A base airline and alliance cooperative airline allocation method, device and medium

Abstract

This invention discloses a method, equipment, and medium for airline allocation in collaboration between base airlines and alliances, relating to the field of airport operation planning technology. By predicting the alliance's capacity demand within a preset future timeframe, and integrating base airline capacity growth forecasts, alliance collaboration needs, and terminal capacity constraints, a multi-objective optimization model is constructed. This model achieves precise matching and dynamic adjustment between airlines and terminals, strengthens the adaptability constraint between the alliance's future total capacity and the terminal's design capacity, ensures that the alliance's future total capacity in the current allocation does not exceed the terminal's design capacity, and provides quantitative capacity recommendations for future terminal planning. It solves the problems of traditional methods neglecting alliance collaboration, capacity adaptability, and planning foresight, reduces cross-terminal transfer costs, improves airport operational efficiency and passenger experience, and provides a scientific basis for future terminal planning. It is applicable to the long-term planning and daily scheduling of large multi-terminal hub airports.

Description

Technical Field

[0001] This invention relates to the field of airport operation planning technology, specifically to a method, equipment, and medium for airline allocation in collaboration between base airlines and alliances. Background Technology

[0002] With the rapid development of the air transport industry, multi-terminal layouts at large hub airports have become the mainstream model for meeting capacity growth. Airline allocation, as a core element of airport operation planning, directly impacts transfer efficiency, resource utilization, and passenger experience. Base airlines and their affiliated airline alliances have strong route network synergy, and traditional allocation methods have four major drawbacks:

[0003] First, the long-term capacity growth of base airlines and alliances was not fully predicted, resulting in a lack of foresight in the allocation plan and insufficient terminal capacity adaptability. Second, the "matching relationship between the future total capacity of the alliance and the design capacity of the terminal" was ignored, which could easily lead to terminal overload after the alliance's development. Third, the demand for clustered operations of airlines within the same alliance was ignored, and the same airlines in the same alliance were dispersed to different terminals, resulting in a large number of cross-terminal transfers. Fourth, the focus was on a single objective, and the balanced optimization of multiple objectives such as cross-airline transfer volume, resource utilization rate and capacity adaptability was not achieved.

[0004] Therefore, there is an urgent need for an airline allocation method that takes into account capacity forecasting, alliance coordination, capacity adaptation, and multi-objective optimization, while also providing support for future terminal planning. Summary of the Invention

[0005] The technical problem this invention aims to solve is that traditional methods lack foresight, neglect alliance collaboration, and have insufficient capacity adaptability. The goal is to provide a method, equipment, and medium for allocating airlines between base airlines and alliances. By predicting the alliance's capacity demand within a predetermined future timeframe, and integrating base airline capacity growth forecasts, alliance collaboration needs, and terminal capacity constraints, a multi-objective optimization model is constructed. This model achieves precise matching and dynamic adjustment between airlines and terminals, strengthens the adaptability constraint between the alliance's future total capacity and the terminal's design capacity, ensures that the alliance's future total capacity in the current allocation does not exceed the terminal's design capacity, and provides quantitative capacity recommendations for future terminal planning. This solves the problems of traditional methods neglecting alliance collaboration, capacity adaptability, and planning foresight, reduces cross-terminal transfer costs, improves airport operational efficiency and passenger experience, and provides a scientific basis for future terminal planning. It is applicable to the long-term planning and daily scheduling of large multi-terminal hub airports.

[0006] This invention is achieved through the following technical solution:

[0007] The first aspect of this invention provides a method for airline allocation in collaboration between a base airline and an alliance, comprising the following specific steps:

[0008] Collect airport operation data, construct a capacity demand forecasting model, and obtain the alliance's capacity demand forecast results within a preset future time period;

[0009] Establish a multi-dimensional constraint system;

[0010] With the goals of minimizing cross-airline transfer volume, maximizing terminal resource utilization, minimizing base airline operating costs, and optimizing passenger satisfaction, the model implicitly aims to maximize the compatibility between the alliance's future capacity and the terminal's designed capacity. It combines the alliance's capacity demand forecasts for the future and a multi-dimensional constraint system to construct a multi-objective optimization model.

[0011] A hybrid intelligent algorithm is used to solve the multi-objective optimization model and obtain the optimal allocation scheme;

[0012] Establish a real-time data monitoring mechanism to collect actual flight operation data, terminal resource usage status, dynamic changes in alliance cooperation data, and actual growth data of alliance capacity. When the deviation between the actual alliance capacity and the predicted value exceeds a preset threshold, trigger the multi-objective optimization model to be re-solved and dynamically adjust the allocation scheme.

[0013] Furthermore, the airport operation data includes base airline operation data, airline alliance related data, terminal resource data, and external influencing factor data;

[0014] The base airline's operational data includes route network, flight takeoffs and landings, passenger throughput, transfer ratio, and capacity growth trend;

[0015] The airline alliance-related data includes the affiliation of alliance airlines, the connection relationship of routes within the alliance, the frequency of cross-alliance cooperation, and alliance transit priority agreements;

[0016] The terminal resource data includes the building area of ​​each terminal, the number and type of boarding gates, the capacity of security checkpoints, the capacity of baggage handling systems, the design capacity and peak capacity.

[0017] The data on external influencing factors include regional economic data, tourism market demand, and policy regulation information.

[0018] Furthermore, the construction of the capacity demand forecasting model to obtain the alliance's capacity demand forecast results within a preset future time period includes:

[0019] The capacity demand forecasting model includes the ARIMA time series model and the gradient boosting tree model;

[0020] The ARIMA time series model is used to capture the time series characteristics of the capacity demand of individual airlines, and the preliminary future capacity prediction values ​​of each airline's ARIMA time series model are obtained.

[0021] Gradient boosting tree model is used to explore the nonlinear relationship between external influencing factors and the capacity demand of individual airlines, and to obtain the preliminary future capacity prediction values ​​of each airline's gradient boosting tree model.

[0022] The preliminary future capacity predictions of each airline are obtained by weighted fusion of the preliminary future capacity predictions of the ARIMA time series model and the gradient boosting tree model. The weighting coefficients of the weighted fusion are adjusted in reverse iteration based on the prediction error of each model in the historical prediction period. The model with the prediction error at a first set threshold receives the first weight, and the model with the prediction error at a second set threshold receives the second weight. The first set threshold is less than the set threshold, and the first weight is greater than the second weight.

[0023] Based on the collaborative strategy in the alliance-related data, a collaborative coefficient is introduced to correct the preliminary future capacity forecast of each airline, so as to obtain the future capacity forecast of a single airline.

[0024] The alliance's capacity demand forecast for the future is obtained by summing the future capacity forecasts of all individual airlines within the alliance.

[0025] Furthermore, the predicted capacity demand of the alliance within the preset future timeframe includes:

[0026] The forecast values ​​for the number of takeoffs and landings and passenger throughput of base airlines within a predetermined time period, as well as the forecast values ​​for the total future capacity of airline alliances obtained by summing up the forecast results of individual airlines.

[0027] Furthermore, the multi-dimensional constraint system includes: alliance future capacity-terminal design capacity adaptation constraint, same-alliance aggregation constraint, terminal capacity constraint, and operational efficiency constraint;

[0028] The constraint on the matching of future capacity of the alliance with the design capacity of the terminal is as follows: when an airline alliance is assigned to a certain terminal, its total future capacity prediction value shall not exceed the design capacity of the terminal. That is, for any alliance i and the assigned terminal j, the total future capacity prediction value of alliance i shall not be greater than the design capacity of terminal j.

[0029] The same alliance aggregation constraint means that base airlines and partner airlines of the same airline alliance are preferentially assigned to the same or adjacent terminals, and the proportion of the number of alliance airlines across terminals does not exceed a preset threshold.

[0030] The terminal capacity constraint is that the resource occupancy of each terminal shall not exceed the design peak capacity.

[0031] The operational efficiency constraint is that the average turnaround time and transfer time of airline flights do not exceed the industry standard threshold.

[0032] Furthermore, the mathematical expression of the multi-objective optimization model is:

[0033] ;

[0034] in, To optimize the objective, For cross-airline transfer volume, For resource utilization, For operating costs, For passenger satisfaction, Penalties for violations related to the alliance's future capacity and terminal design capacity, when hour , Let K be the projected future capacity of airline k in alliance i; otherwise... It is the square of the overcapacity ratio. For cross-airline transfer volume weighting coefficients, This is the resource utilization rate weighting coefficient. This is the weighting factor for operating costs. This is a weighting coefficient for passenger satisfaction. The weighting coefficient for the penalty items for capacity violations. .

[0035] Furthermore, the method of using a hybrid intelligent algorithm to solve the multi-objective optimization model specifically includes:

[0036] The multi-objective optimization model is solved using a hybrid optimization algorithm of simulated annealing and adaptive particle swarm optimization. The particle swarm optimization algorithm is used for global optimization, while the simulated annealing algorithm is used to escape local optima. The algorithm parameters are dynamically adjusted through an adaptive mechanism. The solution process includes:

[0037] Chromosomes are constructed based on real-number encoding, and the gene loci of the chromosomes include decision variables such as the proportion of flights allocated by airlines in the terminal and / or the design capacity of the terminal.

[0038] The constraints are transformed into penalty terms in the form of penalty functions and incorporated into the fitness function;

[0039] The fitness of each chromosome is evaluated based on the fitness function of the fusion penalty term;

[0040] The population is iteratively evolved based on fitness to obtain the optimal allocation scheme between airlines and terminals.

[0041] Furthermore, after obtaining the optimal airline-terminal allocation plan, it also includes:

[0042] Determine whether it is a future terminal planning scenario. If so, output the suggested design capacity of each planned terminal simultaneously. The suggested design capacity is dynamically adjusted according to the alliance's capacity growth.

[0043] A second aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for coordinating airline allocation between a base airline and an alliance.

[0044] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for coordinating airline allocation between a base airline and an alliance.

[0045] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0046] Strengthen the compatibility constraints between the alliance's future capacity and the terminal's design capacity to avoid terminal overload problems caused by the alliance's development from the source;

[0047] It achieves the dual functions of operational allocation and planning recommendations, solving the current airline allocation problem while providing quantitative capacity recommendations for future terminal planning, reducing the blind spots in planning, and lowering the cost of later renovations;

[0048] Introducing collaborative forecasting of total alliance capacity avoids the cumulative effect of forecasting bias from a single airline, thereby improving the accuracy of alliance capacity forecasting and providing reliable data support for capacity adaptation constraints.

[0049] By combining clustering constraints and capacity constraints, we can reduce the volume of transfers between terminals while improving resource utilization, thus achieving a balance between operational efficiency and forward-looking planning. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0051] Figure 1 This is a flowchart illustrating the allocation method between base airlines and alliance-cooperative airlines in an embodiment of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0053] As one possible implementation method, such as Figure 1As shown, this embodiment provides a method for allocating airlines between base airlines and alliances, including the following specific steps:

[0054] Step 1: Data Acquisition Phase;

[0055] Five core data categories are collected: base airline operational data (route layout, takeoffs and landings, passenger throughput, capacity growth rate, etc. over the past 5 years) to analyze the operational characteristics and growth patterns of individual airlines; airline alliance correlation data (alliance airline affiliation, route connections within the alliance, alliance capacity synergy strategies, etc.) to provide a basis for predicting the total alliance capacity and clustering constraints; terminal resource data (existing terminal building area, number of gates, design capacity, peak capacity, etc., and for planned terminals, basic information such as planning area and expected service life needs to be collected) to support the construction of capacity constraints; external influencing factor data (regional GDP growth rate, tourist traffic, policy support, etc.) to improve the accuracy of the prediction model; and historical capacity synergy data of the alliance (correlation of capacity growth among airlines within the alliance, overall route expansion plans of the alliance, etc.) to ensure the accuracy of the total alliance capacity prediction.

[0056] Step 2: Base airline and alliance capacity forecasting phase;

[0057] Based on the collected airport operation data, a capacity demand forecasting model is constructed to obtain the capacity demand forecast results of the alliance within a preset time period. This embodiment adopts a two-level forecasting architecture of single-airline forecasting and alliance collaborative correction. The capacity demand forecasting model includes an ARIMA time series model and a gradient boosting tree model.

[0058] First, the ARIMA time series model is used to capture the time series characteristics of capacity demand for individual airlines. Then, the gradient boosting tree model is used to explore the nonlinear relationship between external influencing factors and capacity demand for individual airlines, thereby obtaining preliminary future capacity forecasts for each airline. Specifically, the process involves using an ARIMA time-series model to capture the time-series characteristics of capacity demand for individual airlines, obtaining preliminary future capacity forecasts for each airline's ARIMA time-series model; using a gradient boosting tree model to explore the nonlinear relationship between external influencing factors and capacity demand for individual airlines, obtaining preliminary future capacity forecasts for each airline's gradient boosting tree model; and then combining the preliminary future capacity forecasts from the ARIMA time-series model and the gradient boosting tree model through weighted fusion to obtain the preliminary future capacity forecast for each airline. The weighting coefficients for the weighted fusion are based on the prediction errors of each model within the historical prediction period. A reverse iterative adjustment is performed, with the prediction error at a first set threshold receiving the first weight, and the prediction error at a second set threshold receiving the second weight. If the first threshold is less than the set threshold, the first weight is greater than the second weight. The final future capacity predictions for individual airlines from the ARIMA time-series model and the future total capacity predictions for the alliance from the gradient boosting tree model are then integrated through weighted fusion to obtain the alliance's capacity demand prediction results for a predetermined future timeframe. The weighting coefficients are adjusted iteratively based on the mean squared error (MSE) to ensure that the prediction deviation for individual airlines is controlled within 5%, and the prediction deviation for the total alliance capacity is controlled within 8%. The system outputs the future capacity predictions for base airlines and the alliance for the next 3-5 years (based on existing terminal allocation) or 5-10 years (based on future terminal planning).

[0059] Secondly, based on the collaborative strategies in the alliance's associated data, a collaborative coefficient is introduced. (Reflecting the degree of capacity synergy between airline k and other airlines in alliance i, with a value of 0.8-1.2), the predicted value for a single airline is corrected to obtain the final predicted future capacity value for a single airline. Finally, for all airlines within the alliance... Sum k to obtain the predicted future total capacity of alliance i. The alliance's capacity demand forecast for the future within a predetermined timeframe includes: the predicted capacity demand of individual airlines, such as the number of flight takeoffs and landings and passenger throughput, as well as the alliance's capacity demand forecast for the future within a predetermined timeframe, which is a summary of the predicted capacity demand of individual airlines.

[0060] Step 3: Constraint Construction Phase;

[0061] A multi-dimensional constraint system is constructed, with the constraint of matching the alliance's future capacity with the terminal's design capacity as the core constraint:

[0062] S31. Alliance Future Capacity - Terminal Design Capacity Adaptation Constraint: This is the core constraint of the method in this embodiment, requiring that the total future capacity forecast of an airline alliance allocated to a certain terminal must not exceed the design capacity of that terminal. The mathematical expression for the alliance future capacity - terminal design capacity adaptation constraint is: For any , ,like (i.e., airline k in alliance i is assigned to terminal j), then ,in The future capacity forecast for airline k in Alliance i (from the previous forecast model). This represents the projected total future capacity of Alliance i. The design capacity of terminal building j.

[0063] For existing terminals, strict requirements must be met. ( For the future total capacity of the alliance, (Design capacity for terminal j); for future planned terminals, this constraint can be applied in reverse—based on the projected total future capacity of the target service alliance. Combined with a certain capacity redundancy factor (usually 1.1-1.2 to cope with sudden growth), the recommended design capacity of the planned terminal is output. If multiple alliances are assigned to the same terminal, the sum of the future total capacity of all alliances must be less than or equal to the terminal's design capacity. (J is the set of alliances assigned to terminal j).

[0064] S32. Same-coalition clustering constraint. The mathematical expression for the same-coalition clustering constraint is:

[0065] ;

[0066] Where A represents the set of airline alliances, and T represents the set of terminals. This represents the state variable where airline k in alliance i is assigned to terminal j (1 for assigned, 0 for unassigned). The threshold for the proportion of airlines from different terminal alliances (range 0.05-0.15).

[0067] Priority will be given to assigning base airlines and partner airlines of the same alliance to the same or adjacent terminals, with the number of alliance airlines operating across terminals not exceeding 10%, thereby reducing inter-terminal transfers within the alliance.

[0068] S33. Terminal capacity constraints: Limit the utilization rate of boarding gates and the load of security checkpoints in each terminal to no more than 85% of the design peak capacity, forming a double guarantee with the alliance capacity constraints.

[0069] S34. Operational efficiency constraints: The average turnaround time for flights is required to be ≤45 minutes, the transfer time within the same terminal is required to be ≤60 minutes, and the transfer time across terminals is required to be ≤120 minutes (if there is cross-terminal allocation).

[0070] S35. Policy Compliance Constraints: Complies with the "Regulations on the Operation and Management of Civil Airports", the airline alliance cooperation agreement and the airport land use planning requirements.

[0071] Step 4: Multi-objective optimization model construction stage;

[0072] The core objective is to minimize cross-airline transfer volume (weight 0.3), while also considering maximizing resource utilization (weight 0.25), minimizing base airline operating costs (weight 0.15), optimizing passenger satisfaction (weight 0.1), and maximizing alliance-terminal capacity fit (weight 0.2). The weights are determined using an analytic hierarchy process combined with entropy weighting. The weight of capacity fit is determined by expert scoring and historical overcapacity loss data to ensure its priority. A capacity violation penalty term is introduced into the multi-objective optimization function. When the alliance's future capacity exceeds the terminal's design capacity, the penalty term significantly increases the objective function value, guiding the algorithm to converge towards a capacity-fit solution. The mathematical expression of the multi-objective optimization model is as follows:

[0073] ;

[0074] in, To optimize the objective, For cross-airline transfer volume, For resource utilization, For operating costs, For passenger satisfaction, Penalties for violations related to the alliance's future capacity and terminal design capacity, when hour , The future capacity forecast of Alliance i in airline k, otherwise It is the square of the overcapacity ratio. These are weighted coefficients, and their sum is 1. For cross-airline transfer volume weighting coefficients, This is the resource utilization rate weighting coefficient. This is the weighting factor for operating costs. This is a weighting coefficient for passenger satisfaction. The weighting coefficient for the penalty items for capacity violations. ,in The weight should be no less than 0.2 to ensure the priority of capacity adaptation constraints.

[0075] Step 5: Optimization and solution stage;

[0076] A hybrid optimization algorithm, SA-APSO, is used to solve the multi-objective optimization model. Specifically, it employs a simulated annealing-adaptive particle swarm optimization algorithm to solve the multi-objective optimization model. The particle swarm optimization algorithm is used for global optimization, while the simulated annealing algorithm is used to escape local optima. Algorithm parameters are dynamically adjusted through an adaptive mechanism. The solution process includes: constructing chromosomes based on real-number encoding, where the gene positions of the chromosomes correspond to the airline's flight allocation ratio in the terminal and / or the terminal's design capacity decision variables; assigning weights to the constraints of the multi-objective optimization model, and transforming the weighted constraints into penalty terms that are incorporated into the fitness function; evaluating the chromosome fitness based on the fitness function with the fused penalty terms; and obtaining the optimal airline-terminal allocation scheme through population iterative evolution.

[0077] Specifically, a dedicated optimization strategy is designed to address capacity constraints: The airline-terminal allocation relationship and the proposed capacity of planned terminals are jointly transformed into a chromosome using real-number encoding. The first half of the chromosome corresponds to the allocation status, and the second half corresponds to the proposed capacity of the planned terminals. The degree of constraint violation is transformed into a penalty term, with the penalty coefficient for future capacity over-capacity of the alliance being three times that of other violations, ensuring that capacity constraints are prioritized. The global search capability of simulated annealing is utilized to escape local optima, combined with the local optimization advantages of adaptive particle swarm optimization to improve solution accuracy. Based on the fitness of the chromosome, iterative evolution of the population is performed, with a population size of 150-200 and 250-300 iterations. When the fitness value change rate is ≤0.001 for 20 consecutive generations, the optimal allocation scheme and (in the case of a planning scenario) the proposed design capacity of each planned terminal are output. The proposed capacity must include a confidence interval (calculated based on prediction bias) and an adjustment threshold.

[0078] Step 6: The dynamic adjustment phase of the real-time data monitoring mechanism;

[0079] Through airport operation management systems, airline scheduling systems, and alliance collaboration platforms, real-time data on actual flight takeoffs and landings, terminal resource load, alliance cooperation dynamics, and actual alliance capacity growth are collected. An early warning mechanism is constructed using the standard deviation method combined with trend analysis: when the deviation between the actual alliance capacity and the predicted value exceeds 10%, or when trend prediction indicates that the alliance capacity will reach 90% of the terminal's design capacity within the next year, the multi-objective optimization model is re-solved; for existing terminals, airline terminal allocation and resource quotas are dynamically adjusted; for planned terminals, the recommended design capacity is revised based on the latest alliance capacity growth data, and an adjustment report is output to the planning department.

[0080] As one possible implementation, this embodiment provides an existing terminal allocation scenario. Taking a multi-terminal hub airport in Southwest China as an example, it applies a base airline and alliance-coordinated airline allocation method for airline allocation. The specific steps are as follows:

[0081] Step 1. Data Collection Phase: Collect operational data from the past 5 years for 3 base airlines (A Alliance airlines / B Alliance airlines / C Alliance airlines) (150 domestic routes, 40 international routes, and an average annual growth rate of 10% in passenger throughput); related data from the 3 airline alliances (20 cooperating airlines within the same alliance, an average of 8,000 connecting flights per year, and an average capacity synergy coefficient β_ik of 1.05 for airlines within the alliance); resource data from the 3 existing terminals (Terminal A: design capacity of 9,000 passengers / hour, peak capacity of 11,000 passengers / hour; Terminal B: design capacity of 10,000 passengers / hour, peak capacity of 12,000 passengers / hour; Terminal C: design capacity of 7,000 passengers / hour, peak capacity of 9,000 passengers / hour); and external data such as a regional GDP growth rate of 6% and an average annual growth rate of 12% in tourist traffic.

[0082] Step 2. Capacity Prediction Phase: Using the ARIMA-Gradient Boosting Tree Fusion Model and Alliance Collaborative Correction, the following predictions are made for the next 3 years: Alliance A's total capacity will be 0.9 million passengers / hour, Alliance B's will be 0.8 million passengers / hour, and Alliance C's will be 0.6 million passengers / hour.

[0083] Step 3. Constraint Construction Stage: Set the future capacity of the alliance to be less than or equal to the design capacity of the terminal (e.g., for airlines in Alliance A, 0.9 ≤ 1.0, suitable for Terminal B; for airlines in Alliance B, 0.8 ≤ 0.9, suitable for Terminal A), the threshold for the proportion of airlines from the same alliance to other terminals is α = 0.1, and the terminal resource utilization rate is less than or equal to 85%.

[0084] Step 4. Optimize the model building stage: Determine the target weights (Inter-airline transit volume) (Resource utilization rate) (Operating costs) (Passenger satisfaction) (Capacity fit) Establish a multi-objective model that includes capacity violation penalties.

[0085] Step 5. Optimization and Solution Stage: The SA-APSO algorithm is used to solve the problem. The population size is 150, the number of iterations is 250, and the optimal solution is output as follows: airlines from Alliance A are centrally allocated to Terminal B (capacity 0.9≤1.0, suitable), airlines from Alliance B are allocated to Terminal A (0.8≤0.9, suitable), and airlines from Alliance C are allocated to Terminal C (0.6≤0.7, suitable). The proportion of airlines from the same alliance that are allocated to different terminals is 8.5%.

[0086] Step 6. Dynamic Adjustment Phase: Real-time monitoring shows that the actual passenger throughput of the airlines in Alliance A exceeds the predicted value by 12%, and the total capacity of the alliance rises to 0.99 million passengers / hour (close to the design capacity of Terminal B of 1.0). This triggers the multi-objective optimization model to be solved again, and some feeder routes of the airline are adjusted to the available boarding gates of Terminal B to ensure capacity matching.

[0087] The results show that the scheme reduced inter-airline transfers by 28.3%, increased terminal resource utilization by 18.6%, reduced operating costs for base airlines by 10.2%, increased passenger satisfaction by 22%, and eliminated the risk of terminal overcapacity for the next three years.

[0088] As one possible implementation, this embodiment provides a future terminal planning scenario. Taking the planning of a future terminal (tentatively referred to as Terminal D) at an eastern hub airport as an example, it applies a capacity allocation method involving base airlines and alliance airlines to provide capacity suggestions:

[0089] Step 1. Data Collection Phase: Collect operational data (average annual capacity growth of 11% over the past 5 years), alliance expansion plans (15 new international routes in the next 5 years), and regional economic growth expectations (average annual GDP growth of 7.5%) from the 8 airlines in the target service alliance (A alliance airlines).

[0090] Step 2. Capacity Forecasting Phase: The total capacity of Alliance A airlines is projected to reach 0.95 million passengers per hour over the next 5 years, with a forecast deviation of ±5%.

[0091] Step 3. Constraints and Model Construction: Based on the capacity adaptation constraint, the recommended design capacity of Terminal D is determined by reverse derivation and combined with the capacity redundancy coefficient of 1.1 = 0.95 × 1.1 = 10,450 passengers / hour. Considering the actual demand, it is rounded to 10,000 passengers / hour, with a confidence interval of 0.99-11,000 passengers / hour.

[0092] Step 4. Output Results: Output the suggested design capacity of Terminal D to the planning department as 10,000 passengers / hour (in line with the actual demand of no more than 10,000), and suggest setting a capacity warning threshold of 9,000 passengers / hour (i.e., 90% of the design capacity). When the actual capacity of the alliance reaches this threshold, the adjustment mechanism will be activated.

[0093] The planning recommendations have been adopted by the airport. Compared with traditional experience-based estimation methods, the matching degree between the planned capacity and the alliance's needs has been improved by 40%, and it is expected to reduce the transformation costs by approximately RMB 120 million over the next 10 years.

[0094] As one possible implementation, this embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a method for allocating airlines in collaboration with alliances.

[0095] As one possible implementation, this embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for coordinating airline allocation between a base airline and an alliance.

[0096] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for allocating airlines in collaboration between base airlines and alliances, characterized in that, The specific steps include the following: Collect airport operation data, construct a capacity demand forecasting model, and obtain the alliance's capacity demand forecast results within a preset future time period; Establish a multi-dimensional constraint system; With the goals of minimizing cross-airline transfer volume, maximizing terminal resource utilization, minimizing base airline operating costs, and optimizing passenger satisfaction, the model implicitly aims to maximize the compatibility between the alliance's future capacity and the terminal's designed capacity. It combines the alliance's capacity demand forecasts for the future and a multi-dimensional constraint system to construct a multi-objective optimization model. A hybrid intelligent algorithm is used to solve the multi-objective optimization model and obtain the optimal allocation scheme; Establish a real-time data monitoring mechanism to collect actual flight operation data, terminal resource usage status, dynamic changes in alliance cooperation data, and actual growth data of alliance capacity. When the deviation between the actual alliance capacity and the predicted value exceeds a preset threshold, trigger the multi-objective optimization model to be re-solved and dynamically adjust the allocation scheme.

2. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, The airport operation data includes base airline operation data, airline alliance related data, terminal resource data, and external influencing factor data; The base airline's operational data includes route network, flight takeoffs and landings, passenger throughput, transfer ratio, and capacity growth trend; The airline alliance-related data includes airline alliance affiliation, route connections within the same alliance, frequency of cross-alliance cooperation, and alliance transfer priority agreements; The terminal resource data includes the building area of ​​each terminal, the number and type of boarding gates, the capacity of security checkpoints, the capacity of baggage handling systems, the design capacity and peak capacity. The data on external influencing factors include regional economic data, tourism market demand, and policy regulation information.

3. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, The construction of the capacity demand forecasting model to obtain the alliance's capacity demand forecast results within a preset future time period includes: The capacity demand forecasting model includes the ARIMA time series model and the gradient boosting tree model; The ARIMA time series model is used to capture the time series characteristics of the capacity demand of individual airlines, and the preliminary future capacity prediction values ​​of each airline's ARIMA time series model are obtained. Gradient boosting tree model is used to explore the nonlinear relationship between external influencing factors and the capacity demand of individual airlines, and to obtain the preliminary future capacity prediction values ​​of each airline's gradient boosting tree model. The preliminary future capacity predictions of each airline are obtained by weighted fusion of the preliminary future capacity predictions of the ARIMA time series model and the gradient boosting tree model. The weighting coefficients of the weighted fusion are adjusted in reverse iteration based on the prediction error of each model in the historical prediction period. The model with the prediction error at a first set threshold receives the first weight, and the model with the prediction error at a second set threshold receives the second weight. The first set threshold is less than the set threshold, and the first weight is greater than the second weight. Based on the collaborative strategy in the alliance-related data, a collaborative coefficient is introduced to correct the preliminary future capacity forecast of each airline, so as to obtain the future capacity forecast of a single airline. The alliance's capacity demand forecast for the future is obtained by summing the future capacity forecasts of all individual airlines within the alliance.

4. The method for allocating base airlines and alliance-cooperative airlines according to claim 3, characterized in that, The alliance's capacity demand forecast results for the future preset time period include: The forecast values ​​for the number of takeoffs and landings and passenger throughput of base airlines within a predetermined time period, as well as the forecast values ​​for the total future capacity of airline alliances obtained by summing up the forecast results of individual airlines.

5. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, The multi-dimensional constraint system includes: alliance future capacity-terminal design capacity adaptation constraint, same alliance aggregation constraint, terminal capacity constraint, and operational efficiency constraint. The constraint on the matching of future capacity of the alliance with the design capacity of the terminal is as follows: when an airline alliance is assigned to a certain terminal, its total future capacity prediction value shall not exceed the design capacity of the terminal. That is, for any alliance i and the assigned terminal j, the total future capacity prediction value of alliance i shall not be greater than the design capacity of terminal j. The same alliance aggregation constraint means that base airlines and partner airlines of the same airline alliance are preferentially assigned to the same or adjacent terminals, and the proportion of the number of alliance airlines across terminals does not exceed a preset threshold. The terminal capacity constraint is that the resource occupancy of each terminal shall not exceed the design peak capacity. The operational efficiency constraint is that the average turnaround time and transfer time of airline flights do not exceed the industry standard threshold.

6. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, The mathematical expression of the multi-objective optimization model is: ； in, To optimize the objective, For cross-airline transfer volume, For resource utilization, For operating costs, For passenger satisfaction, Penalties for violations related to the alliance's future capacity and terminal design capacity, when hour , The future capacity forecast of Alliance i in airline k, otherwise It is the square of the overcapacity ratio. For cross-airline transfer volume weighting coefficients, This is the resource utilization rate weighting coefficient. This is the weighting factor for operating costs. This is a weighting coefficient for passenger satisfaction. The weighting coefficient for the penalty items for capacity violations. .

7. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, The method of using a hybrid intelligent algorithm to solve the multi-objective optimization model specifically includes: The multi-objective optimization model is solved using a hybrid optimization algorithm of simulated annealing and adaptive particle swarm optimization. The particle swarm optimization algorithm is used for global optimization, while the simulated annealing algorithm is used to escape local optima. The algorithm parameters are dynamically adjusted through an adaptive mechanism. The solution process includes: Chromosomes are constructed based on real-number encoding, and the gene loci of the chromosomes include decision variables such as the proportion of flights allocated by airlines in the terminal and / or the design capacity of the terminal. The constraints are transformed into penalty terms in the form of penalty functions and incorporated into the fitness function; The fitness of each chromosome is evaluated based on the fitness function of the fusion penalty term; The population is iteratively evolved based on fitness to obtain the optimal allocation scheme between airlines and terminals.

8. The method for allocating base airlines and alliance-cooperative airlines according to claim 1, characterized in that, After obtaining the optimal airline-terminal allocation plan, it also includes: Determine whether it is a future terminal planning scenario. If so, output the suggested design capacity of each planned terminal simultaneously. The suggested design capacity is dynamically adjusted according to the alliance's capacity growth.

9. An electronic 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 program, it implements the airline allocation method for base airlines and alliance collaboration as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the airline allocation method for base airlines and alliance collaboration as described in any one of claims 1 to 8.

Images