A resource dynamic scheduling method and system for post-disaster emergency repair of highways

Abstract

The present application relates to a kind of resource dynamic scheduling method and system for highway post-disaster emergency repair, method includes: obtaining emergency repair information;Emergency repair information is input into double-layer optimization model and obtains optimal global scheduling strategy, double-layer optimization model includes upper global strategy optimization submodel and lower layer loading and unloading plan dynamic optimization submodel, upper global strategy optimization submodel is used to generate candidate global scheduling strategy according to emergency repair information, and according to the fitness of candidate global scheduling strategy, population is iterated evolution, obtains optimal global scheduling strategy;Lower layer loading and unloading plan dynamic optimization submodel according to candidate global scheduling strategy executes specific travel path planning, and simulation optimization and evaluation are carried out to loading and unloading plan, obtain the total travel time of vehicle of candidate global scheduling strategy and global scheduling performance penalty, and return to upper global strategy optimization submodel as fitness value.Compared with prior art, the present application significantly improves scheduling efficiency and adaptability.

Description

Technical Field

[0001] This invention relates to the field of emergency management and operations optimization technology, and in particular to a method and system for dynamic resource scheduling for emergency repair of highways after disasters. Background Technology

[0002] In the context of climate change, proactively preventing and mitigating disaster risks has become a global consensus. As the lifeline for emergency rescue during disasters, improving the inherent safety level and enhancing emergency response capabilities of the highway network is urgently needed. Disaster events disrupt transportation, severely impacting disaster relief operations, material transport, and daily traffic needs. Ensuring smooth traffic flow within the critical "golden 72 hours" after a disaster is paramount. Coordinating safety and development, rapidly and efficiently organizing repair resources, and repairing damaged road sections to restore the road network's functionality as quickly as possible are of paramount importance.

[0003] The highway emergency repair resource scheduling problem is an extremely complex combinatorial optimization problem, essentially a variant of the Vehicle Routing Problem (VRP). Existing research often employs traditional operations research methods or static intelligent algorithms (such as standard genetic algorithms and ant colony optimization) to solve this problem. These methods typically assume that the road network state and resource requirements remain constant throughout the scheduling cycle, i.e., performing a one-time static optimization to generate a complete scheduling plan from start to finish.

[0004] However, in real-world emergency repair scenarios, the situation is far more complex and dynamic than this:

[0005] 1) Information ambiguity: In the early stages of a disaster, information such as the specific location of the disaster site, the extent of damage, and the amount of resources required may not be entirely accurate and will be dynamically updated as the repair work progresses.

[0006] 2) Environmental dynamics: New disaster sites may appear during the emergency repair process, or a repaired section of road may be interrupted again due to secondary disasters.

[0007] 3) State Interactivity: The execution of a scheduling plan will change the system state. After a road is repaired, the topology of the entire road network and the shortest paths for vehicles will change accordingly, which will directly affect subsequent scheduling decisions.

[0008] Traditional static scheduling methods cannot effectively cope with the above dynamic and uncertainty. The scheduling scheme generated by them is often very fragile. Once the actual situation deviates from the preset model, the entire scheme may become no longer optimal, or even completely infeasible, leading to resource waste, missed rescue opportunities, etc., greatly restricting the feasibility, timeliness and global optimality of scheduling decisions. For example, patent application CN116913090A discloses a city road network post-disaster recovery method and system considering traffic resilience. The method abstracts the city road network as a weighted undirected graph, establishes an upper optimization decision model and a lower traffic distribution model, and uses discrete particle swarm algorithm and Frank-Wolfe algorithm to iteratively solve, to generate a static damaged road repair sequence and engineering team scheduling scheme. The method can only perform one-time global static optimization based on complete information at the initial stage of post-disaster, and cannot perform online rescheduling or dynamic feedback adjustment according to uncertain factors such as post-disaster information ambiguity, environmental dynamic changes and new disaster points, leading to the fragility and invalidation of the scheduling scheme in real emergency scenarios, and the inability to effectively cope with the complex and changeable repair demand within the "golden 72 hours".

[0009] In particular, when multiple continuous disaster points occur on the same road section, the material demand of some disaster points in the middle part of the road section will face the problem of being unable to be satisfied or delayed to be satisfied. At this time, the static model basically cannot give a feasible solution. Therefore, there is an urgent need for a dynamic scheduling method for highway emergency repair resources that can adapt to changes in the post-disaster highway transportation system to improve the feasibility and effectiveness of the scheduling scheme. SUMMARY

[0010] The purpose of the present application is to overcome the defects of the prior art and provide a resource dynamic scheduling method and system for post-disaster emergency repair of highways, which significantly improves the scheduling efficiency and adaptability.

[0011] The purpose of the present application can be achieved by the following technical solutions:

[0012] A resource dynamic scheduling method for post-disaster emergency repair of highways, comprising the following steps:

[0013] Obtaining emergency repair information;

[0014] The emergency repair information is input into a double-layer optimization model to obtain an optimal global scheduling strategy, the double-layer optimization model comprising an upper-layer global strategy optimization sub-model and a lower-layer loading and unloading plan dynamic optimization sub-model, the upper-layer global strategy optimization sub-model being configured to generate a candidate global scheduling strategy according to the emergency repair information, and to perform iterative evolution on a population consisting of multiple candidate global scheduling strategies according to fitness of the candidate global scheduling strategy to obtain the optimal global scheduling strategy; the lower-layer loading and unloading plan dynamic optimization sub-model being configured to perform specific travel path planning according to the candidate global scheduling strategy, and to perform simulation optimization and evaluation on a loading and unloading plan to obtain vehicle total travel time and global scheduling performance penalty of the candidate global scheduling strategy, and to return the vehicle total travel time and global scheduling performance penalty as fitness of the candidate global scheduling strategy to the upper-layer global strategy optimization sub-model.

[0015] Further, the emergency repair information comprises multiple types of initial road network topology state, transport capacity distribution and resource distribution.

[0016] Further, the upper-layer global strategy optimization sub-model performs iterative evolution on the population consisting of multiple candidate global scheduling strategies according to fitness of the candidate global scheduling strategy by using an NSGA-II algorithm to obtain the optimal global scheduling strategy, and the specific steps comprise:

[0017] S101, after obtaining fitness of all candidate global scheduling strategies in the entire population, non-dominated sorting and congestion degree calculation are adopted to select excellent individuals in the population into the next generation;

[0018] S102, a sequence-oriented crossover operator is adopted to randomly select one task segment delimited by a warehouse delimiter from parent P1 and parent P2, and then exchange the two task segments to generate offspring O1 and offspring O2;

[0019] S103, perturbation is performed on the offspring sequence, and the perturbation comprises randomly exchanging positions of two disaster points in the sequence, moving one disaster point to another position of the sequence, and randomly adding or deleting one access to a certain disaster point;

[0020] S104, steps S101 to S103 are repeated until a maximum iteration number or solution set convergence is reached to obtain the optimal global scheduling strategy.

[0021] Further, the lower-layer loading and unloading plan dynamic optimization sub-model adopts a rolling horizon optimization framework, and after receiving a certain candidate global scheduling strategy transmitted by the upper-layer global strategy optimization module, performs simulation optimization and evaluation on the candidate global scheduling strategy, and the specific steps comprise:

[0022] S201, setting a prediction time domain [t, t+Tp] and an execution time domain [t, t+Te] at a current decision time t, and establishing a vehicle loading and unloading plan optimization sub-problem in the prediction time domain [t, t+Tp] to obtain a vehicle visiting sequence;

[0023] S202, solving the vehicle loading and unloading plan optimization sub-problem for vehicle distribution tasks in the prediction time domain [t, t+Tp] to determine a vehicle driving path and a material loading and unloading plan in future time;

[0024] S203, executing the vehicle driving path and the material loading and unloading plan in the corresponding execution time domain [t, t+Te];

[0025] S204, after the end of the execution time domain [t, t+Te], advancing the system time to t+Te, and updating the road network traffic state and the resource holding state of each disaster point, each warehouse and each vehicle according to the plan execution result;

[0026] S205, taking t+Te as a new current decision time, returning and repeating the sub-steps S201 to S204 until a preset simulation end condition is met.

[0027] Further, the decision variables of the vehicle loading and unloading plan optimization sub-problem include the number of repair resources loaded by the vehicle at the material storage center and the number of repair resources unloaded by the vehicle at the disaster point, the constraint conditions are the vehicle load at the current time, the real-time inventory of the warehouse and the real-time remaining demand of the disaster point, and the global scheduling performance penalty is minimized as the objective function of the vehicle loading and unloading plan optimization sub-problem.

[0028] Further, the global scheduling performance penalty is minimized as:

[0029]

[0030] In the formula, Z2 is the global scheduling performance penalty, C S is a preset unit material shortage penalty factor, is the material shortage amount of the disaster point f at time t, C H is a preset unit material accumulation and idling penalty factor, is the material redundancy amount of the disaster point f at time t, F is a set of disaster points, and T is a set of times.

[0031] Further, in step S202, according to the solving result of the vehicle loading and unloading plan optimization sub-problem and the road network topology state at the corresponding time, the shortest path of the vehicle from the current position to the next node in the visiting sequence is calculated to obtain the vehicle driving path in future time.

[0032] Further, in step S204, the road network traffic state and resource holding state are updated according to the planned execution result, and the specific steps include: according to the executed plan execution result, updating the resource inventory, cumulative consumption, shortage and idle amount of each disaster point, and the action and position of the vehicle, judging whether the cumulative consumption of the disaster point reaches the maximum demand, if yes, recording the time of the disaster point being opened, and updating the road network topology state accordingly.

[0033] Further, the preset simulation end condition is that all disaster points are repaired and the road network is completely opened.

[0034] According to another aspect of the present application, a resource dynamic scheduling system for post-disaster emergency repair of highways is provided, comprising:

[0035] An emergency repair information acquisition module is configured to acquire emergency repair information.

[0036] A global scheduling strategy optimization module is configured to input the emergency repair information into a double-layer optimization model to obtain an optimal global scheduling strategy, wherein the double-layer optimization model comprises an upper-layer global strategy optimization sub-model and a lower-layer loading and unloading plan dynamic optimization sub-model.

[0037] Compared with the prior art, the present application has the following beneficial effects:

[0038] 1. By constructing a double-layer optimization model, the upper-layer global strategy optimization sub-model is used to generate a candidate global scheduling strategy according to the emergency repair information, and the population composed of multiple candidate global scheduling strategies is iteratively evolved according to the fitness of the candidate global scheduling strategy to obtain an optimal global scheduling strategy; the lower-layer loading and unloading plan dynamic optimization sub-model is used to perform specific driving path planning according to the candidate global scheduling strategy, and to simulate, optimize and evaluate the loading and unloading plan to obtain the total driving time of the vehicle and the global scheduling performance penalty of the candidate global scheduling strategy, and the total driving time of the vehicle and the global scheduling performance penalty are returned to the upper-layer global strategy optimization sub-model as the fitness value of the candidate global scheduling strategy, thereby avoiding searching in a huge and full of invalid solution mixed space, greatly reducing the solution space, improving the convergence speed and solution quality of the algorithm, and significantly improving the scheduling efficiency.

[0039] 2. The lower layer loading and unloading plan dynamic optimization sub-model in the application adopts a rolling horizon optimization (RHO) framework, simulates and optimizes and evaluates a candidate global scheduling strategy after receiving the candidate global scheduling strategy delivered by the upper layer global strategy optimization module, dynamically adjusts the scheduling plan according to the real-time updated road network state and resource information in the repair process, effectively responds to emergencies, and improves the robustness of the generated scheduling scheme.

[0040] 3. The double-layer optimization model in the application adopts a multi-objective optimization algorithm, and strictly obeys real-time constraints such as vehicle capacity, inventory limit, and road network connectivity, ensures that the finally generated scheduling scheme is always executable and effective in the entire dynamic execution process, can still give a high-quality feasible solution in a complex scenario (multiple consecutive disaster points in multiple road sections) that cannot be solved by traditional static scheduling methods, and improves the adaptability of the algorithm. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 A flowchart of a resource dynamic scheduling method for post-disaster emergency repair of highways according to the application is shown.

[0042] Figure 2 An algorithm flowchart of the fusion of rolling horizon optimization and NSGA-II algorithm is shown.

[0043] Figure 3 An illustration of the chromosome crossover operator in the embodiment is shown. DETAILED DESCRIPTION

[0044] The application will be described in detail below in conjunction with the drawings and specific embodiments. The embodiments are implemented on the premise of the technical solution of the application, and detailed implementation methods and specific operation processes are given, but the protection scope of the application is not limited to the following embodiments.

[0045] English abbreviations involved:

[0046] Non-dominated Sorting Genetic Algorithm II, NSGA-II

[0047] Receding Horizon Optimization, RHO

[0048] Discrete Event Simulation, DES

[0049] Embodiment 1

[0050] The embodiment provides a resource dynamic scheduling method for post-disaster emergency repair of highways, as shown in the following steps: Figure 1

[0051] S1, obtaining emergency repair information.

[0052] The emergency repair information comprises multiple types of initial road network topology state, transport capacity distribution and resource distribution.

[0053] S2, inputting the emergency repair information into a double-layer optimization model to obtain an optimal global scheduling strategy.

[0054] The double-layer optimization model comprises an upper-layer global strategy optimization submodel and a lower-layer loading and unloading plan dynamic optimization submodel. The upper-layer global strategy optimization submodel is used for generating a candidate global scheduling strategy according to the emergency repair information, and iteratively evolving a population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategy, so as to obtain the optimal global scheduling strategy. The lower-layer loading and unloading plan dynamic optimization submodel executes specific driving path planning according to the candidate global scheduling strategy, and simulates, optimizes and evaluates the loading and unloading plan, so as to obtain the total driving time of vehicles and the global scheduling performance penalty of the candidate global scheduling strategy, and returns the total driving time of vehicles and the global scheduling performance penalty to the upper-layer global strategy optimization submodel as the fitness value of the candidate global scheduling strategy.

[0055] The present application adopts multi-objective optimization, aiming to achieve the best balance between efficiency and effect at the same time. Specifically, the following two objectives are included:

[0056] Minimizing total driving time (Z1): this objective focuses on the scheduling efficiency of the repair resources, and strives to complete the task with the lowest transportation time. Its mathematical expression form is as follows:

[0057]

[0058] In the formula, Z1 is the total driving time, K, N and T are the sets of vehicles, nodes and time respectively; is a decision variable, and is 1 if the vehicle k drives from the node i to the node j at time t, otherwise 0; ω ij is the driving time from the node i to the node j.

[0059] Minimizing global scheduling performance penalty (Z2): this objective comprehensively evaluates the satisfaction effect of the scheduling scheme on the demand, aiming to avoid material shortage and resource idle waste. Its mathematical expression form is as follows:

[0060]

[0061] In the formula, Z2 is the global scheduling performance penalty, C S is a preset unit material shortage penalty factor, ​To determine the material shortage at disaster point f at time t, C H The pre-set penalty factor for stockpiling and leaving idle unit materials. Let F be the material redundancy at disaster point f at time t, where F is the set of disaster points and T is the time set.

[0062] The core decision variables of this invention include two categories: vehicle travel route and loading / unloading plan, as shown in the table below.

[0063] Table 1 Core Decision Variables

[0064]

[0065] The two-level optimization model includes the following key constraints:

[0066] 1) Constraints on updating the resource status of disaster-damaged sites

[0067] Actual material consumption at disaster site f at time t Depends on its current stockpile of materials Compared with the current vehicle unloading volume The sum of the expected consumption Expected maximum material consumption D fmax The minimum of the differences between the total amount consumed and the cumulative amount consumed. The remaining amount at disaster point f at time t+1. Then based on the holding quantity at time t Received supply amount Consumption at time t It is dynamically updated. Its mathematical expression is as follows:

[0068]

[0069] In the formula, This represents the actual amount of resources consumed at disaster point f at time t. The amount of supplies at time t. Let K be the amount of unloaded by vehicle K at time t at disaster point f. For the projected consumption, D fmax This represents the estimated maximum consumption of materials.

[0070] 2) Constraints on updating the repair status of disaster sites

[0071] When the cumulative repair resource consumption of disaster point f reaches its demand D fmax When the damaged point is considered to have been repaired, it is considered to have been reopened. This constraint provides a precise basis for determining whether the road network topology has changed. Its mathematical expression is as follows:

[0072]

[0073] In the formula, Df(t) is the actual material consumption of disaster point f at time t fmax Df(t) is the actual material consumption of disaster point f at time t Df(t) is the actual material consumption of disaster point f at time t Df(t) is the actual material consumption of disaster point f at time t

[0074] 3) Dynamic road network connectivity constraint

[0075] This constraint prohibits any vehicle from passing through disaster point f as a passing node before it is repaired. Its logic can be expressed as: if vehicle k visits an unrepaired disaster point f at time t It cannot depart from this point to any other node except the last passing point. This constraint ensures that all path planning is based on the current real-time, truly passable road network state, which is a direct embodiment of the dynamic adaptability of the invention. Its mathematical expression is as follows:

[0076]

[0077] In the formula, Df(t) is the actual material consumption of disaster point f at time t if Df(t) is the actual material consumption of disaster point f at time t Df(t) is the actual material consumption of disaster point f at time t if Df(t) is the actual material consumption of disaster point f at time t Df(t) is the actual material consumption of disaster point f at time t

[0078] In addition to the above core constraints, in order to ensure that the scheduling scheme is completely feasible in physics and logic, the model also includes a series of conventional technical constraints. Those skilled in the art can easily construct the complete mathematical form of these constraints according to the technical idea of the invention. These conventional constraints mainly include:

[0079] 1) Vehicle path constraint

[0080] It includes stipulating that all vehicles must start from the warehouse and finally return to the warehouse; ensuring that the in-degree and out-degree of the vehicle at any intermediate node are consistent; and ensuring that at any time, a vehicle can only be in a unique state of driving or staying.

[0081] 2) Vehicle action logic constraint

[0082] It includes limiting the vehicle to load materials only at the pre-set warehouse and unload materials only at the disaster point; and ensuring that the driving and staying state transition of the vehicle conforms to common logic.

[0083] 3) Physical and resource limitation constraint

[0084] The load of the vehicle does not exceed its maximum capacity; the total travel time or total mileage of the vehicle in a single trip does not exceed the upper limit; the cumulative unloading of all vehicles does not exceed the cumulative loading; and the total delivery of the warehouse center does not exceed its initial inventory.

[0085] Through the organic combination of the above objective function and constraint conditions, the application constructs an optimization model that can accurately reflect the dynamic nature of real repair engineering and guide the generation of high-quality scheduling schemes, providing a solid foundation for subsequent system module solving.

[0086] As shown in Figure 2 , the upper global strategy optimization sub-model executes the iterative evolution of the population composed of multiple candidate global scheduling strategies through the NSGA-II algorithm as follows:

[0087] The upper global strategy optimization sub-model generates a set of high-quality initial solutions using a heuristic greedy strategy. For each vehicle, the "nearest principle" is adopted to induce it to preferentially visit the disaster point closest to it and with a demand for materials; when the vehicle is empty, it goes to the nearest warehouse for replenishment.

[0088] The NSGA-II algorithm uses a hierarchical decoupling chromosome coding scheme, with each chromosome only encoding one high-level access sequence composed of the labels of disaster points and warehouses and separators for distinguishing different vehicles, defining the order of vehicle visits to disaster points and trip division. The chromosomes in the population are coded as shown in Figure 3 , where the chromosome is a one-dimensional integer array that only encodes the access sequence.

[0089] Disaster point identifier: represented by a positive integer, such as F1, F2,...

[0090] Warehouse separator: represented by a specific positive integer, such as C1, C2,... It serves as a partitioning symbol for different trips of the same vehicle.

[0091] Vehicle separator: represented by a negative integer, such as -1. It serves as a partitioning symbol for different vehicles.

[0092] An example of a chromosome is: [C1, F2, F1, C1, F4, F3, F1, C2, -1, C1, F4, F2, C2].

[0093] The decoding of this code is completely handed over to the lower module. It only defines the sequence of tasks, for example, the access sequence of vehicle 1 is [C1, F2, F1, C1, F4, F3, F1, C2], starting from warehouse C1, passing through disaster points F2, F1, returning to C1 for replenishment, and then continuing to disaster points F4, F3, F1, and returning to C2; the access sequence of vehicle 2 is [C1, F4, F2, C2], starting from warehouse C1, passing through disaster points F4, F3, and returning to C2. This design completely decouples the combination of the access sequence and the numerical decision of the loading and unloading amount, greatly improving the robustness and efficiency of the algorithm.

[0094] The upper global strategy optimization sub-model iteratively evolves the population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategies by the NSGA-II algorithm, to obtain the optimal global scheduling strategy, and the specific steps include:

[0095] S101, after obtaining the fitness of all candidate global scheduling strategies in the entire population, non-dominated sorting and congestion calculation are adopted to select excellent individuals in the population into the next generation;

[0096] S102, a sequence-oriented crossover operator is adopted, a task segment delimited by a warehouse separator is randomly selected from each of the parent P1 and the parent P2, and then the two task segments are exchanged to generate the offspring O1 and the offspring O2;

[0097] S103, the offspring sequence is disturbed, and the disturbance includes randomly exchanging the positions of two disaster points in the sequence, moving a disaster point to another position in the sequence, and randomly increasing or deleting an access to a disaster point;

[0098] S104, steps S101 to S103 are repeated until the maximum number of iterations or the solution set converges, to obtain the optimal global scheduling strategy.

[0099] The lower loading and unloading plan dynamic optimization sub-model adopts a rolling horizon optimization (RHO) framework, simulates and optimizes and evaluates a candidate global scheduling strategy (chromosome) after receiving the candidate global scheduling strategy from the upper global strategy optimization module, and the specific steps include:

[0100] S201, at the current decision time t, set the prediction time domain [t, t+Tp] and the execution time domain [t, t+Te], and establish a vehicle loading and unloading plan optimization sub-problem in the prediction time domain [t, t+Tp] to obtain the access sequence of the vehicle.

[0101] Reset the simulation time t = 0, the road network is restored to the initial state after the disaster, and the access sequence table of all vehicles is decoded according to the access sequence. The entire scheduling period is divided into a series of continuous prediction time domains Tp and shorter execution time domains Te, for example, Tp = 10 and Te = 5. According to the current known road network state and the vehicle travel path passed down by the upper layer, the node sequence that each vehicle can access in the [t, t + Tp] is intercepted. For example, the access sequence of vehicle 1 in this prediction time domain is [C1, F2, F1, C1, F4], and the access sequence of vehicle 2 is [C1, F4, F2].

[0102] S202, for the vehicle distribution task in the prediction time domain [t, t + Tp], solve the vehicle loading and unloading plan optimization sub-problem to determine the vehicle travel path and material loading and unloading plan in the future time.

[0103] The decision variables of the vehicle loading and unloading plan optimization sub-problem include the number of rescue resources loaded by the vehicle at the material storage center and the number of rescue resources unloaded by the vehicle at the disaster site, the constraint conditions are the current vehicle load, the real-time inventory of warehouse C1, the real-time remaining demand of disaster sites F1, F2 and F4, and the objective function of the vehicle loading and unloading plan optimization sub-problem is to minimize the global scheduling performance penalty. According to the solving result of the vehicle loading and unloading plan optimization sub-problem and the road network topology state at the corresponding time, the Dijkstra or A* algorithm is used to calculate the shortest path of the vehicle from the current position to the next node in the access sequence, and the vehicle travel path in the future time is obtained.

[0104] S203, execute the vehicle travel path and material loading and unloading plan in the corresponding execution time domain [t, t + Te].

[0105] The vehicle only executes the material scheduling plan in the execution time domain [t, t + Te] in the optimization sub-problem solution, and the vehicle travels and loads and unloads materials according to this plan. For example, vehicle 1 can only execute part of the plan in the prediction time domain in [t, t + Te]: vehicle 1 loads 60 tons of goods at C1 and unloads 40 tons at disaster site F2, and drives to the next disaster site F1, t + Te is in the process of driving to disaster site F1; vehicle 2 loads 40 tons of goods at C2 and drives to disaster site F4, t + Te is in the process of driving to disaster site F4.

[0106] S204, after the execution time domain [t, t + Te] ends, the system time is advanced to t + Te, and the road network resistance state and the resource holding state of each disaster site, each warehouse and each vehicle are updated according to the plan execution result.

[0107] The road network pass-through state and resource holding state are updated according to the planned execution result, and the specific steps include: according to the executed planned execution result, updating the resource inventory, cumulative consumption, shortage and idle amount of each disaster point and the action and position of the vehicle, judging whether the cumulative consumption of the disaster point reaches the maximum demand thereof, if yes, recording the time of rush-through, and updating the road network topology state accordingly. In this stage, the system state is changed due to events (road section repair, resource delivery, resource consumption, etc.) through discrete event simulation. For example, when vehicle 1 arrives at F2, the state is updated, and the update steps are as follows:

[0108] Event processing: vehicle 1 unloads 40 tons of materials at disaster point F2 at time 3.85;

[0109] State variable update: update the position of the vehicle to F2, reduce the load of the vehicle by 40 tons, and the warehouse inventory remains unchanged, and the received material amount of disaster point F2 is increased by 40 tons;

[0110] Road network topology update: check whether the cumulative received material of F2 has met the repair demand thereof. If yes, declare that F2 has been repaired globally, and immediately update the global road network map, and set the damaged road section connected therewith as passable. This update will immediately affect the path planning of all subsequent vehicles.

[0111] S205, take t+Te as a new current decision time, return and repeat sub-steps S201 to S204 until a preset simulation end condition is met.

[0112] Rolling iteration: based on the updated system state, enter the next rolling period until a preset simulation end condition is met, and the preset simulation end condition is that all disaster points are completed repair and the road network is completely opened.

[0113] Through repeated iteration of RHO, the optimal resource distribution scheme and the dynamic process of road network repair under the vehicle scheduling scheme provided by the upper NSGA-II can be simulated and optimized. After the entire access sequence simulation optimization is completed, the cumulative total travel time Z1 and the total scheduling performance penalty Z2 are extracted from the simulation log, and the pair of {Z1, Z2} values are returned to the upper global strategy optimization sub-model as the fitness value.

[0114] Through the above implementation, the strong global search ability of the genetic algorithm and the fine dynamic decision-making ability of the rolling time domain optimization are successfully combined, and the effective solution to the resource scheduling problem of highway emergency repair in a complex dynamic environment is realized.

[0115] Example 2

[0116] The embodiment provides a resource dynamic scheduling system for highway post-disaster emergency repair, which comprises:

[0117] an emergency repair information acquisition module, configured to acquire emergency repair information;

[0118] a global scheduling strategy optimization module, configured to input the emergency repair information into a double-layer optimization model to obtain an optimal global scheduling strategy, the double-layer optimization model comprising an upper-layer global strategy optimization sub-model and a lower-layer loading and unloading plan dynamic optimization sub-model, the upper-layer global strategy optimization sub-model being configured to generate candidate global scheduling strategies according to the emergency repair information, and to perform iterative evolution on a population consisting of the candidate global scheduling strategies according to the fitness of the candidate global scheduling strategies to obtain the optimal global scheduling strategy; the lower-layer loading and unloading plan dynamic optimization sub-model being configured to perform specific travel path planning according to the candidate global scheduling strategies, and to simulate optimization and evaluation on the loading and unloading plans to obtain the total travel time of the vehicles and the global scheduling performance penalty of the candidate global scheduling strategies, and to return the total travel time of the vehicles and the global scheduling performance penalty as the fitness value of the candidate global scheduling strategies to the upper-layer global strategy optimization sub-model.

[0119] The system can further comprise a data interaction interface configured to receive real-time dynamic information such as road network state updates and new disaster reports from external data sources such as traffic management centers, meteorological departments, and handheld terminals of on-site repair teams.

[0120] Meanwhile, the system can further comprise a user interface module configured to show the Pareto optimal solution set to a decision maker, and to present the finally selected scheduling scheme in a visualized manner such as an electronic map or a Gantt chart. The decision maker can issue scheduling instructions to the terminal devices of the repair vehicles through the interface.

[0121] The rest is the same as in Embodiment 1.

[0122] The above detailed the preferred embodiments of the present application. It should be understood that those of ordinary skill in the art can make many modifications and variations without creative effort based on the concept of the present application. Therefore, any technical solution obtained by logical analysis, reasoning or limited experiments based on the prior art according to the concept of the present application shall be within the protection scope defined by the claims.

Claims

1. A resource dynamic scheduling method for post-disaster emergency repair of highways, characterized in that, The method comprises the following steps: obtaining emergency repair information; inputting the emergency repair information into a double-layer optimization model to obtain an optimal global scheduling strategy, wherein the double-layer optimization model comprises an upper-layer global strategy optimization sub-model and a lower-layer loading and unloading plan dynamic optimization sub-model, the upper-layer global strategy optimization sub-model is configured to generate a candidate global scheduling strategy according to the emergency repair information, and to perform iterative evolution on a population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategy to obtain an optimal global scheduling strategy; the lower-layer loading and unloading plan dynamic optimization sub-model is configured to perform specific travel path planning according to the candidate global scheduling strategy, and to perform simulation optimization and evaluation on a loading and unloading plan to obtain a total vehicle travel time and a global scheduling performance penalty of the candidate global scheduling strategy, and to return the total vehicle travel time and the global scheduling performance penalty as the fitness value of the candidate global scheduling strategy to the upper-layer global strategy optimization sub-model; wherein the upper-layer global strategy optimization sub-model performs iterative evolution on the population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategy by using an NSGA-II algorithm to obtain an optimal global scheduling strategy, and the specific steps comprise: S101, after obtaining the fitness of all candidate global scheduling strategies in the entire population, non-dominated sorting and congestion degree calculation are adopted to select excellent individuals in the population into the next generation; S102, a sequence-oriented crossover operator is adopted to randomly select one task segment delimited by a warehouse delimiter from each of a parent P1 and a parent P2, and then exchange the two task segments to generate a child O1 and a child O2; S103, perturbation is performed on the child sequence, and the perturbation comprises randomly exchanging the positions of two disaster points in the sequence, moving one disaster point to another position of the sequence, and randomly increasing or deleting an access to a certain disaster point; S104, steps S101 to S103 are repeated until a maximum iteration number or solution set convergence is reached to obtain an optimal global scheduling strategy; the lower-layer loading and unloading plan dynamic optimization sub-model adopts a rolling horizon optimization framework, and after receiving a certain candidate global scheduling strategy transmitted by the upper-layer global strategy optimization sub-model, performs simulation optimization and evaluation on the candidate global scheduling strategy, and the specific steps comprise: S201、At a current decision moment , set a prediction time domain and an execution time domain , and establish a vehicle loading and unloading plan optimization sub-problem in the prediction time domain to obtain a vehicle access sequence; S202、for the prediction time domain Solve the vehicle loading and unloading plan optimization sub-problem to determine the vehicle driving path and material loading and unloading plan in the future time for the vehicle distribution task in the prediction time domain. S203, performing a vehicle travel path in a corresponding execution time domain a vehicle travel path and a cargo handling plan within the corresponding execution time domain S204、in the execution time domain After the end, the system time is advanced to , and the road network traffic state and the resource holding state of each disaster point, each warehouse, and each vehicle are updated according to the planned execution result. The specific steps include: updating the resource inventory, cumulative consumption, shortage, and idle amount of each disaster point and the action and position of the vehicle according to the executed planned execution result, judging whether the cumulative consumption of the disaster point reaches the maximum demand, if yes, recording the time of breaking through, and updating the road network topology state accordingly; S205, will As the new current decision point, return and repeat sub-steps S201 to S204 until the preset simulation termination condition is met.

2. The method for dynamic scheduling of resources for post-disaster emergency repair of highways according to claim 1, characterized in that, The emergency repair information comprises multiple types of initial road network topology state, transport capacity distribution and resource distribution.

3. The method for dynamic scheduling of resources for post-disaster emergency repair of highways according to claim 1, characterized in that, The decision variables of the vehicle loading and unloading plan optimization sub-problem comprise the number of repair resources loaded by a vehicle at a material storage center and the number of repair resources unloaded by the vehicle at a disaster point, the constraint conditions are the current vehicle load, the real-time inventory of the warehouse and the real-time remaining demand of the disaster point, and the minimum global scheduling performance penalty is taken as the objective function of the vehicle loading and unloading plan optimization sub-problem.

4. The method for dynamic scheduling of resources for post-disaster emergency repair of highways according to claim 3, characterized in that, The minimum global scheduling performance penalty is: In the formula, For global scheduling performance penalties, The preset unit material shortage penalty factor, For the disaster point at time t The shortage of supplies The pre-set penalty factor for stockpiling and leaving idle unit materials. For the disaster point at time t The amount of surplus materials For disaster-damaged sites, It is a time set.

5. The method for dynamic scheduling of resources for post-disaster emergency repair of highways according to claim 1, characterized in that, In step S202, the shortest path of a vehicle from the current position to the next node in the access sequence is calculated according to the solution of the vehicle loading and unloading plan optimization sub-problem and the road network topology state at the corresponding time to obtain the vehicle travel path in the future time.

6. The method for dynamic scheduling of resources for post-disaster emergency repair of highways according to claim 1, characterized in that, The preset simulation end condition is that all disaster points are repaired and the road network is completely opened.

7. A resource dynamic scheduling system for post-disaster emergency repair of highways, characterized in that, The method comprises the following steps: The emergency repair information acquisition module is configured to acquire emergency repair information. The global scheduling strategy optimization module is configured to input the emergency repair information into a double-layer optimization model to obtain an optimal global scheduling strategy, wherein the double-layer optimization model comprises an upper-layer global strategy optimization sub-model and a lower-layer loading and unloading plan dynamic optimization sub-model. The upper-layer global strategy optimization sub-model is configured to generate a candidate global scheduling strategy according to the emergency repair information, and perform iterative evolution on a population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategy to obtain the optimal global scheduling strategy. The lower-layer loading and unloading plan dynamic optimization sub-model is configured to perform specific travel path planning according to the candidate global scheduling strategy, and perform simulation optimization and evaluation on the loading and unloading plan to obtain the total travel time of vehicles and the global scheduling performance penalty of the candidate global scheduling strategy, and return the total travel time of vehicles and the global scheduling performance penalty as the fitness value of the candidate global scheduling strategy to the upper-layer global strategy optimization sub-model. The upper-layer global strategy optimization sub-model is configured to perform iterative evolution on the population composed of multiple candidate global scheduling strategies according to the fitness of the candidate global scheduling strategy by using an NSGA-II algorithm to obtain the optimal global scheduling strategy, and the specific steps comprise: S101. After obtaining the fitness of all candidate global scheduling strategies in the entire population, non-dominated sorting and congestion degree calculation are performed to select excellent individuals in the population into the next generation; S102. A sequence-oriented crossover operator is used to randomly select a task segment delimited by a warehouse separator from parent P1 and parent P2, and then exchange the two task segments to generate offspring O1 and offspring O2; S103. The offspring sequence is disturbed, and the disturbance comprises randomly exchanging the positions of two disaster points in the sequence, moving a disaster point to another position in the sequence, and randomly adding or deleting an access to a disaster point; S104. Steps S101 to S103 are repeated until the maximum number of iterations or solution set convergence is reached to obtain the optimal global scheduling strategy; The lower-layer loading and unloading plan dynamic optimization sub-model uses a rolling horizon optimization framework to perform simulation optimization and evaluation on the candidate global scheduling strategy after receiving the candidate global scheduling strategy passed by the upper-layer global strategy optimization sub-model, and the specific steps comprise: S201、At a current decision moment , set a prediction time domain and an execution time domain , and establish a vehicle loading and unloading plan optimization sub-problem in the prediction time domain to obtain a vehicle access sequence; S202, Regarding the prediction time domain For the vehicle delivery tasks within the time frame, solve the vehicle loading and unloading plan optimization subproblem to determine the vehicle driving routes and material loading and unloading plans in the future. S203, performing a vehicle travel path in a corresponding execution time domain a vehicle travel path and a cargo handling plan within the corresponding execution time domain S204、in the execution time domain After the end, the system time is advanced to , and the road network traffic state and the resource holding state of each disaster point, each warehouse, and each vehicle are updated according to the planned execution result. The specific steps include: updating the resource inventory, cumulative consumption, shortage, and idle amount of each disaster point and the action and position of the vehicle according to the executed planned execution result, judging whether the cumulative consumption of the disaster point reaches the maximum demand, if yes, recording the time of breaking through, and updating the road network topology state accordingly; S205, will As the new current decision point, return and repeat sub-steps S201 to S204 until the preset simulation termination condition is met.

Images