[0034] Step S34: Set time reliability constraints to ensure that emergency rescue services can reach each disaster site within a specified time threshold. The formula used is as follows: ; ;

[0035] In the formula, Indicates the time threshold. Indicates that it can be within a time threshold Internal coverage of disaster-stricken areas The collection of candidate facilities Indicates the maximum rescue time threshold;

[0036] Step S35: Set material reliability constraints to ensure that the material support at each disaster site reaches the prescribed service level. The formula used is as follows: ;

[0037] In the formula, This represents the coefficient of service level for material support. Describing a scenario Lower disaster point Demand weight, Describing a scenario Lower disaster point right The basic demand for this type of material;

[0038] Step S36: Set facility capacity constraints to ensure that the material storage capacity of each temporary emergency facility does not exceed its maximum capacity. The formula used is as follows: ;

[0039] In the formula, Units The volume of such materials Indication facilities right Maximum storage capacity for this type of material;

[0040] Step S37: Material Pre-positioning Quantity Decision. Based on the solution results of the optimization model, determine the material pre-positioning quantity for each temporary emergency facility. The formula used is as follows: ;

[0041] In the formula, Indicates in facilities Pre-set Quantity of this type of material.

[0042] Furthermore, in step S5, the multi-target material distribution route planning specifically includes the following steps:

[0043] Step S51: Set the overall objective function, setting three optimization objectives: minimizing the cumulative number of deaths, minimizing the transit time of supplies, and minimizing the degree of unfairness in allocation. The formula used is as follows: ; ; ; ;

[0044] In the formula, Represents the overall objective function. Represents the function to be minimized. This indicates the cumulative number of deaths. Indicates the assembly time for the rescue operation. Indicates the unit time step; This indicates the total time the supplies were in transit. This indicates the assembly of rescue vehicles. Indicates vehicle The set of feasible paths Indicates vehicle Driving route Time required To indicate whether a path is selected 0-1 variables; Indicates the degree of unfairness in distribution. Indicates allocation to disaster-stricken areas The total amount of supplies, This indicates the urgency of the revised demand;

[0045] Step S52: Multi-objective normalization processing. A weighted summation method is used to transform multiple objectives into a single objective. The formula used is as follows: ; ;

[0046] In the formula, , and This represents the weighting coefficient of each objective. , and This represents the maximum possible value of each objective used for normalization.

[0047] Step S53: Construct a mixed-integer programming model, and construct a mixed-integer programming model with the goal of minimizing the comprehensive objective;

[0048] Step S54: Hybrid Large Neighborhood Search-Simulated Annealing Algorithm Design. A hybrid optimization algorithm is designed to solve the hybrid integer programming model. The specific steps are as follows:

[0049] Step S541: Initialization, generating initial solution Set the initial temperature Termination temperature Cooling rate Markov chain length ;

[0050] Step S542: Iterate through the outer loop to determine if the current temperature is higher than the termination temperature. If so, proceed to step S543; otherwise, go to step S548.

[0051] Step S543: Inner loop iteration, execute steps S544 to S546 until the number of inner loop iterations reaches the length of the Markov chain and the iteration stops;

[0052] Step S544: Destruction operation, removing several affected points from the current solution. An improved destruction operator is used to select the removal targets, and the formula used is as follows: ;

[0053] In the formula, Indicates the disaster-stricken area The probability of being selected. This represents the set of disaster-stricken points in the current path. Indicates the disaster-stricken area correlation function Indicates the randomness control parameter;

[0054] Step S545: Repair operation, using a greedy insertion heuristic to re-insert the removed disaster points into the path, selecting the insertion position that minimizes the increment of the objective function;

[0055] Step S546: Accept the criterion and calculate the objective function value of the new solution. and the objective function value of the current solution. A comparison is made; if the objective function value of the new solution is less than that of the current solution, the new solution is accepted; otherwise, the solution is rejected based on probability. Accept the new interpretation;

[0056] Step S547: Cooling operation, multiply the current temperature by a coefficient less than 1, and return to step S542;

[0057] Step S548: Output the optimal solution.

[0058] The beneficial effects achieved by the present invention using the above solution are as follows:

[0059] (1) In view of the problem that traditional post-earthquake road network capacity assessment relies on static data and is difficult to update dynamically, resulting in inaccurate route planning, this scheme extracts influencing factors from multiple levels and updates the posterior traffic probability of each road segment by combining Bayesian network theory with real-time observation data, so as to realize the dynamic perception of road network capacity and provide accurate basic data for subsequent route planning.

[0060] (2) In view of the problem that the number of injured people after the earthquake is highly uncertain and the robustness of traditional deterministic models is poor, this scheme constructs an emergency facility collaborative site selection-allocation model based on p-robust optimization theory, which provides feasible and efficient site selection and allocation schemes under various uncertain scenarios, and significantly improves the resilience of the emergency collaborative response network.

[0061] (3) In view of the problem that the traditional rescue time minimization model fails to directly optimize the number of deaths, this scheme establishes an optimization model with the goal of minimizing the cumulative number of deaths, establishes a functional relationship between the arrival time of rescue vehicles and the cumulative mortality rate of disaster sites, designs a hybrid large neighborhood search-simulated annealing algorithm, and significantly improves the global search performance and solution efficiency of the algorithm through improved destruction operators and greedy insertion heuristics. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating a method for distributing emergency relief supplies after an earthquake, as proposed in this invention.

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0064] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0065] Example 1, see Figure 1 The present invention provides a method for distributing emergency relief supplies after an earthquake, the method comprising the following steps:

[0066] Step S1: Data acquisition and preprocessing. Collect basic geographic information data, road network structure data, and historical earthquake disaster data of the earthquake-stricken area from the National Earthquake Science Data Center and the Geographic Information Public Service Platform. Obtain real-time aftershock monitoring data through the seismic network. Clean and register the real-time monitoring data and normalize it into a unified format.

[0067] Step S2: Post-earthquake road network traffic capacity assessment. Based on Bayesian network, construct a dynamic assessment model of road network traffic capacity after earthquake, calculate the traffic probability of each road segment after earthquake, and update the traffic probability according to real-time monitoring data.

[0068] Step S3: Emergency facility site selection and material pre-positioning. Based on the p-robust optimization model, a collaborative site selection and allocation model for emergency facilities is constructed to determine the construction location of temporary emergency facilities and the material pre-positioning quantity of each facility.

[0069] Step S4: Assess the urgency of needs at disaster-stricken areas, construct a model for assessing the urgency of needs at disaster-stricken areas, and calculate the weight of material needs for each disaster-stricken area.

[0070] Step S5: Multi-objective material distribution route planning, constructing a multi-objective optimization model, and planning material distribution routes;

[0071] Step S6: Dynamic adjustment and replanning. Based on real-time monitoring data, dynamically adjust the material distribution route and trigger the replanning mechanism.

[0072] Step S7: Results visualization and decision support. Generate an earthquake emergency relief material deployment plan, presenting the optimal route, key road sections, and emergency facility layout in a visual format, and providing decision-making suggestions.

[0073] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S2, the post-earthquake road network traffic capacity assessment specifically includes the following steps:

[0074] Step S21: Extraction of seismic damage influencing factors for road sections. For each road section, extract seven influencing factors: seismic intensity, site category, subgrade soil quality, liquefaction degree, bridge distribution, tunnel distribution, and slope stability, and construct a road section seismic damage feature vector.

[0075] Step S22: Bayesian network structure construction. Using road segments as parent nodes, network edges are constructed based on the connectivity between road segments, and child nodes are established based on the connectivity of OD pairs. The formula used is as follows: ; ;

[0076] In the formula, Represents the road network topology. Represents the set of road segment nodes. This represents the set of connectivity relationships between road segments. Indicates road segment nodes, Indicates the total number of road segments;

[0077] Step S23: Calculation of the passability probability of the roadbed section. The average earthquake damage index is calculated based on the fuzzy comprehensive evaluation method, and then the passability probability of the roadbed section is calculated. The formula used is as follows: ; ;

[0078] In the formula, Indicates the road segment index. Indicates the first This section of road, Indicates road segment The average earthquake damage index. Indicates road segment The probability of passage of the roadbed portion. This represents the bridge seismic damage factor index, with values ​​ranging from 1 to 7. Indicates road segment No. Quantitative values ​​of individual bridge seismic damage factors;

[0079] Step S24: Calculate the probability of passage of the bridge section. Calculate the average seismic damage index of the bridge based on the empirical formula for bridge seismic damage. Determine the seismic damage level of the bridge by referring to a table based on the average seismic damage index, and then obtain the probability of passage of the bridge section. The formula used is as follows: ;

[0080] In the formula, Indicates road segment The average earthquake damage index of bridges. Indicates factor classification index, Indicates the first The first type of bridge seismic damage factor Statistical coefficients for classification A variable to indicate whether a bridge belongs to this category;

[0081] Step S25: Calculate the prior traffic probability of each road segment. Based on historical earthquake disaster data and the fuzzy comprehensive evaluation method, calculate the prior traffic probability of each road segment: For road segments without bridges, the prior traffic probability is the roadbed portion traffic probability; for road segments containing bridges, the traffic probability is the product of the roadbed portion traffic probability and the bridge portion traffic probability, using the following formula: ;

[0082] In the formula, This represents the prior probability of passage for a road segment. Indicates road segment The probability of passage for the bridge section;

[0083] Step S26: Acquisition of observation data and update of posterior probability. Real-time traffic status observation data of road segments are acquired through UAV reconnaissance, satellite remote sensing, and IoT sensors. The posterior probability of road segments is updated based on Bayes' theorem, using the following formula: ;

[0084] In the formula, This represents real-time traffic status observation data. This indicates the data observed in real-time traffic conditions. road section under the condition The posterior probability of passage. This represents the likelihood function, i.e., the road segment. Data observed during passage The probability, Represents the marginal probability of the observed data;

[0085] Step S27: Calculate the path passability probability. For any path composed of multiple road segments connected in series, its passability probability is the product of the posterior passability probabilities of each road segment. The formula used is as follows: ;

[0086] In the formula, Indicates a path. Representing a path The probability of passage, Representing a path The set of included road segments;

[0087] Step S28: Calculate the connectivity probability of OD pairs. For any OD pair from the start point to the end point, if there are multiple parallel paths, the connectivity probability of the OD pair is the probability that at least one path is passable. The formula used is as follows: ;

[0088] In the formula, This represents the connectivity probability of an OD pair. This represents the set of all paths contained in the OD pair.

[0089] By performing the aforementioned operations, this solution addresses the problem that traditional post-earthquake road network capacity assessment relies on static data and is difficult to update dynamically, leading to inaccurate route planning. It extracts influencing factors from multiple levels and updates the posterior probability of each road segment using Bayesian network theory combined with real-time observation data, thereby achieving dynamic perception of road network capacity and providing accurate basic data for subsequent route planning.

[0090] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S3, the selection of emergency facility sites and the pre-positioning of supplies specifically include the following steps:

[0091] Step S31: Classify emergency facilities into temporary emergency facilities and fixed emergency facilities. Temporary emergency facilities are material transfer stations that are quickly built after the earthquake, while fixed emergency facilities are existing material storage warehouses.

[0092] Step S32: Scenario Construction and Parameter Setting. Design multi-level earthquake disaster scenarios and the probability of occurrence for different scenarios. The formulas used are as follows: ; ;

[0093] In the formula, Represents a set of scenarios. Indicates the total number of scenarios. To describe a scenario, Indicates a scenario index. Indicates the first The probability of each scenario occurring;

[0094] Step S33: Construction of p-robust optimization model. With resilience and cost as objective functions, a collaborative site selection and allocation model for emergency facilities is constructed. The formula used is as follows: ;

[0095] In the formula, Represents the objective function value. This indicates the assembly point for temporary emergency candidate facilities. Indicates facility point index, Indicates at candidate point The cost of constructing temporary emergency facilities, Whether at the candidate point The 0-1 decision variables for constructing temporary emergency facilities Indicates the index of the disaster-affected point. Indicates a collection of disaster-stricken areas. Indicates the index of material types. Represents a set of material types. Indicates from facility point to the disaster site The rescue time Indicates the situation From the facility point To the disaster-stricken areas Delivery Quantity of this type of material;

[0096] Step S34: Set time reliability constraints to ensure that emergency rescue services can reach each disaster site within a specified time threshold. The formula used is as follows: ; ;

[0097] In the formula, Indicates the time threshold. Indicates that it can be within a time threshold Internal coverage of disaster-stricken areas The collection of candidate facilities Indicates the maximum rescue time threshold;

[0098] Step S35: Set material reliability constraints to ensure that the material support at each disaster site reaches the prescribed service level. The formula used is as follows: ;

[0099] In the formula, This represents the coefficient of service level for material support. Describing a scenario Lower disaster point Demand weight, Describing a scenario Lower disaster point right The basic demand for this type of material;

[0100] Step S36: Set facility capacity constraints to ensure that the material storage capacity of each temporary emergency facility does not exceed its maximum capacity. The formula used is as follows: ;

[0101] In the formula, Units The volume of such materials Indication facilities right Maximum storage capacity for this type of material;

[0102] Step S37: Material Pre-positioning Quantity Decision. Based on the solution results of the optimization model, determine the material pre-positioning quantity for each temporary emergency facility. The formula used is as follows: ;

[0103] In the formula, Indicates in facilities Pre-set Quantity of this type of material.

[0104] By performing the aforementioned operations, this solution addresses the issues of high uncertainty in the number of injured after an earthquake and poor robustness of traditional deterministic models. Based on p-robust optimization theory, it constructs a collaborative site selection and allocation model for emergency facilities, providing feasible and efficient site selection and allocation schemes under various uncertain scenarios, and significantly improving the resilience of the emergency collaborative response network.

[0105] Example 4, see Figure 1 This embodiment is based on the above embodiment. In step S4, the assessment of the urgency of the disaster-stricken area's needs specifically includes the following steps:

[0106] Step S41: Extract basic indicators, extract disaster severity indicators for each disaster-stricken point, including earthquake intensity, population density, number of injured, and degree of road interruption;

[0107] Step S42: Calculate the time sensitivity coefficient. The time sensitivity coefficient of each disaster-affected point is calculated using the following formula: ;

[0108] In the formula, Indicates time, Indicates the disaster-stricken area Time sensitivity coefficient, Indicates the disaster-stricken area The cumulative mortality rate is a function of the change over time;

[0109] Step S43: Construction of a dynamic evolution model for the number of wounded. Based on the pattern of worsening injuries over time, a dynamic evolution model for the number of wounded is constructed. The formula used is as follows: ;

[0110] In the formula, Indicates the disaster-stricken area At any moment The number of injured people awaiting rescue Indicates the initial number of wounded. Indicates the rate of deterioration of injury coefficient. Indicates time, Indicates the natural index;

[0111] Step S44: Comprehensive calculation of demand urgency. The urgency of demand at the disaster-stricken area is calculated by combining various indicators. The formula used is as follows: ; ;

[0112] In the formula, Indicates the disaster-stricken area The urgency of the demand , , , and These represent the weighting coefficients of each indicator. , , , , These represent the maximum values ​​of the indicators used for normalization.

[0113] Step S45: Fairness Perceived Difference Correction. The urgency of needs is adjusted based on the fairness perceived differences among disaster victims, using the following formula: ;

[0114] In the formula, This indicates the urgency of the revised demand. This represents the average level of urgency of needs across all disaster-affected areas. This represents the fairness perception moderating coefficient, when... A value greater than 0 indicates that disaster victims exhibit jealousy and demand more resources from those with above-average needs. A value below 0 indicates that disaster victims have a sense of compassion and are willing to relinquish resources to areas with below-average demand.

[0115] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S5, the multi-target material distribution route planning specifically includes the following steps:

[0116] Step S51: Set the overall objective function, setting three optimization objectives: minimizing the cumulative number of deaths, minimizing the transit time of supplies, and minimizing the degree of unfairness in allocation. The formula used is as follows: ; ; ; ;

[0117] In the formula, Represents the overall objective function. Represents the function to be minimized. This indicates the cumulative number of deaths. Indicates the assembly time for the rescue operation. Indicates the unit time step; This indicates the total time the supplies were in transit. This indicates the assembly of rescue vehicles. Indicates vehicle The set of feasible paths Indicates vehicle Driving route Time required To indicate whether a path is selected 0-1 variables; Indicates the degree of unfairness in distribution. Indicates allocation to disaster-stricken areas The total amount of supplies, This indicates the urgency of the revised demand;

[0118] Step S52: Multi-objective normalization processing. A weighted summation method is used to transform multiple objectives into a single objective. The formula used is as follows: ; ;

[0119] In the formula, , and This represents the weighting coefficient of each objective. , and This represents the maximum possible value of each objective used for normalization.

[0120] Step S53: Construct a mixed-integer programming model. Construct a mixed-integer programming model with the goal of minimizing the overall objective, including vehicle path constraints, material flow conservation constraints, time window constraints, and vehicle capacity constraints.

[0121] Step S54: Hybrid Large Neighborhood Search-Simulated Annealing Algorithm Design. A hybrid optimization algorithm is designed to solve the hybrid integer programming model. The specific steps are as follows:

[0122] Step S541: Initialization, generating initial solution Set the initial temperature Termination temperature Cooling rate Markov chain length ;

[0123] Step S542: Iterate through the outer loop to determine if the current temperature is higher than the termination temperature. If so, proceed to step S543; otherwise, go to step S548.

[0124] Step S543: Inner loop iteration, execute steps S544 to S546 until the number of inner loop iterations reaches the length of the Markov chain and the iteration stops;

[0125] Step S544: Destruction operation, removing several affected points from the current solution. An improved destruction operator is used to select the removal targets, and the formula used is as follows: ;

[0126] In the formula, Indicates the disaster-stricken area The probability of being selected. This represents the set of disaster-stricken points in the current path. Indicates the disaster-stricken area correlation function Indicates the randomness control parameter;

[0127] Step S545: Repair operation, using a greedy insertion heuristic to re-insert the removed disaster points into the path, selecting the insertion position that minimizes the increment of the objective function;

[0128] Step S546: Accept the criterion and calculate the objective function value of the new solution. and the objective function value of the current solution. A comparison is made; if the objective function value of the new solution is less than that of the current solution, the new solution is accepted; otherwise, the solution is rejected based on probability. Accept the new interpretation;

[0129] Step S547: Cooling operation, multiply the current temperature by a coefficient less than 1, and return to step S542;

[0130] Step S548: Output the optimal solution.

[0131] By performing the aforementioned operations, this solution addresses the problem that traditional rescue time minimization models fail to directly optimize the number of deaths. Instead, it establishes an optimization model aimed at minimizing the cumulative number of deaths, establishes a functional relationship between the arrival time of rescue vehicles and the cumulative mortality rate at disaster sites, and designs a hybrid large neighborhood search-simulated annealing algorithm. Through improved destruction operators and greedy insertion heuristics, the global search performance and solution efficiency of the algorithm are significantly improved.

[0132] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S6, the dynamic adjustment and replanning specifically includes the following steps:

[0133] Step S61: Real-time data acquisition, using drone reconnaissance, IoT sensors, and mobile phone signaling to collect real-time information on road condition changes and aftershocks;

[0134] Step S62: Trigger condition judgment, determine whether the replanning trigger conditions are met, including: changes in road traffic status, changes in road network structure caused by new aftershocks, new disaster-stricken areas, and significant changes in the demand for relief supplies;

[0135] Step S63: Dynamic path adjustment. For ongoing rescue missions, local path re-optimization is performed based on the current location and remaining tasks.

[0136] Step S64: Global replanning. When the triggering condition meets the preset threshold, global replanning is triggered, and the process returns to step S2 to re-execute the road network capacity assessment and route planning.

[0137] Example 7, see Figure 1 This embodiment is based on the above embodiment. In step S7, the result visualization and decision support specifically includes the following steps:

[0138] Step S71: Generate an emergency relief supplies deployment plan, including an emergency facility layout map, a supplies delivery route map, key road section markers, and an estimated arrival time.

[0139] Step S72: Mark critical and vulnerable road segments, calculate the sensitivity of each road segment based on a Bayesian network, and identify critical and vulnerable road segments in the road network. The formula used is as follows: ;

[0140] In the formula, Indicates road segment The sensitivity reflects the degree to which changes in the traffic probability of a road segment affect the connectivity probability of the OD.

[0141] Step S73: Provide decision-making recommendations, including suggestions for early reinforcement of key vulnerable road sections, suggestions for optimization of emergency facility layout, and alternative solutions for material delivery routes;

[0142] Step S74: Visual presentation, using GIS maps to display various decision-making information, supporting interactive queries and scheme adjustments.

[0143] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0144] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0145] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for delivering emergency relief supplies after an earthquake, characterized in that: The method includes the following steps: Step S1: Data acquisition and preprocessing. Collect basic geographic information data, road network structure data, and historical earthquake disaster data of the earthquake-stricken area. Obtain real-time aftershock monitoring data through the seismic network. Clean and register the real-time monitoring data and normalize it into a unified format. Step S2: Post-earthquake road network traffic capacity assessment. Based on Bayesian network, construct a dynamic assessment model of road network traffic capacity after earthquake, calculate the traffic probability of each road segment after earthquake, and update the traffic probability according to real-time monitoring data. Step S3: Emergency facility site selection and material pre-positioning. Based on the p-robust optimization model, a collaborative site selection and allocation model for emergency facilities is constructed to determine the construction location of temporary emergency facilities and the material pre-positioning quantity of each facility. Step S4: Assess the urgency of needs at disaster-stricken areas, construct a model for assessing the urgency of needs at disaster-stricken areas, and calculate the weight of material needs for each disaster-stricken area. Step S5: Multi-objective material distribution route planning, constructing a multi-objective optimization model, and planning material distribution routes; Step S6: Dynamic adjustment and replanning. Based on real-time monitoring data, dynamically adjust the material distribution route and trigger the replanning mechanism. Step S7: Results visualization and decision support. Generate an earthquake emergency relief material deployment plan, presenting the optimal route, key road sections, and emergency facility layout in a visual format, and providing decision-making suggestions.

2. The method for distributing emergency relief supplies after an earthquake according to claim 1, characterized in that: In step S2, the post-earthquake road network traffic capacity assessment specifically includes the following steps: Step S21: Extraction of seismic damage influencing factors for each road segment. Extract influencing factors for each road segment and construct a seismic damage feature vector for each road segment. Step S22: Bayesian network structure construction: with road segments as parent nodes, network edges are constructed based on the connectivity between road segments, and child nodes are constructed based on the connectivity of OD pairs, thus establishing a Bayesian network that reflects the road network topology. Step S23: Calculate the passability probability of the roadbed section. Calculate the average earthquake damage index based on the fuzzy comprehensive evaluation method, and calculate the passability probability of the roadbed section. Step S24: Calculate the probability of passage of the bridge section. Calculate the average seismic damage index of the bridge based on the empirical formula for bridge seismic damage. Determine the seismic damage level of the bridge by looking up the table based on the average seismic damage index of the bridge, and obtain the probability of passage of the bridge section. Step S25: Calculate the prior traffic probability of road segments. Based on historical earthquake disaster data and fuzzy comprehensive evaluation method, calculate the prior traffic probability of each road segment: For road segments without bridges, the prior traffic probability is the roadbed part traffic probability; for road segments with bridges, the traffic probability is the product of the roadbed part traffic probability and the bridge part traffic probability. Step S26: Acquisition of observation data and update of posterior probability. Acquire real-time traffic status observation data of the road segment and update the posterior probability of the road segment based on Bayes' theorem. Step S27: Calculate the path passability probability. For any path composed of multiple road segments connected in series, its passability probability is the product of the posterior passability probabilities of each road segment. Step S28: Calculate the connectivity probability of OD pairs. For any OD pair from the starting point to the ending point, if there are multiple parallel paths, the connectivity probability of the OD pair is the probability that at least one path is passable.

3. The method for distributing emergency relief supplies after an earthquake according to claim 1, characterized in that: In step S3, the selection of emergency facility locations and pre-positioning of supplies specifically includes the following steps: Step S31: Classify emergency facilities, dividing them into temporary emergency facilities and fixed emergency facilities; Step S32: Scenario construction and parameter setting, designing multi-level earthquake disaster scenarios and the probability of occurrence of different scenarios; Step S33: p-robust optimization model construction, using resilience and cost as objective functions, to construct an emergency facility collaborative site selection and allocation model; Step S34: Set time reliability constraints to ensure that emergency rescue services can reach each disaster site within the specified time threshold; Step S35: Set material reliability constraints to ensure that the material support at each disaster site reaches the prescribed service level; Step S36: Set facility capacity constraints to ensure that the material storage capacity of each temporary emergency facility does not exceed its maximum capacity; Step S37: Material pre-positioning decision: Based on the solution results of the optimization model, determine the material pre-positioning quantity for each temporary emergency facility.

4. The method for distributing emergency relief supplies after an earthquake according to claim 1, characterized in that: In step S5, the multi-target material delivery route planning specifically includes the following steps: Step S51: Set the overall objective function, setting three optimization objectives: minimize the cumulative number of deaths, minimize the transit time of supplies, and minimize the degree of unfairness in allocation; Step S52: Multi-objective normalization processing, using a weighted summation method to transform multiple objectives into a single objective; Step S53: Construct a mixed-integer programming model, and construct a mixed-integer programming model with the goal of minimizing the comprehensive objective; Step S54: Hybrid Large Neighborhood Search-Simulated Annealing Algorithm Design. A hybrid optimization algorithm is designed to solve the hybrid integer programming model. The specific steps are as follows: Step S541: Initialization, generate initial solution, set initial temperature, termination temperature, and cooling rate. Markov chain length; Step S542: Iterate through the outer loop to determine if the current temperature is higher than the termination temperature. If so, proceed to step S543; otherwise, go to step S548. Step S543: Inner loop iteration, execute steps S544 to S546 until the number of inner loop iterations reaches the length of the Markov chain and the iteration stops; Step S544: Destruction operation, remove several affected points from the current solution, and use an improved destruction operator to select the removal objects; Step S545: Repair operation, using a greedy insertion heuristic to re-insert the removed disaster points into the path, selecting the insertion position that minimizes the increment of the objective function; Step S546: Acceptance criterion: Calculate the objective function value of the new solution and compare it with the current solution. If the objective function value of the new solution is less than the objective function value of the current solution, then accept the new solution; otherwise, accept the new solution with a specific probability. Step S547: Cooling operation, multiply the current temperature by a coefficient less than 1, and return to step S542; Step S548: Output the optimal solution.

Images