A method for dangerous goods transportation rescue scheduling based on traffic big data

Abstract

The present application provides a dangerous goods transportation rescue scheduling method based on traffic big data, including: obtaining information on disaster-prone areas and determining disaster distribution characteristics; calculating the risk index of each area by using the event tree analysis method, specifically including: constructing an event tree based on disaster statistics, quantitatively analyzing the event tree and calculating the risk index of each area; analyzing the video surveillance records of each road section and calculating the traffic flow and average vehicle speed at different time periods under different weather conditions; constructing a cellular automaton model to simulate the traffic flow of all road sections at different time periods under different weather conditions; parking rescue vehicles according to the weather conditions and providing an optimal scheduling plan, specifically including: parking rescue vehicles according to the weather conditions and providing an optimal scheduling plan for rescue vehicles when a disaster occurs.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and particularly to a method for dispatching dangerous goods transportation rescue based on traffic big data.

Background Art

[0002] Many vehicles, such as ambulances, fire trucks, police cars, road repair vehicles, etc., have important missions and tasks. Sometimes a delay can pose a threat to life and property safety. Road traffic does not provide special care for rescue, but still operates as usual. This can cause possible traffic jams for these vehicles. Especially those sections where these vehicles often appear and pass through are the necessary roads to ensure the safety of people's lives and property. Therefore, it is necessary to avoid these sections as much as possible during the process of dispatching rescue vehicles, select the best route, and ensure the smooth passage of rescue vehicles. And the location where rescue vehicles are parked will directly affect the efficiency of vehicle dispatching. Therefore, it is necessary to reasonably build parking areas to arrange the parking of rescue vehicles and shorten the dispatching time of rescue vehicles. How to avoid these frequently congested sections and not be delayed by traffic jams during the rescue opportunity, how to build parking areas, and adjust the parking positions of vehicles before a disaster occurs so as to better dispatch vehicles during a disaster is an unsolved problem.

Summary of the Invention

[0003] The present invention provides a method for dispatching dangerous goods transportation rescue based on traffic big data, mainly including:

[0004] Obtaining information on disaster-prone areas and determining the disaster distribution characteristics; calculating the risk index of each area by using the event tree analysis method; analyzing the video surveillance records of each section and calculating the traffic flow and average vehicle speed in each time period under different weather conditions; constructing a cellular automaton model to simulate the traffic flow of all sections in each time period under different weather conditions; determining the location selection for building a parking area according to the cellular automaton model; parking rescue vehicles according to the weather conditions and providing the best dispatching plan;

[0005] Further optionally, the obtaining information on disaster-prone areas and determining the disaster distribution characteristics includes:

[0006] Obtaining all disasters that have occurred and been recorded locally in history; obtaining the type of disaster, the location information where it occurred, the time when it occurred, the weather conditions when the disaster occurred, and the amount of economic loss; counting the number of times the same type of disaster has occurred at all locations where disasters have occurred; for each type of disaster, drawing a spatial distribution map of the disaster on the map according to the number of times the disaster has occurred corresponding to the location where the disaster occurred, and analyzing the distribution characteristics of the disaster.

[0007] Further optionally, the calculating the risk index of each area by using the event tree analysis method includes:

[0008] According to the disaster statistics, analyze the factors leading to disasters, construct an event tree; quantitatively analyze the event tree, calculate the probabilities of occurrence of various consequence events and the average economic loss amount; calculate the risk index of each region based on the probabilities of occurrence of the consequence events and the average economic loss amount; including: constructing an event tree according to the disaster statistics; quantitatively analyzing the event tree and calculating the risk index of each region;

[0009] The construction of the event tree according to the disaster statistics specifically includes:

[0010] According to the disaster statistics, analyze the factors leading to disasters, determine the initial event of the disaster; judge the safety functions related to the occurrence of the disaster; starting from the initial event, use branches to represent the event development paths, and draw the event tree according to the event development process; remove the safety functions in the branches that are irrelevant to the initial event or the disaster, have contradictory or inconsistent functional relationships, and simplify the event tree.

[0011] The quantitative analysis of the event tree and the calculation of the risk index of each region specifically include:

[0012] Statistically analyze the occurrence frequencies of events in each link of the event tree; statistically analyze the occurrence frequency of the initial event; approximate the occurrence frequency of the event as the probability, and calculate the probabilities of occurrence of subsequent events and various consequence events based on the occurrence frequency of the initial event and the events in each link of the event tree. Calculate the average economic loss amount of the consequence events according to the economic loss amount of the disasters that occur; calculate the risk index of each region based on the probabilities of occurrence of the consequence events and the average economic loss amount.

[0013] Further optionally, the analysis of the video surveillance videos of each road section and the calculation of the traffic flow and average vehicle speed in each time period under different weather conditions include:

[0014] Retrieve the video surveillance videos of all road sections in different time periods under different weather conditions; calculate the traffic flow and average vehicle speed of all road sections in different time periods under different weather conditions according to the video surveillance videos.

[0015] Further optionally, the construction of a cellular automaton model to simulate the traffic flow of all road sections in each time period under different weather conditions includes:

[0016] Obtain the lengths and numbers of lanes of all road sections; obtain the speed limits of all road sections and the durations of traffic lights; determine the driving rules of vehicles according to the traffic flow, average vehicle speed, lengths, numbers of lanes, speed limits and traffic light durations of road sections in different time periods under different weather conditions, and construct a cellular automaton model.

[0017] Further optionally, the determination of the location for building a parking area according to the cellular automaton model includes:

[0018] According to the distribution characteristics of historical disasters, investigate all the locations near the areas with dense disaster occurrences where parking areas can be built, and estimate the costs of building parking areas at each location; according to the cellular automaton model, combined with the time and weather conditions of historical disasters, calculate the average time taken to pass through each road section; according to the average time taken to pass through each road section, use the ant colony algorithm to solve the shortest time and the corresponding paths for rescue vehicles from each parking area construction point to all locations; according to the risk index of each region and the shortest time for rescue vehicles from each parking area construction point to all locations, combined with the construction costs of parking areas, use the genetic algorithm to solve the optimal location for building parking areas; including: calculating the average time taken to pass through all road sections according to the cellular automaton model; using the ant colony algorithm to solve the shortest time and the corresponding paths for dispatching rescue vehicles; using the genetic algorithm to solve the optimal location for building parking areas;

[0019] The calculating of the average time taken to pass through all road sections according to the cellular automaton model specifically includes:

[0020] Count the number of disasters occurring under various weather conditions and at various time periods; simulate the time taken to pass through all road sections under various weather conditions and at various time periods through the cellular automaton model; calculate the weighted average of the time taken to pass through all road sections with the proportion of the number of disasters occurring under various weather conditions and at various time periods as the weight.

[0021] The using of the ant colony algorithm to solve the shortest time and the corresponding paths for dispatching rescue vehicles specifically includes:

[0022] Mark the time taken to pass through each road section on the map; select a location as the end point for executing the ant colony algorithm; S1: Randomly place ants at different parking area construction points; S2: Each ant selects the next road section; S3: Judge whether there are unselected road sections. If so, execute S2, otherwise execute S4; S4: Update the pheromone concentration of each road section; Repeat the execution of S1 - S4 in sequence until the preset number of iterations is reached; Output the shortest time and the corresponding paths from all parking area construction points to the locations where disasters have occurred; Select the remaining locations as the end points and repeat the execution of the ant colony algorithm until all locations are selected, and output the shortest time and the corresponding paths from each parking area construction point to all locations.

[0023] The using of the genetic algorithm to solve the optimal location for building parking areas specifically includes:

[0024] Determine the objective function of the optimal site selection problem for building a parking area according to the construction cost and scheduling time; select a coding method and determine the length of the chromosome according to the solution accuracy; initialize the population according to the length of the chromosome; S1: Calculate the fitness of the population according to the objective function; S2: Select and replicate individuals with high fitness; S3: Perform crossover and mutation operations on the chromosomes; repeat S1 - S3 until the preset number of iterations is reached; read the result of the last iteration of the algorithm, decode the chromosome, and obtain the optimal distribution plan.

[0025] Further optionally, the parking of rescue vehicles according to weather conditions and providing the optimal scheduling plan includes:

[0026] Obtain the weather conditions of the local area for the next day, park rescue vehicles near the parking areas in the areas with dense disaster occurrences according to the disaster distribution that has occurred under the weather conditions of the next day; when a disaster occurs, according to the solution result of the ant colony algorithm, dispatch rescue vehicles along the optimal path obtained by the algorithm; including: parking rescue vehicles according to weather conditions; providing the optimal scheduling plan for rescue vehicles when a disaster occurs;

[0027] The parking of rescue vehicles according to weather conditions specifically includes:

[0028] Obtain the weather conditions of the next day through the local weather forecast; screen the disaster information that has occurred under the weather conditions of the next day; according to the spatial distribution map of disasters, park rescue vehicles in advance in the parking areas closest to the areas with dense disaster occurrences.

[0029] The providing of the optimal scheduling plan for rescue vehicles when a disaster occurs specifically includes:

[0030] Obtain the location, time, and weather conditions of the disaster occurrence; simulate the time taken to pass through all road sections using a cellular automaton model according to the time and weather conditions of the disaster occurrence; obtain the parking area of the rescue vehicle closest to the disaster occurrence location; use the ant colony algorithm to solve the optimal path from the parking area of the rescue vehicle closest to the disaster occurrence location to the disaster occurrence location, and dispatch rescue vehicles along the optimal path to implement the rescue.

[0031] The technical solutions provided in the embodiments of the present invention may include the following beneficial effects:

[0032] In the present invention, the disaster distribution characteristics and the risk index of each region can be determined based on historical disaster information, the traffic conditions of all road sections at different time periods under different weather conditions can be studied according to traffic big data, the optimal site selection for building a parking area can be determined through the disaster distribution characteristics, the risk index of each region, and the traffic conditions of the road sections, and the optimal scheduling plan for rescue vehicles can be provided. This enables rescue vehicles not to be delayed due to congestion, minimizing casualties and property losses to the greatest extent.

Description of the Drawings

[0033] Figure 1 This is a flowchart of a dangerous goods transportation rescue scheduling method based on traffic big data according to the present invention.

[0034] Figure 2 This is a schematic diagram of a dangerous goods transportation rescue scheduling method based on traffic big data according to the present invention.

Detailed implementation manners

[0035] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Figure 1 This is a flowchart of a dangerous goods transportation rescue scheduling method based on traffic big data according to the present invention. As Figure 1 shown, a dangerous goods transportation rescue scheduling method based on traffic big data in this embodiment may specifically include:

[0037] Step 101: Obtain information on disaster-prone areas and determine the disaster distribution characteristics.

[0038] Obtain all disasters that have occurred and been recorded locally in history; obtain the types of disasters, location information of the disasters, occurrence time, weather conditions during the disasters, and economic loss amounts; count the number of occurrences of the same type of disaster at all disaster-occurring locations; for each type of disaster, draw a spatial distribution map of the disasters on the map according to the number of disaster occurrences corresponding to the disaster-occurring locations, and analyze the disaster distribution characteristics; the types of disasters include: flood, fire, chemical disaster, epidemic infectious disease, and traffic accident; the size of the points in the spatial distribution map of the disasters reflects the number of disaster occurrences, and the larger the point in the distribution map, the more times the disaster has occurred at that location; the disaster distribution characteristics include the size, density, and shape of the points. There are a total of 14 weather conditions, namely overcast, sunny, sunny turning to cloudy, fog, light rain, moderate rain, heavy rain, rainstorm, thunderstorm, light snow, moderate snow, heavy snow, sleet, and hail.

[0039] Step 102: Calculate the risk index of each region by using the event tree analysis method.

[0040] According to the disaster statistics, analyze the factors leading to the disasters, construct an event tree; quantitatively analyze the event tree, calculate the probability of occurrence of each consequence event and the average economic loss amount; calculate the risk index of each region according to the probability of occurrence of the consequence event and the average economic loss amount.

[0041] Construct an event tree according to the disaster statistics.

[0042] According to the disaster statistics, analyze the factors leading to disasters, determine the initial events of disasters; judge the safety functions related to the occurrence of disasters; starting from the initial events, use branches to represent the event development paths, and draw an event tree according to the event development process; remove the safety functions in the branches that are irrelevant to the initial events or disasters, have conflicting or inconsistent functional relationships, and simplify the event tree. Event tree analysis is a method to systematically study how the initial events as hazard sources form a sequential logical relationship with subsequent events and ultimately lead to accidents. The initial event is a harmful event or dangerous event in the development process of a disaster before it occurs, such as machine failure, equipment damage, energy escape or out-of-control, human error, etc. The safety function is the function that eliminates or mitigates its impact when the initial event occurs to maintain the safe operation of the system. For example: an alarm system that reminds the operator of the occurrence of the initial event, measures taken by the operator according to the alarm or work procedures, and physical protection measures (such as dikes, fire dikes, and waste oil tanks) that limit the impact of the initial event. During the process of drawing the event tree, the state where the function is exerted (also known as the successful state) should be drawn on the upper branch, and the state where the function cannot be exerted (also known as the failure state) should be drawn on the lower branch until the system failure or disaster is reached.

[0043] Quantitatively analyze the event tree and calculate the risk index of each area.

[0044] Statistically analyze the occurrence frequency of each event link in the event tree; statistically analyze the occurrence frequency of the initial event; approximate the occurrence frequency of the event as the probability, and calculate the probability of subsequent events and each consequence event according to the occurrence frequency of the initial event and each event link in the event tree. According to the economic loss amount of the disaster, calculate the average economic loss amount of the consequence event; according to the probability of the consequence event occurring and the average economic loss amount, calculate the risk index of each area. The calculation formula of the risk index: the probability of the consequence event occurring × the average economic loss amount.

[0045] Step 103, analyze the video surveillance records of each road section and calculate the traffic flow and average vehicle speed in each time period under different weather conditions.

[0046] Retrieve the video surveillance records of all road sections in different time periods under different weather conditions; according to the video surveillance records, calculate the traffic flow and average vehicle speed of all road sections in each time period under different weather conditions; the time periods for retrieving the surveillance records are selected as the morning peak period, the morning stable period, the afternoon period, the evening period, and the night period. The traffic flow is the number of vehicles passing through a certain road section per unit time, and the calculation formula of the traffic flow: traffic flow = number of passing vehicles / time.

[0047] Step 104, construct a cellular automaton model to simulate the traffic flow of all road sections in each time period under different weather conditions.

[0048] Obtain the lengths and number of lanes of all road segments; obtain the speed limits of all road segments and the durations of traffic lights; determine the driving rules of vehicles based on the traffic flow, average vehicle speed, lengths, number of lanes, speed limits, and traffic light durations of road segments in different weather conditions at different time periods, and construct a cellular automaton model. A cellular automaton is a grid dynamics model with discrete time, space, and states, where spatial interactions and time causal relationships are local, and it has the ability to simulate the spatio-temporal evolution process of complex systems. Due to traffic rule restrictions, vehicle driving has a certain regularity. Therefore, a cellular automaton model can be constructed based on the driving rules of vehicles to simulate the traffic flow of road segments. Regarding each vehicle as an independent cell for simulation can better solve the randomness of the traffic flow. The driving rules of vehicles include: basic forward rule, lane-changing rule, acceleration rule, and deceleration rule. The basic forward rule is determined by the average speed of the vehicle; lane-changing rule: when there is a vehicle ahead, the driver will try to change lanes. If in the middle lane, the probabilities of changing left and right are equal. If in the edge lane, the driver will change to the middle lane; acceleration rule: when the road ahead is very clear, according to practical experience, the driver tends to accelerate, but the peak speed will not exceed the speed limit of the road segment; deceleration rule: when there is a vehicle ahead or the traffic light turns red, the driver will decelerate. The driver often also decelerates due to non-traffic factors, but the possibility of decelerating due to non-traffic factors is relatively small.

[0049] Step 105, determine the location for building a parking area according to the cellular automaton model.

[0050] According to the distribution characteristics of historical disasters, investigate all locations near areas with dense disaster occurrences where parking areas can be built, and estimate the costs of building parking areas at each location; according to the cellular automaton model, combined with the time and weather conditions of historical disaster occurrences, calculate the average time taken to pass through each road segment; according to the average time taken to pass through each road segment, use the ant colony algorithm to solve for the shortest time and the corresponding paths for rescue vehicles from each parking area construction point to all locations; according to the risk index of each region and the shortest time for rescue vehicles to reach all locations from each parking area construction point, combined with the construction cost of the parking area, use the genetic algorithm to solve for the optimal location for building a parking area.

[0051] Calculate the average time taken to pass through all road segments according to the cellular automaton model.

[0052] Count the number of disasters occurring under various weather conditions and at different time periods; simulate the time taken to pass through all road segments under various weather conditions and at different time periods using a cellular automaton model; calculate the weighted average of the time taken to pass through all road segments with the proportion of the number of disasters occurring under various weather conditions and at different time periods as the weight. For example, among all the disasters recorded in the local history, 300 cases occurred in the afternoon with foggy weather, 200 cases occurred during the morning rush hour with heavy rain, and 100 cases occurred in the evening with sunny weather; the time taken to pass through a certain road segment under foggy weather in the afternoon, heavy rain during the morning rush hour, and sunny weather in the evening is simulated by the cellular automaton to be 240s, 720s, and 600s respectively. Then the weighted average of the time taken to pass through this road segment is 240*(300 / 500)+720*(200 / 500)+600*(100 / 500) = 552s.

[0053] Use the ant colony algorithm to solve for the shortest time and the corresponding path for dispatching rescue vehicles.

[0054] Please refer to Figure 2 , mark the time taken to pass through each road segment on the map; select a location as the end point for implementing the ant colony algorithm; S1: Randomly place ants at different parking area construction points; S2: Each ant selects the next road segment; S3: Determine whether there are unselected road segments. If so, execute S2; otherwise, execute S4; S4: Update the pheromone concentration of each road segment; Repeat S1 - S4 in sequence until the preset number of iterations is reached; Output the shortest time and the corresponding path from all parking area construction points to the locations where disasters have occurred; Select the remaining locations as the end points and repeat the ant colony algorithm until all locations are selected, and output the shortest time and the corresponding path from each parking area construction point to all locations. The ant colony algorithm is inspired by the foraging behavior of ants. Since ants have no vision, when searching for a food source, they will release a pheromone on the path they pass through and can sense the pheromones released by other ants. The magnitude of the pheromone concentration represents the distance of the path. The higher the pheromone concentration, the shorter the corresponding path distance. The expression for pheromone update is:

[0055] τ ij (t + 1) = (1 - ρ)τ ij (t) + Δτ ij , 0 < ρ < 1

[0056] Among them,

[0057] i and j respectively represent the starting point and the ending point of each path segment, and τ represents the pheromone concentration from i to j at time t

[0058] Use genetic algorithm to solve the optimal location selection for building a parking area.

[0059] Determine the objective function for the problem of solving the optimal location selection for building a parking area based on the construction cost and scheduling time; select a coding method and determine the length of the chromosome according to the solution accuracy; initialize the population according to the length of the chromosome; S1: Calculate the fitness of the population according to the objective function; S2: Select and replicate individuals with high fitness; S3: Perform crossover and mutation operations on the chromosomes; repeat S1 - S3 until the preset number of iterations is reached; read the result of the last iteration of the algorithm, decode the chromosome, and obtain the optimal distribution plan. The essence of the genetic algorithm is to gradually evolve through population search technology according to the principle of survival of the fittest, and finally obtain the best or near-optimal distribution plan of the coupons. The optimal location selection plan for building a parking area should meet the following conditions: the total cost of building the parking area is as low as possible, and the total time for dispatching rescue vehicles to all locations with high risk indexes is as short as possible. Therefore, the objective function can refer to the following formula: - the total cost of building the parking area × ∑ (the dispatching time of the rescue vehicle × the risk index), where ∑ represents the summation over all locations. It is recommended to use the roulette wheel selection method for individual selection in S2; it is recommended to use single-point crossover for the chromosome crossover form, that is, only randomly set a crossover point in the individual coding string, and then exchange part of the chromosomes of two paired individuals at this point; it is recommended that the probabilities of chromosome crossover and mutation be 0.6 and 0.001 respectively, which can be adjusted according to the actual iteration effect of the algorithm; it is recommended to use binary coding for the chromosome coding method. The formula for calculating the length of the binary-coded chromosome:

[0060]

[0061] where d is the solution accuracy of the problem, and the domain is [U min , U max , then the number of bits of the code is m + 1

[0062] The formula for decoding the binary-coded chromosome:

[0063]

[0064] Step 106, park the rescue vehicles according to the weather conditions and provide the best scheduling plan.

[0065] Obtain the weather conditions of the local area for the next day, park the rescue vehicles near the parking areas in the areas where disasters occurred intensively according to the disaster distribution that occurred under the weather conditions of the next day; when a disaster occurs, dispatch the rescue vehicles according to the best path obtained by the ant colony algorithm according to the solution result of the ant colony algorithm.

[0066] Park the rescue vehicles according to the weather conditions.

[0067] Obtain the weather conditions for the next day through the local weather forecast; screen the disaster information that has occurred historically under the weather conditions of the next day; according to the spatial distribution map of disasters, park the rescue vehicles in advance in the parking area closest to the disaster-prone area. If the weather conditions of the next day are similar to those of the current day, there is no need to adjust the parking area of the rescue vehicles.

[0068] Provide the best dispatching plan for rescue vehicles during disasters.

[0069] Obtain the location, time, and weather conditions of the disaster; use the cellular automaton model to simulate the time taken to pass through all road sections according to the time and weather conditions of the disaster; obtain the parking area of the rescue vehicle closest to the disaster site; use the ant colony algorithm to solve the best path from the parking area of the rescue vehicle closest to the disaster site to the disaster site, and dispatch the rescue vehicle according to the best path to carry out the rescue.

[0070] The above only lists some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and transformations can be made. As long as the improvements and transformations are made on the basis of the basic principles of the present invention, they shall be regarded as falling within the protection scope of the present invention.

Claims

1. A method for dangerous goods transportation rescue scheduling based on traffic big data, characterized in that, The method includes: Obtaining information on disaster-prone areas and determining the characteristics of disaster distribution; Calculating the risk index of each area using the event tree analysis method, including: analyzing the factors leading to disasters based on disaster statistics, constructing an event tree; quantitatively analyzing the event tree, calculating the probability of occurrence of each consequence event and the average economic loss amount; calculating the risk index of each area based on the probability of occurrence of the consequence event and the average economic loss amount; Analyzing the video surveillance records of each road section and calculating the traffic flow and average vehicle speed in each time period under different weather conditions; Constructing a cellular automaton model to simulate the traffic flow of all road sections in each time period under different weather conditions; Determining the location for building a parking area based on the cellular automaton model, including: examining all locations where a parking area can be built near areas with dense disaster occurrences according to the distribution characteristics of historical disasters, estimating the cost of building a parking area at each location; counting the number of disasters occurring under various weather conditions and in each time period; simulating the time taken to pass through all road sections under various weather conditions and in each time period through the cellular automaton model; calculating the weighted average of the time taken to pass through all road sections with the proportion of the number of disasters occurring under various weather conditions and in each time period as the weight; solving the shortest time and the corresponding path for the rescue vehicle to reach all locations from each parking area construction point according to the average time taken to pass through each road section using the ant colony algorithm; solving the optimal location for building a parking area using the genetic algorithm based on the risk index of each area and the shortest time for the rescue vehicle to reach all locations from each parking area construction point, combined with the construction cost of the parking area; The solving of the optimal location for building a parking area using the genetic algorithm specifically includes: determining the objective function of the optimal location solving problem for building a parking area based on the construction cost and scheduling time; selecting a coding method and determining the length of the chromosome according to the solving accuracy; initializing the population according to the length of the chromosome; S1: calculating the fitness of the population according to the objective function; S2: selecting and replicating individuals with high fitness; S3: performing crossover and mutation operations on the chromosome; repeating S1 - S3 until the preset number of iterations is reached; reading the result of the last iteration of the algorithm, decoding the chromosome, and obtaining the optimal location plan for building a parking area; Parking rescue vehicles according to the weather conditions and providing the best scheduling plan.

2. The method according to claim 1, wherein, The obtaining of information on disaster-prone areas and determining the characteristics of disaster distribution includes: Obtaining all disasters that have occurred and been recorded locally in history; obtaining the type of disaster, location information of occurrence, time of occurrence, weather conditions during the disaster occurrence, and economic loss amount; counting the number of occurrences of the same type of disaster at all locations where disasters have occurred; For each type of disaster, drawing a spatial distribution map of the disaster on the map according to the number of disaster occurrences corresponding to the location of the disaster occurrence, and analyzing the distribution characteristics of the disaster.

3. The method according to claim 2, wherein, The analyzing of the factors leading to disasters based on disaster statistics and constructing an event tree specifically includes: According to the disaster statistics, analyze the factors leading to disasters, determine the initial events of disasters, and judge the safety functions related to the occurrence of disasters. Starting from the initial events, use branches to represent the event development paths and draw an event tree according to the event development process. Remove the safety functions in the branches that are irrelevant to the initial events or disasters, have contradictory or inconsistent functional relationships, and simplify the event tree. For the quantitative analysis of the event tree, calculate the probabilities of occurrence of each consequence event and the average economic loss amount. According to the probabilities of occurrence of the consequence events and the average economic loss amount, calculate the risk indices of each region, specifically including: Statistically analyze the frequencies of occurrence of each event in the event tree. Statistically analyze the frequency of occurrence of the initial event. Approximate the frequency of occurrence of the event as the probability, and calculate the probabilities of occurrence of subsequent events and each consequence event based on the frequency of occurrence of the initial event and each event in the event tree. Calculate the average economic loss amount of the consequence event according to the economic loss amount of the disaster. Calculate the risk indices of each region according to the probabilities of occurrence of the consequence events and the average economic loss amount.

4. The method according to claim 3, wherein, For the analysis of the video surveillance videos of each road section and the calculation of the traffic flow and average vehicle speed in each time period under different weather conditions, it includes: Retrieve the video surveillance videos of all road sections in different time periods under different weather conditions. According to the video surveillance videos, calculate the traffic flow and average vehicle speed of all road sections in different time periods under different weather conditions.

5. The method according to claim 4, wherein, For the construction of a cellular automaton model to simulate the traffic flow of all road sections in each time period under different weather conditions, it includes: Obtain the lengths and numbers of lanes of all road sections. Obtain the speed limits and signal durations of traffic lights of all road sections. Determine the driving rules of vehicles according to the traffic flow, average vehicle speed, lengths, numbers of lanes, speed limits, and signal durations of road sections in different time periods under different weather conditions, and construct a cellular automaton model.

6. The method according to claim 5, wherein, For the solution of the shortest time and corresponding paths of rescue vehicles from each parking area construction point to all locations using the ant colony algorithm according to the average time taken to pass through each road section, it specifically includes: Mark the time taken to pass through each road section on the map. Select a location as the end point for the execution of the ant colony algorithm. S1: Randomly place ants at different parking area construction points; S2: Each ant selects the next road section; S3: Determine whether there are unselected road sections. If so, execute S2, otherwise execute S4; S4: Update the pheromone concentration of each road section. Sequentially repeat the execution of S1 - S4 until the preset number of iterations is reached; Output the shortest time and corresponding paths from all parking area construction points to the locations where disasters have occurred. Select the remaining locations as the end points and repeat the execution of the ant colony algorithm until all locations are selected, and output the shortest time and corresponding paths from each parking area construction point to all locations.

7. The method according to claim 6, wherein, For parking rescue vehicles according to the weather conditions and providing the best dispatching plan, it includes: Obtain the weather conditions of the next day through the local weather forecast. Screen the disaster information that has occurred historically under the weather conditions of the next day. According to the spatial distribution map of disasters, park the rescue vehicles in advance in the parking areas closest to the disaster - prone areas. When a disaster occurs, obtain the location, time, and weather conditions of the disaster. The time taken to pass through all sections is simulated using a cellular automaton model according to the time of the disaster occurrence and weather conditions; the parking area of the rescue vehicle closest to the disaster occurrence location is obtained; the ant colony algorithm is used to solve the optimal path from the parking area of the rescue vehicle closest to the disaster occurrence location to the disaster occurrence location, and the rescue vehicle is dispatched according to the optimal path to implement the rescue.

Images