High-speed emergency rescue facility site selection and resource scheduling method, system and program

Abstract

The invention relates to a highway emergency rescue facility site selection and resource scheduling method, system and program. The method comprises the following steps: establishing a network diagram containing emergency demand points and emergency facility points to be selected based on highway accident historical data; dividing traffic accidents in the historical data into a plurality of situations, and calculating emergency demands of different emergency demand points under the traffic accidents in different situations; based on the emergency demand of each emergency demand point, by taking the maximization of the emergency resource supply capability and the minimization of the rescue travel time as targets and combining the rescue deadline constraint and the facility point number constraint, establishing an emergency facility point site selection model and solving, and selecting an emergency facility point; and on the basis of the selected emergency facility points, establishing an emergency resource scheduling model and solving the emergency resource scheduling model by taking the shortest emergency rescue time and the lowest rescue cost as targets for the place where the accident occurs, and obtaining scheduling schemes of each emergency demand point under different accident conditions. According to the invention, emergency rescue facilities can be reasonably arranged on the expressway, and resources can be effectively scheduled when an accident occurs.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation informatization, and specifically relates to a method, system and program for site selection and resource scheduling of high-speed emergency rescue facilities. Background Art

[0002] After a traffic accident occurs, if timely and effective rescue measures cannot be provided, secondary accidents may occur, resulting in more serious casualties and economic losses. If traffic accident rescue points can be reasonably set up in the highway network, it can not only reduce the cost of emergency management, but also ensure the timeliness and effectiveness of rescue, and avoid or reduce casualties and property losses caused by untimely rescue. Highway traffic accidents have the characteristics of uncertainty, suddenness and time-varying, and emergency rescue work is more difficult to carry out. In order to further improve the efficiency of emergency resource utilization and the level of emergency services, it is of great significance to study the optimal layout of highway emergency rescue center site selection and emergency resource scheduling.

[0003] At present, emergency plans for traffic accidents often adopt the method of manually determining plans, manually allocating resources, and manually dispatching resources. This decision-making that relies entirely on manual experience has certain differences and misjudgments, resulting in low rescue efficiency. Summary of the invention

[0004] The first purpose of the present invention is to solve the above-mentioned highway traffic accident emergency rescue problem, including the emergency rescue facility site selection and emergency resource scheduling problems, and to provide a highway emergency rescue facility site selection and resource scheduling method.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] A method for site selection and resource scheduling of high-speed emergency rescue facilities, comprising:

[0007] Obtain historical data on highway accidents, count sections with high accident rates as emergency demand points, and determine candidate emergency facility points based on the highway network. Establish a network diagram with emergency demand points and candidate emergency facility points as points, and roads connecting emergency demand points and between emergency demand points and candidate emergency facility points as arcs.

[0008] Traffic accidents in historical data are divided into several situations, and the emergency demands under different situations of traffic accidents at different emergency demand points are calculated; for the emergency facility points to be selected in the network diagram, based on the emergency demand of each emergency demand point, with the goal of maximizing the emergency resource supply capacity and minimizing the rescue travel time, combined with the rescue deadline constraint and the facility point quantity constraint, an emergency facility point site selection model is established and solved to select the emergency facility point;

[0009] Based on the selected emergency facility points, an emergency resource scheduling model is established and solved for the location where the accident occurred, with the goal of minimizing emergency rescue time and minimizing rescue costs, to obtain the scheduling plans for each emergency demand point under different accident situations.

[0010] As a preferred implementation, the emergency demands at different emergency demand points under different traffic accident situations are calculated based on the following formula:

[0011]

[0012]

[0013] Where N i (k) is the demand point N i The emergency demand in case of traffic accident k; α ik is the demand point N i The probability of a traffic accident situation k occurring at I ik For the traffic accident situation k for demand point N i The influence coefficient of is the demand point N for each traffic accident situation k i The traffic capacity of C i is the demand point N i The design capacity of Q i is the demand point N i The average traffic flow of the road section; n ik is the demand point N i The number of traffic accidents in k situations; N is the demand point N i The total number of .

[0014] As a preferred implementation, the objective function of the accident emergency resource facility point screening model is set as follows:

[0015]

[0016]

[0017] Where N j (k) is the number of emergency facility points N to be selected when traffic accident k occurs j The amount of supplies that can be provided within the rescue period; N i (k) is the emergency demand point N i Emergency needs in the event of a traffic accident k; t ij For rescue vehicles from the candidate emergency facility point N j To emergency demand point N i The travel time of the shortest path; y ij is a decision variable, when N i Points are Nj When the rescue range of the point is covered, take y ij =1, otherwise y ij =0, the rescue range is the preset value; x j is a decision variable. When N j When setting up a facility point, take x j =1, otherwise x j =0; I is the set of emergency demand point locations in the network diagram; J is the set of emergency facility points to be selected in the network diagram.

[0018] As a preferred implementation, when the emergency facility site selection model is established, the shortest time required to complete the emergency rescue work is calculated first, and the shortest time is used as a constraint condition, and the minimization of rescue costs is used as the objective function to obtain an emergency rescue material scheduling plan.

[0019] As a preferred implementation method, the shortest time required for the emergency rescue work is to obtain the method of sorting the emergency facilities according to the rescue time based on the time required for rescue at each emergency facility point that can be rescued around the accident point, and calculating the amount of materials at the rescue point. The shortest time is the shortest emergency rescue time of one or more emergency facility points that meet the material needs of the accident point.

[0020] The shortest time required for the emergency rescue work is obtained by solving the following formula:

[0021]

[0022]

[0023] In the formula, Q t+1 , Q t are the remaining material demand at time t and time t+1 respectively; S t is the amount of materials involved in the rescue at time t; x i The amount of supplies provided to rescue point i; n is the total number of rescue points, and a is a constant;

[0024] when When the rescue demand is met, this moment is the shortest emergency rescue time; when When , rescue points are added and the solution is iterated until the rescue demand is met.

[0025] As a preferred implementation, the emergency resource scheduling model also includes a remaining material time conversion constraint:

[0026]

[0027] In the formula, Q0 is the initial amount of materials required for rescue; a is a constant; T is the shortest rescue time; t iis the time when the rescue point i arrives at the accident scene; i The amount of supplies provided to rescue point i.

[0028] As a preferred implementation, the accident emergency resource facility point screening model is solved based on a heuristic algorithm based on Lagrangian relaxation; and the emergency resource scheduling model is solved based on a greedy algorithm.

[0029] Another object of the present invention is to provide a high-speed emergency rescue facility site selection and resource scheduling system, comprising:

[0030] The network diagram construction module obtains historical data on highway accidents, counts sections with high accident rates as emergency demand points, and determines candidate emergency facility points based on the highway network. The network diagram is established with emergency demand points and candidate emergency facility points as points, and roads connecting emergency demand points and between emergency demand points and candidate emergency facility points as arcs.

[0031] The emergency facility site selection module divides the traffic accidents in the historical data into several situations, and calculates the emergency demand under different emergency demand points and different situations of traffic accidents; for the candidate emergency facility points in the network diagram, based on the emergency demand of each emergency demand point, with the goal of maximizing the emergency resource supply capacity and minimizing the rescue travel time, combined with the rescue deadline constraint and the facility point quantity constraint, the emergency facility site selection model is established and solved to select the emergency facility points;

[0032] The emergency resource scheduling module, based on the selected emergency facility points, establishes and solves the emergency resource scheduling model for the location where the accident occurred with the goal of minimizing emergency rescue time and minimizing rescue costs, and obtains the scheduling plan for each emergency demand point under different accident situations.

[0033] A third object of the present invention is to provide a computer program product, comprising a computer program / instruction, which performs the steps of the above method when executed by a processor.

[0034] The present invention first analyzes the layout characteristics of emergency demand in the highway network, and determines the emergency demand of the demand point under a certain traffic accident situation according to the probability of accident occurrence, the scope of accident impact and the average traffic flow of the road section. Then, an emergency facility point site selection model is established based on the demand analysis. The coverage conditions of the facility point are determined by the rescue time limit, and the candidate points with short rescue time and strong material supply capacity are selected to establish the facility point. In the process of studying the highway emergency rescue system, the emergency resource scheduling method is the core of the entire highway traffic incident emergency rescue system. With the shortest emergency rescue end time and the minimum emergency rescue related costs as the goal, the influence of the highway traffic capacity is fully considered, and the mathematical programming method is used to establish an emergency resource scheduling model based on traffic capacity. The model is optimized and solved by the greedy algorithm, and then it is determined whether each emergency rescue point participates in the rescue and the cost capacity of each rescue point is compared, so as to select the appropriate emergency rescue point for effective rescue.

[0035] The present invention has the following beneficial effects:

[0036] (1) The present invention uses historical data of highway traffic accidents to determine the location of emergency demand points based on past accident experience (it can also be combined with future road conditions), and arranges emergency facility points according to the emergency needs of the highway network. It can select the site of emergency rescue facilities based on the high-incidence locations of highway accidents and accident needs, thereby ensuring the rescue time of emergency events, providing a new idea for the layout of highway emergency facilities, and has strong applicability.

[0037] (2) The emergency resource scheduling of the present invention aims to minimize the emergency rescue time and the rescue cost, and comprehensively considers the dynamic changes in the traffic capacity of the highway and the demand for emergency rescue materials at the accident site, selects appropriate emergency rescue points for effective rescue, and provides an emergency material scheduling plan with the lowest rescue cost under the rescue time limit.

[0038] (3) The present invention takes into account different traffic accident scenarios (accident vehicles, traffic volume, proportion of passenger cars and trucks, vehicle speed, impact range, etc.), and can generate active control measures and resource scheduling plan libraries based on different accident scenarios. It plays a key auxiliary role in the accident emergency rescue decision-making of relevant transportation departments. DETAILED DESCRIPTION

[0039] The technical solution of the present invention is further described below in conjunction with specific embodiments.

[0040] Example 1

[0041] This embodiment generates emergency rescue facility layout and emergency resource scheduling plans based on historical data of highway traffic accidents and current accidents and road characteristics, and simulates and evaluates alternative plans in the vehicle traffic pattern simulation model to obtain the optimal plan. This embodiment includes four major steps: A1, road network emergency demand layout and emergency demand analysis; A2, emergency facility site selection model construction and solution; A3, emergency resource scheduling model construction and solution; A4, intelligent rescue plan generation based on vehicle traffic pattern simulation deduction, as follows:

[0042] A1 Road network emergency demand layout and emergency demand analysis

[0043] The objectives of this step are: 1) to establish a highway network diagram to intuitively reflect the layout of traffic accident emergency demand points in the road network; 2) then conduct demand analysis on the emergency demand points to determine the emergency demand situation of the demand points under specific accidents.

[0044] A1.1 Create a network diagram

[0045] The location of the traffic accident emergency demand point in the highway network can be determined based on empirical values, which are the locations where traffic accidents have occurred in recent years. The location of the emergency demand point is determined by counting the high-incidence sections of highways in the selected area. Using a network diagram to describe the highway network can intuitively reflect the location of the traffic accident emergency demand point and the connection between the roads at each point, which is convenient for studying the layout of emergency resources in the highway network. i Indicates emergency demand points; N j = represents the selected accident emergency resource facility point (hereinafter referred to as the candidate emergency facility point); arc e represents the road connection between emergency demand points and the candidate emergency facility point to the emergency demand point. The entire network graph is denoted as G=(N, E), where N represents the set of points and E represents the set of arcs. In this embodiment, toll stations and emergency rescue ports on highways are used as candidate accident emergency resource facility points.

[0046] A1.2 Emergency Demand Analysis

[0047] The emergency resources required for traffic accident emergency demand points are uncertain, which depends on many factors such as the situation of the traffic accident, the traffic conditions of the road where the accident point is located, and the impact of the accident on vehicle traffic.

[0048] The analysis of expressway capacity is as follows:

[0049] The basic traffic capacity of a road section is also called the theoretical traffic capacity. It refers to the maximum number of vehicles that can pass through a lane or a section of a road in a unit time under ideal road, traffic, control and environmental conditions, regardless of the service level. It is the traffic load limit that the road section can bear. The one-way traffic capacity of a highway can be calculated according to the following formula:

[0050] (1)

[0051] Where: C is the traffic capacity, pcu / h; C b is the basic traffic capacity; p is the road familiarity correction coefficient; n is the number of one-way drivable lanes.

[0052] Therefore, the basic one-way capacity of the expressway :

[0053] (2)

[0054] Road familiarity correction factor f p It depends on the driver's age, driving experience and familiarity with the highway section, and is generally within the range of 0.90 to 1.00.

[0055] This embodiment assumes that K is a set of various traffic accident situations, k is a certain traffic accident situation, and k∈K. In order to describe the uncertain demand of traffic emergency demand points, this article introduces the following three parameters for possible traffic accident situations.

[0056] (1) α ik Represents the emergency demand point N i The probability of a traffic accident situation k occurring at the specified time is obtained based on historical data. i The number of traffic accidents in k situations n ik and emergency demand point N i The total number of traffic accidents N ik ,and:

[0057] (3)

[0058] (2) I ik Indicates the traffic accident situation k for the emergency demand point N i The impact coefficient of the emergency demand point N within the specified time is obtained based on historical data. i Number of traffic accidents in situation k at n ik , each time a traffic accident occurs, the emergency demand point N i Capacity 、Emergency demand point Ni Design capacity C i ,but:

[0059] (4)

[0060] N refers to the emergency demand point N i The total number of .

[0061] (3) Q i Represents the emergency demand point N i The average traffic flow of the road section. Therefore, the emergency demand point N i The emergency demand N in case of traffic accident k i (k) can be calculated as follows:

[0062] (5)

[0063] A2 Construction and solution of emergency facility location model

[0064] A2.1 Establishment of mathematical model

[0065] First, make the model assumptions

[0066] Assumption 1: After a traffic accident occurs, the travel time of the rescue vehicle from the emergency facility to the traffic accident site is known.

[0067] Assumption 2: After a traffic accident occurs, the shortest rescue route for emergency vehicles to the accident site does not change.

[0068] Assumption 3: The reserves of traffic accident emergency facilities are durable materials, such as medical kits, stretchers, ambulances, etc., which can be stored for a long time. In the embodiment, the number of rescue vehicles is selected to represent the amount of materials, which is convenient for calculation and comparison with t ij correspond.

[0069] Based on the above assumptions, the following mathematical model can be established:

[0070] (6)

[0071] (7)

[0072] (8)

[0073] (9)

[0074] (10)

[0075] (11)

[0076] (12)

[0077] Where: I is the set of traffic accident emergency demand point locations in a given road network (i.e., the network diagram established in A1.1); J is the set of accident emergency resource facility point locations to be selected in a given road network; N j (k) is the number of emergency facility points N to be selected when traffic accident k occurs j The number of rescue vehicles that can be provided within the rescue period; N i (k) is the emergency demand point N i The emergency demand when a traffic accident occurs is as shown in formula (5); t ij For rescue vehicles from the candidate emergency facility point N j To emergency demand point N i The travel time of the shortest path (min); y ij is a decision variable, when N i Points are N j When the rescue range of the point is covered, take y ij =1, otherwise y ij =0, the rescue range is a predefined parameter, which is determined comprehensively based on factors such as the rescue time limit and actual needs. In this embodiment, it is defined as a distance of 5 km; j is a decision variable. When N j When setting up a facility point, take x j =1, otherwise x j =0; P is the number of accident emergency resource facilities required for a given highway network; t max It is the limit period (min) for emergency rescue in traffic accidents.

[0078] Formula (6) is the objective function, which represents the optimal location for seeking a strong emergency resource supply capacity and a short rescue travel time.

[0079] Formulas (7)-(12) are constraints:

[0080] Formula (7) indicates that if the candidate emergency facility point N j The rescue range covers the emergency demand points N i , then it should be in N j Emergency facilities should be set up at designated locations.

[0081] Formula (8) limits the number of facility points that can be established.

[0082] Formula (9) represents each emergency demand point N i All need to be 1 emergency facility point N j covered by the rescue scope.

[0083] Formula (10) means that if N jThe rescue range of the point covers N i point, then from N j Point to N i The shortest path travel time of the point needs to be within the rescue limit.

[0084] Formula (11) and Formula (12) define x j and ij An integer variable with a value of 0 or 1.

[0085] A2.2 Model solution

[0086] The above model is a linear model and can be solved by the existing heuristic algorithm based on Lagrangian relaxation. Relax formulas (9) and (10) into the objective function and set the Lagrangian multipliers u i , i , and the following model is obtained:

[0087] (13)

[0088] The constraints are formulas (7), (8), (11), and (12).

[0089] The Lagrange multipliers in this model can be defined as:

[0090] (14)

[0091] in:

[0092] (15)

[0093] By making the decision variables The solution and slack variables The adjustment of slack variables can be achieved through an iterative process, and the iterative equation is as follows:

[0094] (16)

[0095] Where: M k and M k + 1 is the Lagrange multiplier for the kth and k+1th iterations, α k represents the step size of the kth iteration; g k represents the subgradient; Indicates that the slack variables should satisfy and , as shown in formula (17):

[0096] (17)

[0097] For a given u i and λ i By using the value of , we can find a set of feasible solutions to the problem L and use it as the upper limit UB of the optimal solution. (18)

[0098] All φ i By adding them together, we can get the lower limit of the optimal value of the location selection model, that is, LB. ij If the value is 1 and the rest are 0, we can get a corresponding feasible solution.

[0099] The algorithm steps of the model are as follows:

[0100] (1) Initialize the parameters, M 0 , k=1 The optimal lower limit LB of the site selection model is obtained by formula (17) 0 , by solving the given problem L, we can get the upper limit of the optimal value UB 0 ;

[0101] (2) Through iteration, we get M k , solve the problem L and get the optimal upper limit UB k ;

[0102] (3) For a given M k , find the feasible solution of the site selection model by formula (17) to calculate LB k ;

[0103] (4) Update the optimal upper and lower limits UB and LB:

[0104]

[0105] (5) When UB - LB ≤ 0.3 or , the calculation is completed and the result is output; otherwise, proceed to the next step;

[0106] (6) Update the Lagrange multiplier and iteration step, take k = k + 1, and go to step (2). During the calculation process, the iteration step is:

[0107] (19)

[0108] Among them, the parameter θ k The initial value is 2. When UB is calculated during the iteration k When UB cannot be updated, then:

[0109]

[0110] The Lagrange multipliers are updated as follows:

[0111] (20)

[0112] (twenty one)

[0113] A3 Construction and solution of emergency resource scheduling model

[0114] The core problem of the highway resource scheduling model is how to determine the time required for each rescue point within the rescue range covered by the accident point to reach the accident site and the amount of emergency supplies that the rescue point can provide. When a traffic accident occurs, the traffic command and dispatch center obtains the accident information and estimates the initial amount of rescue supplies required for the accident rescue. The initial amount of supplies will increase with the duration of the emergency event. Therefore, the demand for emergency rescue supplies at the accident point is dynamic. Based on the above description, an optimization plan for highway emergency resource scheduling is established with the goal of minimizing the emergency rescue time T and minimizing the rescue cost Z, and the plan is made to meet the actual situation of the rescue point and various constraints.

[0115] A3.1 Building an emergency resource scheduling model

[0116] Since the demand for emergency rescue materials is changing dynamically and the construction area is relatively complex, the following assumptions are proposed to facilitate the study of this issue:

[0117] The number of materials required at the accident point changes linearly over time, and its growth coefficient is a constant. Based on this assumption, the emergency resource scheduling optimization model is studied:

[0118] (twenty two)

[0119] (twenty three)

[0120] Where: x i is the rescue point i (i.e., the facility point N obtained by solving the model in A3) j ) The amount of materials provided; q i is the total amount of materials stored at rescue point i; Q0 is the initial amount of materials required for rescue; a is the growth coefficient of the amount of rescue materials and this coefficient is a constant; a t is the time conversion coefficient for rescue materials, that is, an exponential function with a as the base and time as the variable. In the above formula, t=T or T-t1; T is the shortest emergency rescue time; t i It is the time from the occurrence of the accident to the arrival of the rescue point i at the accident scene.

[0121] The setting principle of a is: in the jurisdiction, when the demand for relief supplies is expected to increase over time, such as in the early stages of a disaster or when rescue resources are limited, the a coefficient should be set to greater than 1 to reflect the increase in demand for supplies; on the contrary, if the demand for relief supplies is expected to decrease over time, such as when rescue operations are carried out efficiently or when supplies are sufficient, the a coefficient can be set to less than 1 to reflect the decrease or stability of demand for supplies. This setting helps to simulate and optimize emergency resource scheduling more accurately. The default value is generally set to 1.2.

[0122] (twenty four)

[0123] (25)

[0124] In the formula: z1 is the emergency rescue time, T is the shortest emergency rescue time; z2 is the emergency rescue cost; u i The manpower and vehicle costs required for rescue at rescue point i; v i The unit price of the materials provided by rescue point i; x i is the amount of materials involved in the rescue at rescue point i, and n is the total number of rescue points that can provide rescue for the accident point.

[0125] Formula (22) indicates that the amount of rescue materials that can be provided by each emergency rescue point should be less than or equal to the total amount of rescue materials stored at the point; Formula (23) indicates that the sum of the time-converted amount of rescue materials at each emergency rescue point should be greater than the time-converted amount of materials required for the emergency event to meet the rescue needs. The amount of rescue materials is calculated according to different types of materials.

[0126] The highway resource scheduling model can be regarded as a multi-objective optimization problem for solution. When multiple objectives (rescue time and rescue cost) are sought for optimization at the same time, it is usually impossible to obtain the optimal solution for each objective. Therefore, the hierarchical sequence method is used to solve the above model. In emergency rescue work, when the shortest rescue time and the lowest rescue cost cannot be met, the rescue time should usually be the primary consideration. Therefore, in the process of solving the model, the shortest time T required to complete the emergency rescue work should be obtained first, and then this time should be used as a constraint condition, and the traffic capacity of the highway should be comprehensively considered to obtain the emergency rescue material scheduling plan with the lowest rescue cost.

[0127] Define the demand for materials at the accident point at time t, Q t and the amount of materials involved in the rescue S t for:

[0128] (26)

[0129] (27)

[0130] Where: Qt+1 , Q t are the remaining material demand at time t and time t+1 respectively; S t is the amount of materials involved in the rescue at time t.

[0131] The shortest emergency rescue time is to sort the emergency rescue points according to the time they can implement rescue around the accident rescue point, select the emergency rescue points that can implement rescue and compare whether the amount of rescue materials they can provide meets the needs of the rescue points. When the rescue needs are met, this moment is the shortest emergency rescue time. When it is less than 0, it means that the requirement is not only met, but also there is an additional buffer to improve the reliability of rescue; When , it indicates that the rescue point needs more rescue points for rescue, and the above process is repeated until the rescue demand is met. The shortest emergency rescue time T is obtained through iteration, so the above model is simplified to a single-objective model. After simplification, the objective function is the same as formula (25), and the constraints are formulas (22) and (23) with the addition of time limit. The model is as follows:

[0132] (28)

[0133] St:

[0134] (29)

[0135] (30)

[0136] (31)

[0137] Formula (30) indicates that the moment when rescue point i participates in the rescue should be earlier than the shortest time T required to complete the emergency rescue work.

[0138] A3.2 Solving the emergency resource scheduling model

[0139] The greedy algorithm is a method that does not pursue the overall optimal solution, but only hopes to obtain a relatively satisfactory solution for the current step. Generally, a satisfactory solution can be obtained quickly because it saves a lot of time that must be spent on exhausting all possibilities to find the optimal solution. Since the emergency resource scheduling problem is carried out step by step, it is judged whether each step meets the rescue needs. Therefore, the greedy algorithm is used to find the optimal solution for each step. The algorithm flow is as follows:

[0140] Step 1: Obtain accident rescue information, set the current time as the initial time, estimate the rescue demand Q0 according to the on-site situation, and put the rescue points that meet the requirements near the accident point area into set A1.

[0141] Step 2: Carry out rescue work one by one in the order of the time when the rescue points in set A1 arrive at the accident site. The time of arrival at the accident site is t, and the amount of rescue supplies provided is recorded S t .

[0142] Step 3: Determine and calculate the difference between the amount of materials needed for the current accident and the amount of rescue materials already supplied at the rescue point ,like , it means that the amount of materials provided by these rescue points cannot meet the rescue needs, and we need to continue to seek other rescue points in set A1 for rescue. According to formula (26), the rescue demand Q of the accident site at the current moment is calculated: t , iterate and go to step 2; if , it means that the rescue has been completed and the rescue requirements are met. The current moment is the shortest rescue time T, and go to step 4.

[0143] Step 4: Calculate the time conversion amount of the initial rescue demand at the accident site at time T At this time, the remaining material demand is the time conversion amount Calculate the capacity cost ratio of the rescue points participating in the rescue in set A1 at time T , put into another set A2.

[0144] Step 5: Calculate the conversion capacity of the amount of supplies that the selected rescue point i can provide for time T , to determine whether the conversion capacity at that moment meets the rescue demand. If , it means that we need to continue to select rescue point i in set A2 for rescue, and let , go to step 5; if , it means that the amount of rescue materials provided by the rescue point has met the rescue needs, the rescue work is completed, and according to the formula Calculate the amount of supplies provided by the last rescue point i , the calculation ends.

[0145] A4 Intelligent rescue plan generation based on vehicle traffic pattern simulation

[0146] The purpose of generating this plan is to intelligently match rescue plans for accidents, which can serve as a good auxiliary for manual rescue measures, speed up the accident handling process, and reduce decision-making errors caused by relying entirely on manual experience.

[0147] Based on the type of accident vehicle in a highway traffic accident, the specific location of the accident, and the vehicle traffic behavior pattern in the target area within a period of time after the accident, such as flow rate, proportion of passenger cars and trucks, vehicle speed, etc. (this data comes from vehicle profiling), combined with the impact range of the accident, a variety of accident scenarios can be generated.

[0148] In order to generate an intelligent rescue plan library, the present invention parameterizes all modules that constitute the scene and simulates and deduces vehicle traffic patterns for all possible accident scenarios. The simulation content includes: accident scenario simulation, rescue resource scheduling simulation, traffic control measures simulation and simulation data services.

[0149] A4.1 Accident scenario system design

[0150] Accident scenarios are graded and classified according to the type of accident and scope of impact.

[0151] Regarding the classification of highway traffic accident types, the present invention starts from the aspects of vehicle type, accident location, lane, traffic conditions, severity, weather, etc., and combines machine learning, data mining and other analysis methods to explore various types of highway traffic accidents. Table 1 is an example of the classification of accident types (i.e., the aforementioned traffic accident situation k):

[0152] Table 1 Typical accident classification scenarios

[0153]

[0154] A4.2 Design and screening of emergency plan library framework

[0155] Based on the database of accident scenarios, design the storage format of the plan library to realize the storage of plans, design the retrieval mechanism of plans to realize the rapid search of plans, and design the update mechanism of plans to realize the dynamic update of plans;

[0156] Generate active control measures for each accident plan, including speed limit, ramp (toll booth) closure, diversion, signal light, hard shoulder opening, information board release, etc.;

[0157] The accident conditions of the plan and the corresponding control measures are parameterized to form a complete emergency plan alternative library to achieve intelligent screening of plans;

[0158] The evaluation indicators of the design plan include efficiency, cost, risk and other aspects, and the design evaluation method is used to evaluate the effectiveness and feasibility of the plan.

[0159] A4.3 Simulation

[0160] Simulation environment construction: based on road network model, vehicle model, plan library and emergency resource dispatch information, accident scenario simulation, rescue resource dispatch simulation, traffic control measures simulation and simulation data service;

[0161] Evaluation: Efficiency analysis, congestion elimination analysis and indicator evaluation analysis are carried out based on the simulation process data and simulated traffic situation data.

Claims

1. A method for site selection and resource scheduling of high-speed emergency rescue facilities, It is characterized in that include: Obtain historical data on highway accidents, count sections with high accident rates as emergency demand points, and determine candidate emergency facility points based on the highway network. Establish a network diagram with emergency demand points and candidate emergency facility points as points, and roads connecting emergency demand points and between emergency demand points and candidate emergency facility points as arcs. Traffic accidents in historical data are divided into several situations, and the emergency demands under different situations of traffic accidents at different emergency demand points are calculated; for the emergency facility points to be selected in the network diagram, based on the emergency demand of each emergency demand point, with the goal of maximizing the emergency resource supply capacity and minimizing the rescue travel time, combined with the rescue deadline constraint and the facility point quantity constraint, an emergency facility point site selection model is established and solved to select the emergency facility point; Based on the selected emergency facility points, an emergency resource scheduling model is established and solved for the location where the accident occurred, with the goal of minimizing emergency rescue time and minimizing rescue costs, to obtain the scheduling plans for each emergency demand point under different accident situations.

2. The method according to claim 1, It is characterized in that The emergency demand under different traffic accidents at different emergency demand points is calculated based on the following formula: ; ; Where N i (k) is the emergency demand point N i The emergency demand in case of traffic accident k; α ik is the emergency demand point N i The probability of a traffic accident situation k occurring at I ik For the traffic accident situation k to the emergency demand point N i The influence coefficient of is the emergency demand point N when each traffic accident occurs i The traffic capacity of C i is the emergency demand point N i The design capacity of Q i is the emergency demand point N i The average traffic flow of the road section; n ik is the emergency demand point N i The number of traffic accidents in k situations; N is the emergency demand point N i The total number of .

3. The method according to claim 1, It is characterized in that The objective function of the accident emergency resource facility point screening model is set as follows: ; ; Where N j (k) is the number of emergency facility points N to be selected when traffic accident k occurs j The amount of supplies that can be provided within the rescue period; N i (k) is the emergency demand point N i Emergency needs in the event of a traffic accident k; t ij For rescue vehicles from the candidate emergency facility point N j To emergency demand point N i travel time of the shortest path; y ij is a decision variable, when N i Points are N j When the rescue range of the point is covered, take y ij =1, otherwise y ij =0, the rescue range is the preset value; x j is a decision variable. When N j When setting up a facility point, take x j =1, otherwise x j =0; I is the set of emergency demand points in the network diagram; J is the set of emergency facility points to be selected in the network diagram.

4. The method according to claim 1, It is characterized in that When the emergency facility site selection model is established, the shortest time required to complete the emergency rescue work is first calculated, and the shortest time is used as a constraint condition and the minimization of rescue costs is used as the objective function to obtain an emergency rescue material scheduling plan.

5. The method according to claim 4, It is characterized in that The shortest time required for emergency rescue work is obtained by sorting the emergency facilities according to the rescue time based on the time required for rescue at each emergency facility point around the accident site, and calculating the amount of materials at the rescue point. The shortest time is the shortest emergency rescue time of one or more emergency facility points that meet the material needs of the accident site.

6. The method according to claim 4, It is characterized in that The shortest time required for the emergency rescue work is obtained by solving the following formula: ; ; In the formula, Q t+1 , Q t are the remaining material demand at time t and time t+1 respectively; S t is the amount of materials involved in the rescue at time t; x i The amount of supplies provided to rescue point i; n is the total number of rescue points, and a is a constant; when When the rescue demand is met, this moment is the shortest emergency rescue time; when When , rescue points are added and the solution is iterated until the rescue demand is met.

7. The method according to claim 4, It is characterized in that The emergency resource scheduling model also includes the remaining material time conversion constraint: ;; In the formula, Q 0 is the initial amount of materials required for rescue; a is a constant; T is the shortest rescue time; t i is the time when the rescue point i arrives at the accident scene; i The amount of supplies provided to rescue point i.

8. The method according to claim 1, It is characterized in that The accident emergency resource facility point screening model is solved based on a heuristic algorithm based on Lagrangian relaxation; and the emergency resource scheduling model is solved based on a greedy algorithm.

9. A high-speed emergency rescue facility site selection and resource scheduling system, It is characterized in that include: The network diagram construction module obtains historical data on highway accidents, counts sections with high accident rates as emergency demand points, and determines candidate emergency facility points based on the highway network. The network diagram is established with emergency demand points and candidate emergency facility points as points, and roads connecting emergency demand points and between emergency demand points and candidate emergency facility points as arcs. The emergency facility site selection module divides the traffic accidents in the historical data into several situations, and calculates the emergency demand under different emergency demand points and different situations of traffic accidents; for the candidate emergency facility points in the network diagram, based on the emergency demand of each emergency demand point, with the goal of maximizing the emergency resource supply capacity and minimizing the rescue travel time, combined with the rescue deadline constraint and the facility point quantity constraint, the emergency facility site selection model is established and solved to select the emergency facility points; The emergency resource scheduling module, based on the selected emergency facility points, establishes and solves the emergency resource scheduling model for the location where the accident occurred with the goal of minimizing emergency rescue time and minimizing rescue costs, and obtains the scheduling plan for each emergency demand point under different accident situations.

10. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instructions are executed by a processor, the steps of the method described in any one of claims 1 to 8 are implemented.

Images