Airport gate position distribution method and device and storage medium

Abstract

The invention provides an airport parking space distribution method and device and a storage medium, and relates to the technical field of aircraft parking space distribution, and the method comprises the steps: constructing a push-out control model based on an N-SCP strategy; performing simulation analysis on the push-out control model through a time window simulation tool to obtain a departure flight queuing length threshold value and push-out time; according to the departure flight queuing length threshold value and the push-out time, constructing a gate position distribution model; and solving the gate position distribution model based on a genetic algorithm to obtain an airport gate position distribution result. According to the method, the push-out control model and the gate position distribution model are constructed, and the push-out control and gate position distribution process is jointly optimized by using the simulation analysis result of the time window, so that the accuracy and adaptability of the models in the time dimension are improved, scientific decision support is provided for airport operation management, and the airport operation management efficiency is improved. Therefore, the parking space distribution model can comprehensively and accurately reflect the real-time condition and complexity of airport operation.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of aircraft parking stand allocation, and in particular to an airport parking stand allocation method, device and storage medium. Background Art

[0002] The airport's surface traffic system is an organic whole composed of dynamic and static elements such as aircraft, people, and taxiing environment. The surface operation situation, that is, the overall traffic state of airport aircraft taxiing, can express the external characteristics and evolution of airport traffic operation. The purpose of predicting and evaluating the traffic situation is to simulate the airport aircraft traffic flow state in the future time period, obtain the changes and development trends of the airport situation, and deeply understand and master the airport ground traffic state. This state can be described by the aircraft taxiing information per unit time, which is mainly related to factors such as aircraft taxiing flow, speed, density, and the resulting queuing time.

[0003] The push-out process is the process of the aircraft taxiing from the parking position to the take-off runway. Figure 1 As shown in the figure, it mainly includes applying for pushback from the parking position to the tower. After receiving the application, the controller will issue pushback permission to the flights in order according to the sorting results. The aircraft receives the instruction to withdraw the wheel block, connect the tractor, and be pushed out of the specified area by the tractor. The tractor and the aircraft are separated, and the aircraft turns on the auxiliary power unit and relies on its own thrust to move forward to the taxiway, and taxis according to the path assigned by the controller. During the taxiway taxiing process, the controller monitors and directs the aircraft to ensure the safe interval between aircraft. After taxiing to the end of the runway, if there is no aircraft waiting to take off in front, you can apply for takeoff from the tower. After obtaining the takeoff permission from the controller, you can take off and complete the departure process. If there is an aircraft queuing in front when taxiing to the runway, you need to queue up behind and wait for the takeoff permission from the tower controller. You must maintain the specified time interval with the aircraft taking off in front and ensure a safe interval in the airspace.

[0004] From the perspective of modeling theory, there are two main types of modeling models: linear programming models and graph network flow models. In most cases, the goal determines the modeling type and constraints, and the model determines the solution method. From the perspective of the classification of goals, the robustness objective function is often the primary consideration. The passenger-oriented objective function focuses on the attention of transfer passengers. The airline / airport-oriented goal is used to reduce costs and is usually used in combination with passenger-oriented goals or robustness-oriented goals. In terms of solution methods, heuristic models are currently being studied more. In the problem of parking space reallocation, multi-objective problems account for a large proportion. Most of the current solution methods use the method of weighted conversion to a single objective for solution, and few use the Pareto front approximation method.

[0005] A small number of people have combined the redistribution problem with other airport scheduling issues and taxiway research. Studies have shown that taxiway allocation and parking stand allocation will conflict with each other. It is concluded that the linkage relationship between various resources must be fully considered to achieve coordinated operation. However, there are few joint studies on parking stand allocation and pushback control. Therefore, it is urgent to introduce a method that can more accurately predict the airport's parking stand allocation to improve airport operation efficiency. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an airport parking space allocation method, device and storage medium in view of the deficiencies in the prior art.

[0007] The technical solution of the present invention to solve the above technical problem is as follows: A method for allocating parking spaces at an airport comprises the following steps:

[0008] S1. Construct a push control model based on N-SCP strategy;

[0009] S2. Simulating and analyzing the push-out control model using a time window simulation tool to obtain a queue length threshold and a push-out time for departing flights;

[0010] S3, constructing a parking space allocation model according to the departure flight queue length threshold and the push-out time;

[0011] S4. Solving the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

[0012] Another technical solution of the present invention to solve the above technical problem is as follows: an airport parking space allocation device, comprising:

[0013] A push control model building module, used to build a push control model based on the N-SCP strategy;

[0014] A simulation analysis module, used to perform simulation analysis on the push-out control model through a time window simulation tool to obtain a departure flight queue length threshold and a push-out time;

[0015] A parking stand allocation model building module, used to build a parking stand allocation model according to the departure flight queue length threshold and the push-out time;

[0016] The solution module is used to solve the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

[0017] Another technical solution of the present invention to solve the above technical problem is as follows: an airport parking stand allocation device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the airport parking stand allocation method as described above is implemented.

[0018] Another technical solution of the present invention to solve the above technical problem is as follows: a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the airport parking space allocation method as described above is implemented.

[0019] The beneficial effects of the present invention are: a push-out control model and a parking stand allocation model are constructed, and the simulation analysis results of the time window are used to jointly optimize the push-out control and parking stand allocation processes, thereby improving the accuracy and adaptability of the model in the time dimension, providing scientific decision-making support for airport operation management, and enabling the parking stand allocation model to comprehensively and accurately reflect the real-time status and complexity of airport operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of an aircraft push-out process of the prior art of the present invention;

[0021] Figure 2 A schematic diagram of a flow chart of an airport parking space allocation method provided by an embodiment of the present invention;

[0022] Figure 3 A module block diagram of an airport parking space allocation device provided by an embodiment of the present invention;

[0023] Figure 4 A graph showing the relationship between the push-off rate and the number of aircraft queuing on the taxiway provided in an embodiment of the present invention;

[0024] Figure 5 A simulation flow chart provided for an embodiment of the present invention;

[0025] Figure 6 A genetic algorithm flow chart provided for an embodiment of the present invention;

[0026] Figure 7 A schematic diagram of algorithm convergence provided by an embodiment of the present invention;

[0027] Figure 8 One of the flight satisfaction graphs provided by the embodiment of the present invention;

[0028] Fig. 9 The second flight satisfaction graph provided by the embodiment of the present invention;

[0029] Fig.10 A schematic diagram of a pre-allocation solution provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0031] like Figure 2 As shown, the present invention provides an airport parking space allocation method, comprising the following steps:

[0032] S1. Construct a push control model based on N-SCP strategy;

[0033] S2. Simulating and analyzing the push-out control model using a time window simulation tool to obtain a queue length threshold and a push-out time for departing flights;

[0034] S3, constructing a parking space allocation model according to the departure flight queue length threshold and the push-out time;

[0035] S4. Solving the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

[0036] In the above embodiment, a push-out control model and a parking stand allocation model are constructed, and the simulation analysis results of the time window are used to jointly optimize the push-out control and parking stand allocation processes, thereby improving the accuracy and adaptability of the model in the time dimension, providing scientific decision-making support for airport operation management, and enabling the parking stand allocation model to comprehensively and accurately reflect the real-time status and complexity of airport operations.

[0037] In some embodiments, the N-SCP strategy dynamically adjusts the takeoff rate based on the current taxiway queue length so that the takeoff rate decreases as the queue length increases. Therefore, this approach effectively optimizes the push-out control model and enhances its adaptability to changes in congestion levels at different airports. The N-SCP strategy combines the N control strategy with a piecewise step linear function. The N-SCP strategy aims to determine the optimal control rate and establish a threshold for control implementation, taking into account specific N value constraints applicable to airports of different sizes. Figure 4As shown, this strategy sets two thresholds, denoted as 'a' and 'b', and defines two takeoff control rates accordingly, resulting in three different control stages. For example, when the number of aircraft in the taxiway queue has not reached the control value a, the aircraft requesting takeoff is allowed to leave with a probability of 1. Once the number of aircraft in the queue exceeds 'a' but is lower than 'b', the probability of the aircraft requesting takeoff being allowed to leave is denoted as α. When the number of aircraft in the taxiway queue exceeds the b value but does not exceed the threshold N, the probability of the aircraft requesting takeoff being allowed to leave is β, where α>β. Once the number of aircraft in the taxiway queue exceeds the threshold N, no other aircraft is allowed to leave. In this embodiment, the process of aircraft waiting at the boarding gate to enter the taxiway is modeled as a queue system Q1 (M / M / 1 / ∞ / ∞), that is, the first queue system Q1, which has no limit on the queue length but a variable service rate. The aircraft arrival rate is represented by λ, and the pushback control is regarded as the waiter I, whose service rate μ is equal to the pushback release rate. The queuing process is not a straight line, but follows the first-come, first-served (FCFS) principle. The arrival rate λ in Q1 is 1. Similarly, the process of aircraft waiting on the taxiway to enter the runway for takeoff is also modeled as a queue system Q2(M / M / 1 / N / ∞), that is, the second queue system Q2, which has a queue length limit and an arrival rate that varies with time. The runway service is regarded as waiter II, the arrival rate λ represents the arrival efficiency of aircraft entering the taxiway, and the service rate μ represents the average takeoff time. Since the pushback control strategy is implemented in Q1, the arrival rate λ in Q2 depends on the queue length of Q2 at the decision time t.

[0038] Based on the above content, preferably, the S1 includes:

[0039] Based on the N-SCP strategy, the process of the aircraft waiting to enter the taxiway at the boarding gate is modeled as the first queue system Q1 (M / M / 1 / ∞ / ∞), and the process of the aircraft waiting to enter the runway for takeoff on the taxiway is modeled as the second queue system Q2 (M / M / 1 / N / ∞);

[0040] Assumptions are made for the push-out control model, including:

[0041] (1) The runway service time of each type of aircraft follows the same distribution;

[0042] (2) The time interval between the aircraft's takeoff request and the time between the aircraft receiving runway service is equal;

[0043] (3) Considering different aircraft categories, the fuel consumption per unit time is calculated as a weighted average;

[0044] (4) The first queue system Q1 and the second queue system Q2 are independent queues and follow the push-out strategy:

[0045]

[0046] Where i is the first push request of the aircraft, a and b are the control implementation thresholds. When the number of aircraft j in the queue system is greater than or equal to a and less than b, the probability that the aircraft requesting take-off is allowed to leave is α. N is the departure flight queue length threshold. When the number of aircraft j in the queue system is greater than or equal to b and less than N, the probability that the aircraft requesting take-off is allowed to leave is β. When the number of aircraft j in the queue system is greater than or equal to N, the probability that the aircraft requesting take-off is allowed to leave is 0.

[0047] (5) Flights arrive according to the Poisson process and enter from the first queue system Q1 (M / M / 1 / ∞ / ∞), and receive services at the first service station of the first queue system Q1 and the second service station of the second queue system Q2 in turn, and finally leave from the second service station. When the capacity M of the second service station reaches the threshold, new flights are no longer allowed to enter the queue;

[0048] (6) When the length of the second queue system Q2 is less than the capacity M, the first service station of the first queue system Q1 is working. When the queue length on the runway reaches the capacity M, the first service station immediately stops working; it instructs the second service station to complete a service and reduce the queue length to K-1, and then the first server resumes working;

[0049] (7) The time intervals between all flight arrivals and the service hours of each service desk are independent of each other;

[0050] (8) Both service times follow a negative exponential distribution;

[0051] The assumed first queue system Q1 and the second queue system Q2 are connected in series to obtain the push-out control model.

[0052] The above embodiment obtains the push-out control model and the corresponding assumptions of the model. Under the premise of the assumptions, the MATLAB software is used to perform numerical simulation on the continuous process of the departure aircraft from push-out-taxiing-takeoff. Secondly, based on the idea of ​​queuing theory in operations research, the taxiing process of the departure flight is regarded as a queuing system, and the process of the flight taxiing to the runway end is regarded as the input process of the queuing system. Existing empirical data show that the distribution of arrivals obeys a Poisson distribution with certain parameters, the arrival mode is a single arrival, the runway service time obeys a negative exponential distribution, and the queuing rule is the first-come, first-served principle (FCFS).

[0053] The time when the first departing aircraft enters the runway end is assigned as t1, at which time the system's queue waiting time and queue length are both 0 (the first departing aircraft does not need to wait or queue). The above process is dynamically updated, combining the expected arrival time, actual arrival time, and block time of the incoming flight with the off-block time, expected take-off time, and actual take-off time of the departing flight.

[0054] In order to determine the optimal operating cost and its corresponding N value, this embodiment proposes an optimization algorithm based on a continuous time Markov chain (CTMC).

[0055] like Figure 5 As shown, based on the above content, preferably, S2 includes:

[0056] S2.1. Sort all aircraft in ascending order according to their take-off time and construct an event matrix;

[0057] S2.2, initializing the parameters of the push-out control model to obtain event cells;

[0058] S2.3. When the aircraft push-out request i=1, update the event application time and push-out time. When i≥2, update the length of the first queue system Q1:

[0059] S2.4, increase the length of the first queue system Q1 by 1;

[0060] Judge the constraint conditions: if the length of the taxiway queue is greater than the taxiway threshold, the request time and service time are summed and the event matrix is ​​updated according to the sum result;

[0061] If the length of the taxiway queue is less than a, update the time to enter the taxiway = the updated delayed request time, and update the time at the parking stand = the time to enter the taxiway - the service time, and update the length of the parking stand virtual queue plus one;

[0062] If a is less than the taxiway length and less than b, set j = 1, update the pushout rate a(j), and update the pushout request time + service time;

[0063] If the taxiway queue length > the departure flight queue length threshold, record the number of aircraft s that have successfully taken off plus 1 and record, update service time = time spent in the taxiway queue, update departure time = time entering the taxiway + time spent in the taxiway queue; update service time = departure time of the previous aircraft - time entering the taxiway plus time spent in the taxiway queue; update departure time = departure time of the previous aircraft + time spent in the taxiway queue;

[0064] S2.5. Determine the calculated Gj value. If the Gj value is greater than the threshold Gmax, mark the N value as infeasible and execute S2.6. If the Gj value is greater than or equal to the threshold Gmax, return to S2.4. The Gj value is the actual queue length of the second queue system Q2.

[0065] S2.6. Calculate and store the total fuel consumption each time the engine is turned on and the aircraft is waiting for takeoff on the taxiway;

[0066] S2.7. Confirm whether all aircraft are simulated. If so, end. If not, return to S2.2.

[0067] The above embodiment obtains the queue length threshold and push-out time of the departing flight, makes assumptions about the parking stand allocation model, and takes into account the size constraints of the parking stands. Therefore, in this model, two scheduling centers are virtualized, representing the set of large parking stands and the set of small parking stands respectively; the expected time window and the acceptable time window of the scheduling demand of each aircraft are known; the estimated arrival time of each aircraft is known; each time the parking stand is provided, one parking stand must and can only serve one aircraft.

[0068] In order to clearly explain how to convert the highest satisfaction in airport operations into the lowest penalty cost through effective management of scheduling time windows, this embodiment introduces a scheduling time window function based on soft time windows to derive a penalty cost function. This function not only reflects the airport's emphasis on the accuracy of aircraft arrival and departure (push-out) times, but also reflects the reward and punishment mechanism for scheduling efficiency and robustness. When aircraft i can complete the service within its most ideal time window, that is, enter and exit the parking space within the predetermined optimal time range, the penalty cost is set to 0. If the actual service completion time of the aircraft falls on the leading edge or trailing edge of the expected time window, that is, slightly earlier or later than the most ideal time period, the function will give a certain penalty cost according to the degree of deviation. The setting of the penalty cost is intended to encourage airlines and airport ground handling teams to perform tasks as punctually as possible, while also leaving a certain amount of flexibility for slight deviations, ensuring the flexibility and robustness of the parking space pre-allocation system.

[0069] In order to convert the highest airport satisfaction into the lowest penalty cost for exceeding the scheduling time window, the scheduling time window function μ is used based on the soft time. i (t i ) to obtain the penalty cost function As shown in the figure. When aircraft i is in the most desired time window [a i ,b i ] to complete the service, the penalty cost is 0; when the aircraft is in E i <t i <a i or b i <ti <L i If the service is completed within 10 seconds, a certain penalty cost will be given; when the time for the dispatched vehicle to arrive at the delivery point i exceeds the acceptable time window [E i ,L i ], a greater penalty will be imposed. i (t i ) can be expressed as:

[0070]

[0071] Among them, a i =T arrive +T taxi-in , b i =T push -T slot , L i =T application +30-T service , E i =T application -T service .

[0072] Next, the intensity function is predicted based on the crowding degree.

[0073] Grey relational analysis (GRA) is an important branch of grey system theory, which is mainly used to evaluate and analyze the degree of correlation between various factors within the system. This analysis method is particularly suitable for situations where system information is incomplete, the amount of data is small, and the uncertainty is high. The basic principle of grey relational analysis is to quantitatively describe the strength of the relationship between them by comparing the similarities between sequences based on the similarity of geometric forms.

[0074] The theoretical basis of grey relational analysis is based on the grey system theory of system analysis, which was proposed by Chinese scholar Professor Deng Julong in the early 1980s. Grey system theory focuses on systems where some information is known and some is unknown, namely grey systems. In such systems, the uncertainty and incompleteness of data make traditional data analysis methods difficult to apply. Grey relational analysis provides a method to analyze the interaction of internal factors of a system under the condition of insufficient information by constructing a similarity measure between sequences.

[0075] 1) Data preprocessing

[0076] The first step of grey relational analysis is data preprocessing, including data standardization or normalization, the purpose of which is to eliminate the differences in data dimensions and magnitudes so that the data can be compared under the same standards. Through preprocessing, the original data can be converted into dimensionless comparison sequences and reference sequences.

[0077] 2) Determine the reference sequence

[0078] In the grey relational analysis, it is necessary to first determine the reference sequence and one or more comparison sequences. The reference sequence is usually the ideal state of the system or the key factor of the study, while the comparison sequence is other factors or states compared with the reference sequence.

[0079] 3) Correlation calculation

[0080] Calculating the correlation is the core step of grey correlation analysis. First, the difference sequence between the reference sequence and each comparison sequence is calculated, and then the correlation coefficient of the difference sequence is determined by a specific calculation formula, such as the minimum difference and the maximum difference. The calculation of the correlation coefficient usually involves a resolution coefficient, which is used to adjust the sensitivity of the correlation.

[0081] 4) Analyze the association results

[0082] Finally, the degree of association between each comparison sequence and the reference sequence is analyzed based on the calculated association coefficient. The larger the association coefficient, the higher the similarity between the comparison sequence and the reference sequence, and the closer the degree of association.

[0083] In order to reflect the dynamic changes of queuing data, the embodiment of the present invention selects the maximum runway queue number and the longest waiting time for parking stands as the comparison sequence, which can intuitively reflect the service pressure faced by airport parking stands and runways during the observation period. It not only pays attention to the changes in the average level, but also can capture the congestion in extreme cases, providing a basis for formulating penalty strategies for dealing with peak hours.

[0084] Next, construct a parking slot allocation penalty function.

[0085] Compared with the traditional penalty function scenario, where the service can be satisfied by starting within a specified time interval, the penalty function pursued by the parking space pre-allocation problem must not only reflect the preference and guiding nature of specific time, but also take into account the different busyness of airport operations. Previous studies usually only draw a distribution map of the expected aircraft arrival time and judge the busyness of the airport based on the arrival frequency in a certain time period. Although this method is intuitive, it ignores the impact of the aircraft's stopover time on the airport's operating pressure. More importantly, it cannot reflect the prediction and consideration of the airport management pressure. In order to more realistically reflect the complexity and dynamic changes of airport operations, we propose a new methodology, hoping to achieve stricter constraints on the time window by imposing a larger penalty during busy hours when parking space resources become more scarce. To this end, we introduce a situation prediction coefficient based on the traditional penalty function.

[0086]

[0087] Arranged:

[0088]

[0089] in, σ2=α n-a ,σ3=β n-b α a ρ a .

[0090] In order to better reflect the scale of congestion, the coefficient needs to be normalized before use in actual calculations:

[0091]

[0092] in, express By using simulation to push out control results, the function can conduct in-depth analysis of the airport's situation at different busy levels and accurately adjust the time window constraint. The core advantage of this method is that it can quantify the impact of the airport situation and apply these quantitative indicators to the pre-allocation problem of parking spaces. This not only enhances the ability to predict airport operation dynamics, but also significantly improves the robustness of the parking space allocation plan.

[0093] Based on the above content, the two objective functions are weighted and added, so as to transform the multi-objective problem into a single-objective problem for solving. Preferably, S3 includes:

[0094] The distance function is:

[0095]

[0096] The penalty function is:

[0097]

[0098] Among them, C k represents the distance a passenger needs to walk to parking position k, y ik indicates that aircraft i is served by parking stand k, x ijg Indicates that the g-th parking stand serves aircraft i to aircraft j, represents the situation prediction coefficient, represents the penalty cost function;

[0099] The distance function and the penalty function are weighted and added together, and the objective function is:

[0100] minZ=αZ1+βZ2,

[0101] The variables are as follows:

[0102]

[0103] A constraint condition is set for the objective function, and a parking space allocation model is obtained based on the objective function with the constraint condition set, wherein the constraint condition is:

[0104]

[0105]

[0106] t ki =max{E i ,t k(i-1) +s k(i-1)}, (8)

[0107] t k(i-1) +s k(i-1) ≤p i +15, (9)

[0108] s ki -s k(i-1) ≥30, (10)

[0109] α+β=1; (11)

[0110] Where L represents the set of large parking spaces, l represents the number of large parking spaces, S represents the set of small parking spaces, s represents the number of small parking spaces, G represents the set of all parking spaces, F represents the set of aircraft, g represents the index of each parking space, i represents the index of each aircraft, and t represents the number of small parking spaces. ki represents the time when aircraft i starts to occupy parking space k, S ki represents the occupancy time of aircraft i at parking position k, P i It indicates the time when aircraft i will apply for pushback, [E i ,L i ] represents the acceptable time window for aircraft i, α and β represent weights, and 15 and 30 are set time parameters, specifically: 15min represents the fluctuation that can be absorbed, and 30min represents the time that is judged as a delay. i ,b i ] represents the expected time window of aircraft i, T represents the airport stopover time, and δ represents the penalty intensity coefficient.

[0111] Constraints (2) and (3) ensure the capacity limits of large and small parking stands, constraint (4) ensures that each aircraft must be assigned to a parking stand, constraints (5) and (6) ensure that each aircraft has only one parking stand to serve, (7) represents the time window constraint, (8) represents the delay constraint, and (9) represents the airport transit time constraint.

[0112] The above embodiment constructs a parking space allocation model and the constraints corresponding to the model. Next, a genetic algorithm is used to solve the parking space allocation model. A genetic algorithm is an optimization algorithm that simulates the natural evolution mechanism. The population represents: a set of solution codes, each code is called an individual or chromosome. Individual / chromosome representation: a specific representation of the problem solution, such as a binary string, a real number sequence, etc.

[0113] like Figure 6 As shown, based on the above content, preferably, S4 includes:

[0114] S4.1. Based on a real number encoding method, each constraint condition in the parking space allocation model is regarded as a value of a gene and mapped to a preset real number, where the real number is represented as an index of the parking space resource;

[0115] S4.2. In the pre-allocation problem, each aircraft has a model constraint and a parking space has a minimum before and after occupancy constraint. The corresponding genes are required to satisfy the relevant constraints in the parking space allocation model. After the i-th gene individual is determined, the last use time and allocation of the parking space are updated, and the process is cyclically processed until a chromosome is generated. The chromosome indicates that all flights are allocated to the parking space.

[0116] S4.3. Based on the objective function of the parking stand allocation model, a pre-allocation objective function is constructed. The pre-allocation objective function is a weighted combination of two indicators: the minimum number of parking stands for departure aircraft of each airline and the shortest total walking time of all passengers. The two indicators are standardized to obtain the fitness value of each gene individual. The fitness value of each gene individual is added to obtain the total fitness of the population.

[0117] S4.4, based on the roulette wheel selection method, select gene individuals from all genes, calculate the ratio between the selected gene individuals and the total fitness of the population, generate a random number based on the ratio result, and if the random number is less than the fitness value of the selected gene, fill the selected gene into the population, and repeat S4.4 to obtain the filled population;

[0118] Specifically, first traverse each individual in the population, calculate and collect their fitness values, and then add up the fitness values ​​of all individuals to get FitnessSum, which is the total fitness of the population. The fitness of individual i is Fitness(i), and the probability of the individual being selected is proportional to the value of the fitness. The specific calculation formula is as follows:

[0119]

[0120] Subsequently, the algorithm generates a random number r to select an individual, and then generates another random number. If this random number is less than the relative fitness of the individual, Fitness(i), the individual is selected to join the new population. This process is repeated until the new population is fully populated.

[0121] S4.5. Take each chromosome as an allocation scheme, randomly select a flight, randomly allocate a new parking space for the flight, and replace the chromosome in the populated population with a new individual generated by crossover and mutation operations to obtain a new population;

[0122] Specifically, in the two-point crossover method and single-point mutation, the algorithm randomly selects a position on the chromosome (i.e., the mutation point), randomly selects a flight in each chromosome, i.e., an allocation scheme, and randomly allocates a new parking space for it, ensuring that the newly allocated parking space is different from the original one. The new individuals generated by crossover and mutation replace the chromosomes in the current population to form a new population.

[0123] S4.6. Determine whether the set number of iterations is met. If so, execute S4.7. If not, return to S4.1. Specifically, the termination condition is determined by setting the number of iterations, which is set to 100 iterations. The algorithm terminates when the maximum number of iterations is reached, and the optimal solution of the model is output.

[0124] S4.7. Output the results and select the chromosome with the highest fitness as the airport parking space allocation result.

[0125] Preferably, the S4.2 includes:

[0126] S4.2.1. Sort the estimated arrival times of the flights, number the flights based on the sorting, start from parking stand j and flight i, and initialize the current last use time, available time and service time of the parking stand;

[0127] S4.2.2. Initially allocate parking spaces. For each gene individual in the population, starting from the first aircraft, allocate a parking space to each flight in order, and check whether the constraints of aircraft type and parking space size are full: the allocation is feasible if and only if the parking space is greater than or equal to the aircraft type. If it is satisfied, keep the initial allocation result. If it is not satisfied, randomly find another parking space and try to allocate again. Repeat this behavior until all parking spaces are allocated.

[0128] S4.2.3. Update the initial allocation result, store the allocated and unallocated sets of flights, update the new free time after each parking stand ends, and update the time when the flight is scheduled to leave the parking stand + the minimum interval time;

[0129] S4.2.4, allocate again, start with selecting flight i from the unallocated set, randomly select a parking space for it, and check all relevant constraints for the current flight information and available parking space information of this allocation: check the aircraft type matching constraint, the condition ensures that the selected parking space can accommodate the aircraft of the flight, that is, the specifications of the parking space must at least meet the minimum demand of the flight aircraft, that is, the allocation is feasible if and only if the parking space is greater than or equal to the aircraft model; then check the parking space time availability constraint: ensure that the parking space is free during the arrival and departure time of the flight, that is, the last use time of the parking space is less than or equal to the estimated arrival time of this flight. If the allocation result violates the constraint, then loop S4.2.4 in this way. If it is confirmed that the parking space meets all constraints, execute S4.2.5;

[0130] S4.2.5. Allocate the flight to the parking stand and update the last use time and allocation of the parking stand. Once it is confirmed that the parking stand meets all constraints, the algorithm will allocate the flight to this parking stand and update the status of this parking stand, that is, the next available time, to ensure that there will be no conflict with the allocation of subsequent flights, update the initial allocation result, and store the allocated and unallocated sets of flights.

[0131] S4.2.6. Determine whether all flights are assigned to parking stands. If so, a chromosome is generated, and the chromosome indicates that all flights are assigned to parking stands. If not, return to S4.2.4.

[0132] Preferably, the S4.3 includes:

[0133] The walking distance walkdist and penalty function Vol of the objective function of the parking space allocation model are calculated respectively, and the walking distance walkdist is:

[0134]

[0135] Among them, i is the index, Np i For the i-th passenger, Dp i Nb is the distance from the exit of the i-th passenger to the luggage pick-up area, i For the i-th baggage passenger, Db i The distance for the i-th passenger to get his luggage out of the station;

[0136] The penalty function Vol is:

[0137]

[0138] in,

[0139]

[0140] The calculation is based on the situation where the distribution arrival time is in different time intervals. If the distribution arrival time is in the optimal arrival time interval, the penalty cost is 0. If it exceeds the optimal arrival time but does not exceed the tolerable delay time, the penalty cost increases according to the proportion of the arrival time delay. If it exceeds the tolerable delay time, the penalty function reaches its maximum.

[0141] In order to convert different evaluation indicators (such as walking distance, satisfaction and penalty items) into comparable magnitudes, the dimensions of the walking distance walkdist and the penalty function Vol are adjusted in the fitness function fitness. The fitness function fitness is:

[0142]

[0143] Among them, ω1 and ω2 represent the weights of the walking distance walkdist and the penalty function Vol respectively.

[0144] like Figure 3 As shown, an embodiment of the present invention further provides an airport parking space allocation device, comprising:

[0145] A push control model building module, used to build a push control model based on the N-SCP strategy;

[0146] A simulation analysis module, used to perform simulation analysis on the push-out control model through a time window simulation tool to obtain a departure flight queue length threshold and a push-out time;

[0147] A parking stand allocation model building module, used to build a parking stand allocation model according to the departure flight queue length threshold and the push-out time;

[0148] The solution module is used to solve the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

[0149] Preferably, the push control model construction module is used to construct the push control model based on the N-SCP strategy, including:

[0150] Based on the N-SCP strategy, the process of the aircraft waiting to enter the taxiway at the boarding gate is modeled as the first queue system Q1 (M / M / 1 / ∞ / ∞), and the process of the aircraft waiting to enter the runway for takeoff on the taxiway is modeled as the second queue system Q2 (M / M / 1 / N / ∞);

[0151] Assumptions are made for the push-out control model, including:

[0152] (1) The runway service time of each type of aircraft follows the same distribution;

[0153] (2) The time interval between the aircraft's takeoff request and the time between the aircraft receiving runway service is equal;

[0154] (3) Considering different aircraft categories, the fuel consumption per unit time is calculated as a weighted average;

[0155] (4) The first queue system Q1 and the second queue system Q2 are independent queues and follow the push-out strategy:

[0156]

[0157] Where i is the first push request of the aircraft, a and b are the control implementation thresholds. When the number of aircraft j in the queue system is greater than or equal to a and less than b, the probability that the aircraft requesting take-off is allowed to leave is α. N is the departure flight queue length threshold. When the number of aircraft j in the queue system is greater than or equal to b and less than N, the probability that the aircraft requesting take-off is allowed to leave is β. When the number of aircraft j in the queue system is greater than or equal to N, the probability that the aircraft requesting take-off is allowed to leave is 0.

[0158] (5) Flights arrive according to the Poisson process and enter from the first queue system Q1 (M / M / 1 / ∞ / ∞), and receive services at the first service station of the first queue system Q1 and the second service station of the second queue system Q2 in turn, and finally leave from the second service station. When the capacity M of the second service station reaches the threshold, new flights are no longer allowed to enter the queue;

[0159] (6) When the length of the second queue system Q2 is less than the capacity M, the first service station of the first queue system Q1 is working. When the queue length on the runway reaches the capacity M, the first service station immediately stops working; it instructs the second service station to complete a service and reduce the queue length to K-1, and then the first server resumes working;

[0160] (7) The time intervals between all flight arrivals and the service hours of each service desk are independent of each other;

[0161] (8) Both service times follow a negative exponential distribution;

[0162] The assumed first queue system Q1 and second queue system Q2 are connected in series to obtain the push-out control model.

[0163] The above method and device are introduced below through a specific example.

[0164] A simulation analysis was conducted on the push-out process of a total of 549 aircraft from 6:00 to 23:00. According to the calculation results given by the airport, the approach taxiing time of the aircraft at a certain airport was 11.9 minutes, and the departure taxiing time was 9.7 minutes. Combined with the KDE method, the service time before the aircraft submitted an application can be obtained. According to the time window calculation method, the milestone time node of each aircraft can be obtained.

[0165] As shown in Table 1, Table 1 is a table showing some of the results of the N-SCP simulation. Based on the simulation prediction of the launch activity in the previous section, the first 15 aircraft are selected from the prediction results for display.

[0166] Table 1

[0167]

[0168] In order to better meet the needs of different airports, the algorithm parameters need to be set before calculation. The specific settings are as follows:

[0169] Maximum number of iterations (Maxgen=100): Set to stop after 100 generations (iterations), limiting the running time of the algorithm and preventing infinite loops.

[0170] Population size (Y=100): The number of solutions that exist simultaneously, set to 100.

[0171] Crossover probability (croPos=0.7): The probability of whether a chromosome performs a crossover operation, which is set to 0.7 here. A high crossover probability encourages the exchange of information (genes) in the population and helps generate new solutions.

[0172] Mutation probability (mutPos = 0.8): The probability of each gene mutating is set to 0.8. Mutation introduces new genetic diversity, which is an important mechanism for genetic algorithms to avoid premature convergence.

[0173] Weight (w1=0.2, w2=0.8): Used to balance the importance of different objectives in the fitness function. Here, w1 is set to 0.2 and w2 is set to 0.8.

[0174] Next, after rounding the input data, MATLAB is used to implement the design of the pre-allocation solution algorithm, so as to calculate the example. In order to reflect the effectiveness of the model, a total of 40 aircraft with an estimated arrival time of 7:30 to 10:00 were selected as data input. The calculation results are shown in the table. The satisfaction rate is 13.254, and the total target converges to 0.25. The convergence curve is shown in the figure below. Figure 7 shown.

[0175] Table 2 is the aircraft allocation plan table.

[0176] Table 2

[0177]

[0178]

[0179] like Figure 8-Figure 9 As shown in the figure, in terms of satisfaction, the satisfaction of aircraft No. 2, No. 21 and No. 35 is 0, but the service time of No. 2 is just the left boundary of the acceptable service time window, while aircraft No. 21 and No. 35 are only 1 minute and 2 minutes later than the acceptable time window. The estimated arrival times of the three aircraft are 468, 515 and 572. The allocation plan requires aircraft No. 2 to arrive 6 minutes later, and the allocation results require aircraft No. 21 and No. 35 to circle in the air for 23 minutes and 16 minutes respectively. There is no airport constraint on arrival delay in the algorithm. Although aircraft No. 21 needs to circle in the air for a long time, no aircraft exceeds the airport's maximum tolerance for delay of 30 minutes, which shows the feasibility of the pre-allocation model.

[0180] In order to show the effectiveness of the solution, this example sets a situation where the time window is not considered. The results are as follows Fig.10 As shown in the figure: the satisfaction of the aircraft assigned for the first time is higher, but the satisfaction of the aircraft assigned to the same position for the second time is lower, and even the satisfaction is 0. The comparison results show that the model considering the time window has higher satisfaction, that is, the scheme has better robustness, which also indirectly illustrates the scientific nature of the time window setting in the present invention.

[0181] The average service time of 30 minutes is taken to estimate the parking stand capacity. Theoretically, 12 parking stands can serve 48 aircraft within 180 minutes. Due to the setting of time windows and the reasonable allocation of different service times, the 12 parking stands can serve 40 aircraft. This is mainly because the parking stand resources are used reasonably. The reason for the poor utilization of parking stands before 8 o'clock is mainly because there are only 4 aircraft expected to arrive before 8 o'clock in the morning, and 3 of them are very close to 8 o'clock.

[0182] The present invention innovatively constructs a model framework that jointly optimizes pushback control and parking stand allocation. Based on the N-SCP simulation results, including scene situation indicators such as taxiway queue length, predictions are made for key milestone time nodes of flights such as pushback application time and pushback time, providing a basis for the calculation of the time window and situation penalty coefficient of the parking stand pre-allocation model based on the N-SCP pushback strategy, improving the accuracy and adaptability of the model in the time dimension, and providing scientific decision-making support for airport operation management. The pre-allocation model can fully and accurately reflect the real-time status and complexity of airport operations, thereby improving the practicality and flexibility of the model.

[0183] An embodiment of the present invention also provides an airport parking stand allocation device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the airport parking stand allocation method described above is implemented.

[0184] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the airport parking space allocation method as described above is implemented.

[0185] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0186] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0187] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0188] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0189] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0190] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. For such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0191] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for allocating parking spaces at an airport, characterized in that: The steps include: S1. Construct a push control model based on N-SCP strategy; S2. Simulating and analyzing the push-out control model using a time window simulation tool to obtain a queue length threshold and a push-out time for departing flights; S3, constructing a parking space allocation model according to the departure flight queue length threshold and the push-out time; S4. Solving the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

2. The airport parking space allocation method according to claim 1, characterized in that: The S1 includes: Based on the N-SCP strategy, the process of the aircraft waiting to enter the taxiway at the boarding gate is modeled as the first queue system Q1 (M / M / 1 / ∞ / ∞), and the process of the aircraft waiting to enter the runway for takeoff on the taxiway is modeled as the second queue system Q2 (M / M / 1 / N / ∞); Assumptions are made for the push-out control model, including: (1) The runway service time of each type of aircraft follows the same distribution; (2) The time interval between the aircraft's takeoff request and the time between the aircraft receiving runway service is equal; (3) Considering different aircraft categories, the fuel consumption per unit time is calculated as a weighted average; (4) The first queue system Q1 and the second queue system Q2 are independent queues and follow the push-out strategy: Where i is the first push request of the aircraft, a and b are the control implementation thresholds. When the number of aircraft j in the queue system is greater than or equal to a and less than b, the probability that the aircraft requesting take-off is allowed to leave is α. N is the departure flight queue length threshold. When the number of aircraft j in the queue system is greater than or equal to b and less than N, the probability that the aircraft requesting take-off is allowed to leave is β. When the number of aircraft j in the queue system is greater than or equal to N, the probability that the aircraft requesting take-off is allowed to leave is 0. (5) Flights arrive according to the Poisson process and enter from the first queue system Q1 (M / M / 1 / ∞ / ∞), and receive services at the first service station of the first queue system Q1 and the second service station of the second queue system Q2 in turn, and finally leave from the second service station. When the capacity M of the second service station reaches the threshold, new flights are no longer allowed to enter the queue; (6) When the length of the second queue system Q2 is less than the capacity M, the first service station of the first queue system Q1 is working. When the queue length on the runway reaches the capacity M, the first service station immediately stops working; (7) The time intervals between all flight arrivals and the service hours of each service desk are independent of each other; (8) Both service times follow a negative exponential distribution; The assumed first queue system Q1 and the second queue system Q2 are connected in series to obtain the push-out control model.

3. The airport parking space allocation method according to claim 2, characterized in that: The S2 includes: S2.

1. Sort all aircraft in ascending order according to their take-off time and construct an event matrix; S2.2, initializing the parameters of the push-out control model; S2.

3. When the number of aircraft push-out requests i = 1, update the event application time and push-out time. When i ≥ 2, update the length of the first queue system Q1: S2.4, increase the length of the first queue system Q1 by 1; The constraint conditions are judged: if the length of the first queue system Q1 is greater than the taxiway threshold, the request time and the service time are summed, and the event matrix is ​​updated based on the sum result; If the length of the taxiway queue is less than a, update the time to enter the taxiway = the updated delayed request time, and update the time at the parking stand = the time to enter the taxiway - the service time, and increase the length of the first queue system Q1 once; If a<the length of the taxiway<b, let j=1, update the push-out rate a(j), and update the push-out request time + service time; If the taxiway queue length is greater than the departure flight queue length threshold, the number of aircraft s that have successfully taken off is increased by 1 and recorded, and the updated service time = the time spent in the taxiway queue, the updated departure time = the time of entering the taxiway + the time spent in the taxiway queue; the updated service time = the departure time of the previous aircraft - the time of entering the taxiway plus the time spent in the taxiway queue; the updated departure time = the departure time of the previous aircraft + the time spent in the taxiway queue; S2.

5. Determine the calculated Gj value. If the Gj value is greater than the threshold Gmax, mark the N value as infeasible and execute S2.

6. If the Gj value is greater than or equal to the threshold Gmax, return to S2.

4. The Gj value is the actual queue length of the second queue system Q2. S2.

6. Calculate and store the total fuel consumption each time the engine is turned on and the aircraft is waiting for takeoff on the taxiway; S2.

7. Confirm whether all aircraft are simulated. If so, end. If not, return to S2.

2.

4. The method for allocating airport parking spaces according to claim 1, characterized in that: The S3 includes: A distance function and a penalty function are constructed based on the N-SCP strategy. The distance function is: The penalty function is: Among them, C k represents the distance a passenger needs to walk to parking position k, y ik indicates that aircraft i is served by parking stand k, x ijg Indicates that the g-th parking stand serves aircraft i to aircraft j, represents the situation prediction coefficient, represents the penalty cost function; The distance function and the penalty function are weighted and added together, and the objective function is: minZ=αZ1+βZ2, Wherein, a constraint condition is set for the objective function, and a parking space allocation model is obtained based on the objective function with the constraint condition set, wherein the constraint condition is: t ki =max{E i ,t k(i-1) +s k(i-1) }, t k(i-1) +s k(i-1) ≤p i +15, s ki -s k(i-1) ≥30, α+β＝1； Where L represents the set of large parking spaces, l represents the number of large parking spaces, S represents the set of small parking spaces, s represents the number of small parking spaces, G represents the set of all parking spaces, F represents the set of aircraft, g represents the index of each parking space, i represents the index of each aircraft, and t represents the number of small parking spaces. ki represents the time when aircraft i starts to occupy parking space k, S ki represents the occupancy time of aircraft i at parking position k, P i It indicates the time when aircraft i will apply for pushback, [E i ,L i ] represents the acceptable time window for aircraft i, α and β represent weights, and 15 and 30 are the set time parameters.

5. The method for allocating airport parking spaces according to claim 1, characterized in that: The S4 includes: S4.

1. Based on a real number encoding method, each constraint condition in the parking space allocation model is regarded as a value of a gene and mapped to a preset real number, where the real number is represented as an index of the parking space resource; S4.

2. In the pre-allocation problem, each aircraft has a model constraint and a parking space has a minimum before and after occupancy constraint. The corresponding genes are required to satisfy the relevant constraints in the parking space allocation model. After the i-th gene individual is determined, the last use time and allocation of the parking space are updated, and the process is cyclically processed until a chromosome is generated. The chromosome indicates that all flights are allocated to the parking space. S4.

3. Based on the objective function of the parking stand allocation model, a pre-allocation objective function is constructed. The pre-allocation objective function is a weighted combination of two indicators: the minimum number of parking stands for departure aircraft of each airline and the shortest total walking time of all passengers. The two indicators are standardized to obtain the fitness value of each gene individual. The fitness value of each gene individual is added to obtain the total fitness of the population. S4.4, based on the roulette wheel selection method, select gene individuals from all genes, calculate the ratio between the selected gene individuals and the total fitness of the population, generate a random number based on the ratio result, and if the random number is less than the fitness value of the selected gene, fill the selected gene into the population, and repeat S4.4 to obtain the filled population; S4.

5. Take each chromosome as an allocation scheme, randomly select a flight, randomly allocate a new parking space for the flight, and replace the chromosome in the populated population with a new individual generated by crossover and mutation operations to obtain a new population; S4.6, determine whether the set number of iterations is met, if so, execute S5.7, if not, return to S4.1; S4.

7. Output the results and select the chromosome with the highest fitness as the airport parking space allocation result.

6. The method for allocating airport parking spaces according to claim 1, characterized in that: The S4.2 includes: S4.2.

1. Sort the estimated arrival times of the flights, number the flights based on the sorting, start from parking stand j and flight i, and initialize the current last use time, available time and service time of the parking stand; S4.2.

2. Initially allocate parking spaces. For each gene individual in the population, starting from the first aircraft, allocate a parking space to each flight in order, and check whether the constraints of aircraft type and parking space size are full: the allocation is feasible if and only if the parking space is greater than or equal to the aircraft type. If it is satisfied, keep the initial allocation result. If it is not satisfied, randomly find another parking space and try to allocate again. Repeat this behavior until all parking spaces are allocated. S4.2.

3. Update the initial allocation results, and update the new free time after each parking stand ends, as well as the time when the flight is scheduled to leave the parking stand + the minimum interval time; S4.2.4, allocate again, start with selecting flight i from the unallocated set, randomly select a parking space for it, and check all relevant constraints for the current flight information and available parking space information allocated this time: Check the aircraft type matching constraint: the condition ensures that the selected parking space can accommodate the aircraft of the flight, that is, the specifications of the parking space must at least meet the minimum demand of the flight aircraft, that is, the allocation is feasible if and only if the parking space is greater than or equal to the aircraft model; then check the parking space time availability constraint: ensure that the parking space is free during the arrival and departure time of the flight, that is, the last use time of the parking space is less than or equal to the estimated arrival time of this flight. If the allocation result violates the constraint, then loop S4.2.4 in this way. If it is confirmed that the parking space meets all constraints, execute S4.2.5; S4.2.

5. Allocate flights to parking stands and update the last usage time and allocation status of the parking stands; S4.2.

6. Determine whether all flights are assigned to parking stands. If so, a chromosome is generated, and the chromosome indicates that all flights are assigned to parking stands. If not, return to S4.2.

4.

7. The method for allocating airport parking spaces according to claim 1, characterized in that: The S4.3 includes: The walking distance walkdist and penalty function Vol of the objective function of the parking space allocation model are calculated respectively, and the walking distance walkdist is: Among them, i is the index, Np i For the i-th passenger, Dp i Nb is the distance from the exit of the i-th passenger to the luggage pick-up area, i For the i-th baggage passenger, Db i The distance for the i-th passenger to get his luggage out of the station; The penalty function Vol is: in, In order to convert different evaluation indicators into comparable magnitudes, the walking distance walkdist and the penalty function Vol are dimensionalized in the fitness function fitness, and the fitness function fitness is: Among them, ω1 and ω2 represent the weights of the walking distance walkdist and the penalty function Vol respectively.

8. An airport parking space allocation device, characterized in that: include: A push control model building module, used to build a push control model based on the N-SCP strategy; A simulation analysis module, used to perform simulation analysis on the push-out control model through a time window simulation tool to obtain a departure flight queue length threshold and a push-out time; A parking stand allocation model building module, used to build a parking stand allocation model according to the departure flight queue length threshold and the push-out time; The solution module is used to solve the parking stand allocation model based on a genetic algorithm to obtain an airport parking stand allocation result.

9. An airport parking stand allocation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the airport parking stand allocation method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for allocating airport parking spaces according to any one of claims 1 to 7 is implemented.

Images