A flight model allocation method, device and equipment

Abstract

The application discloses a flight model allocation method, device and equipment, the method comprises the following steps: collecting flight information, model information, airport information and economic data of a target airline; extracting a first flight set within a first preset time range from the flight information; setting a same model change penalty threshold for each flight in the first flight set; establishing a target function of the model change penalty threshold; constructing a model allocation model with multiple optimization targets; iteratively searching the target function, and obtaining a candidate model allocation scheme and a candidate model change number corresponding to the output of the model allocation model according to each candidate model change penalty threshold; and outputting a target model change penalty threshold and a target flight model allocation scheme within the first preset time range after iterative solution. The embodiment of the application can automatically adjust the model change penalty threshold, optimize the flight model allocation scheme, and improve the efficiency of flight model allocation.

Description

Technical Field

[0001] This invention relates to the field of aviation flight technology, and in particular to a method, apparatus and equipment for flight type allocation. Background Technology

[0002] With the development of the aviation industry, airlines face increasing competitive pressure due to their limited flight slot resources and complex capacity constraints. To improve operational efficiency and profitability, airlines need to rationally allocate existing flight slot resources by aircraft type. Traditional aircraft allocation methods mainly rely on manual experience and historical data. This approach often struggles to generate optimal aircraft allocation adjustment plans in a short period when faced with rapidly increasing demands for large-scale flight adjustments and a complex and volatile market environment. Summary of the Invention

[0003] The purpose of this invention is to provide a method, apparatus, and equipment for flight aircraft type allocation, which can automatically adjust the penalty threshold for changing aircraft types according to market changes, optimize the flight aircraft type allocation scheme, and improve the efficiency of airlines in allocating flight aircraft types.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0005] A first aspect of the present invention provides a method for flight aircraft type allocation, comprising:

[0006] Collect flight information, aircraft type information, airport information, and economic data of the target airline;

[0007] Extract a first set of flights within a first preset time range from the flight information;

[0008] Set the same aircraft type change penalty threshold for each flight in the first flight set; and establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; and the objective function is used to obtain the target aircraft type change penalty threshold;

[0009] Based on the flight information, aircraft type information, airport information, and economic data, an aircraft type allocation model with multi-level optimization objectives is constructed.

[0010] The objective function is iteratively searched, and the candidate model allocation scheme and the number of candidate replacement models are obtained according to the penalty threshold for each candidate replacement model in the iterative search process; the objective function value is updated according to the number of candidate replacement models.

[0011] When the preset number of iterations is reached, or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold, the target aircraft type change penalty threshold and the target flight aircraft type allocation scheme within the first preset time range are obtained.

[0012] Optionally, the objective function for establishing the model replacement penalty threshold includes:

[0013] The penalty threshold for switching models is randomly assigned multiple times;

[0014] After each random assignment, it is determined whether the change in marginal contribution before and after changing the aircraft type for each flight is greater than the aircraft type change penalty threshold; if so, the flight that is greater than the aircraft type change penalty threshold is allowed to change the aircraft type; and the number of all flights that are allowed to change the aircraft type and the sum of the corresponding changes in marginal contribution are counted; wherein, the number of flights that are allowed to change the aircraft type is used to represent the number of aircraft type changes.

[0015] Obtain several data pairs between the number of replacement models and the cumulative sum, and establish a relationship function between the number of replacement models and the cumulative sum;

[0016] Based on the aforementioned relationship function, determine the optimal number of replacement models;

[0017] An objective function for the device replacement penalty threshold is established, with the goal of minimizing the deviation between the number of device replacements corresponding to the device replacement penalty threshold and the optimal number of device replacements.

[0018] Optionally, the aircraft type allocation model is configured with a first set of constraints; wherein the first set of constraints includes flow balance constraints, aircraft type number constraints, flight coverage constraints, aircraft type number constraints, canceled flight constraints, and aircraft type connection number constraints.

[0019] Furthermore, the flow balance constraint is used to constrain the number of aircraft of each type at each flow balance checkpoint in the spatiotemporal network model in the candidate flight aircraft type allocation scheme.

[0020] The aircraft type change quantity constraint is used to restrict the number of aircraft type changes to not exceed the preset upper limit of the number of flights that are allowed to change aircraft types.

[0021] The flight coverage constraint is used to restrict that the same flight can only be operated by a maximum of one aircraft;

[0022] The aircraft type quantity constraint is used to ensure that the number of aircraft of the same type used in the candidate flight aircraft type allocation scheme each day does not exceed the number of available aircraft of that type.

[0023] The flight cancellation constraint is used in the candidate flight aircraft type allocation scheme to ensure that the total number of aircraft with all cancelled flights on the ground at each airport is not less than the minimum number of aircraft required to cover the preset set of cancelled flights.

[0024] The aircraft type connection quantity constraint is used to constrain the conservation of the number of aircraft of each type at each airport in the candidate flight aircraft type allocation scheme.

[0025] Furthermore, the step of constructing an aircraft allocation model with multi-level optimization objectives based on the flight information, aircraft type information, airport information, and economic data includes:

[0026] Based on the flight information and the aircraft type information, allocation decision variables between aircraft type and flight are constructed, and a first-level optimization objective function is established with the maximum actual number of aircraft scheduled as the objective; wherein, the constraints of the first-level optimization objective function are the first set of constraints.

[0027] The optimal solution of the first-level optimization objective function is added as a constraint condition to the first constraint condition set to obtain the second constraint condition set;

[0028] Based on the airport information and the aircraft type information, decision variables for the number of new aircraft of different types and the number of aircraft of different types are constructed for each airport, respectively. A secondary optimization objective function is established with the goal of minimizing the deviation of aircraft layout before and after optimization at each airport. The constraints of the secondary optimization objective function are the second set of constraints.

[0029] The optimal solution of the second-level optimization objective function is added as a constraint condition to the second constraint condition set to obtain the third constraint condition set;

[0030] Based on the aforementioned economic data, obtain revenue data for each flight operated by each aircraft type;

[0031] Based on the allocation decision variables and the revenue data, and with the goal of maximizing the total revenue of the first flight set, a three-level optimization objective function is established; wherein the constraints of the three-level optimization objective function are the third set of constraints.

[0032] Furthermore, the method also includes:

[0033] Based on the target flight aircraft type allocation scheme within the first preset time range, the number of target aircraft type changes within the first preset time range is obtained; and the aircraft type change ratio within the first preset time range is calculated.

[0034] Extract the second set of flights within the second preset time range from the flight information, and count the number of flights within the second preset time range;

[0035] Based on the aircraft type change ratio and the number of flights within the second preset time range, the upper limit of the number of flights that can be changed to different aircraft types within the second preset time range is calculated.

[0036] Based on the upper limit of the number of flights that can change aircraft types within the second preset range, update the constraints on the number of flights that can change aircraft types in the first constraint set to obtain the fourth constraint set.

[0037] With the goal of maximizing the total revenue of the second flight set, a four-level optimization objective function is established to obtain the target flight aircraft type allocation scheme within the second preset time range; wherein, the constraints of the four-level optimization objective function are the fourth set of constraints.

[0038] Furthermore, the updated third constraint set includes the following constraints on the number of replacement models:

[0039] Update the constraint on the number of replacement models included in the third constraint set using the following formula:

[0040]

[0041] in, To assign decision variables, if aircraft type k is used to operate flight f, then otherwise K is the set of all aircraft types; F′ is the second set of flights; k f The aircraft type originally scheduled to operate flight f; weekly The number of flights within the first preset time range; f monthly chg_obj represents the number of flights within the second preset time range, and chg_obj represents the number of target aircraft type changes within the first preset time range.

[0042] Optionally, the iterative search of the objective function specifically includes:

[0043] A particle swarm is generated to represent candidate device replacement penalty thresholds, and the position and velocity of the particle swarm are randomly initialized; wherein each particle in the particle swarm represents a candidate device replacement penalty threshold.

[0044] The position and velocity of each particle in the particle swarm are repeatedly iterated and updated according to the following equation:

[0045] The velocity update equation is

[0046] The position update equation is

[0047] The weight update equation is

[0048] Business acceptability equation is

[0049] Among them, the and It represents the velocity of the i-th particle during the it-th and it+1-th iterations; gbest is the penalty threshold for candidate device switching based on the historical best of particle i at the it-th iteration; it The threshold value for penalizing candidate device switching is the globally optimal threshold value at the it-th iteration. and These are the candidate model switching penalty thresholds for the i-th particle at the it-th and it+1-th iterations, respectively; w it and w it+1 These are the inertia weights at the it-th and it+1-th iterations, respectively; w max and w min These are the preset maximum and minimum values ​​of the inertia weight, respectively; r1 and r2 are random numbers within the interval (0,1); This is the business acceptability value at the it-th iteration, chg it The number of candidate replacement models output by the model allocation model at the it-th iteration.

[0050] A second aspect of the present invention provides a flight aircraft type allocation apparatus for implementing the flight aircraft type allocation method described in any of the first aspects above, the apparatus comprising:

[0051] The information and data acquisition module is used to collect flight information, aircraft type information, airport information, and economic data of the target airline.

[0052] The first flight set acquisition module is used to extract the first flight set within a first preset time range from the flight information;

[0053] The objective function establishment module is used to set the same aircraft type change penalty threshold for each flight in the first flight set; and to establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; and the objective function is used to obtain the target aircraft type change penalty threshold;

[0054] The aircraft type allocation model construction module is used to construct an aircraft type allocation model with multi-level optimization objectives based on the flight information, the aircraft type information, the airport information, and the economic data.

[0055] The objective function solving module is used to iteratively search the objective function and obtain the candidate model allocation scheme and the number of candidate replacement models output by the model allocation model according to the penalty threshold of each candidate replacement model in the iterative search process; and update the objective function value according to the number of candidate replacement models.

[0056] The aircraft type allocation scheme output module is used to obtain the target aircraft type penalty threshold and the target flight aircraft type allocation scheme within a first preset time range when the preset number of iterations is reached or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold.

[0057] A third aspect of the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the flight type allocation method described in any of the first aspects above.

[0058] Compared with existing technologies, embodiments of the present invention provide a method, apparatus, and device for flight aircraft type allocation. The method first extracts a first set of flights within a first preset time range from flight information; sets the same aircraft type change penalty threshold for each flight in the first set; and establishes an objective function for the aircraft type change penalty threshold. Then, it constructs an aircraft type allocation model with multi-level optimization objectives; next, iteratively searches the objective function, and based on each candidate aircraft type change penalty threshold during the iterative search process, obtains the candidate aircraft type allocation scheme and the number of candidate aircraft types to be changed corresponding to the output of the aircraft type allocation model; finally, after iterative solving, it outputs the target aircraft type change penalty threshold and the target flight aircraft type allocation scheme within the first preset time range. Embodiments of the present invention can automatically adjust the aircraft type change penalty threshold according to market changes, optimize the flight aircraft type allocation scheme, and improve the efficiency of airlines in allocating flight aircraft types. Attached Figure Description

[0059] Figure 1 This is a flowchart of a flight aircraft type allocation method provided in the first aspect embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of a spatiotemporal network provided in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of flow balance provided in an embodiment of the present invention;

[0062] Figure 4 This is a flowchart of constructing a spatiotemporal network model and flow balancing provided in an embodiment of the present invention;

[0063] Figure 5This is a fitted curve diagram showing the sum of the marginal contribution changes of the replacement models and the number of replacement models provided in the embodiments of the present invention;

[0064] Figure 6 This is a fitted curve diagram of the change in the marginal contribution gap between different models provided in the embodiments of the present invention;

[0065] Figure 7 This is a flowchart of the multi-objective calculation process in the model allocation model provided in this embodiment of the invention;

[0066] Figure 8 This is a flowchart of the optimization solution for the multi-level objective function provided in this embodiment of the invention;

[0067] Figure 9 This is a schematic diagram illustrating the optimized distribution of flight schedule adjustment types provided in this embodiment of the invention;

[0068] Figure 10 This is a schematic diagram illustrating the changes in marginal contribution before and after flight optimization, provided in an embodiment of the present invention.

[0069] Figure 11 This is a schematic diagram illustrating the daily utilization of each model after optimization, provided in an embodiment of the present invention.

[0070] Figure 12 This is a schematic diagram comparing the average number of tasks on the flight sequence before and after optimization, and the average connection time of flight tasks excluding overnight flights, provided by an embodiment of the present invention.

[0071] Figure 13 This is an algorithm flowchart of an adaptive model-switching penalty threshold model allocation model provided in an embodiment of the present invention;

[0072] Figure 14 This is a graph showing how the acceptable level of services changes with the degree of stability loss, as provided in the embodiments of the present invention.

[0073] Figure 15 This is a structural block diagram of a flight aircraft type allocation device provided in a second aspect embodiment of the present invention;

[0074] Figure 16 This is a structural block diagram of an electronic device provided in a third aspect embodiment of the present invention. Detailed Implementation

[0075] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] First, the flight optimization rules involved in the embodiments of the present invention will be described in detail, as follows:

[0077] (1) Rules for International Flight Aircraft Registration

[0078] The aircraft type used to operate international flights must meet the rules for aircraft registration in the corresponding country or airport. Registration conditions are divided into two types: aircraft type registration and tail number registration. Currently, if either type has a registration record, the aircraft type is considered to be eligible to operate the international flight.

[0079] (2) Time Limit Rules

[0080] When adjusting aircraft types, it is necessary to determine whether the connections between flights meet the minimum turnaround time. This includes the interval between the two connecting flights (pairs) and the stopover time between each segment of a flight pair or connecting flight pair. Scheduled maintenance tasks do not require minimum turnaround time for connections with other tasks.

[0081] (3) Rules for limiting the use of buoys

[0082] Flights are categorized as scheduled or unscheduled. Scheduled flights are valid for 60 days from the approval date if they match the approved route data. Unscheduled flights, when matching route data, must strictly adhere to the date requirements. Furthermore, the departure and arrival airports must be checked against the operational guidelines to ensure the corresponding aircraft type is registered (double R) for each type of flight. After determining the validity of scheduled and unscheduled routes, the last execution date in the flight segment's schedule must be checked and must not exceed 180 days from the current date.

[0083] (4) Position Limit Rules

[0084] Cabin capacity restrictions are divided into total capacity restrictions and sub-cabin restrictions. Cabin types are divided into J, W, and Y classes. The total capacity restriction is that the actual number of passengers booked for the current flight can exceed the total physical capacity of the aircraft type to which the flight is to be allocated by a maximum of two. If the number exceeds two, the flight will not be allocated to that aircraft type. The sub-cabin restriction is that the actual number of J class bookings cannot exceed the physical capacity of the J class of the aircraft type to which the flight is to be allocated. The sum of the actual number of W and Y class bookings can exceed the sum of the physical capacity of the W and Y classes of the aircraft type to which the flight is to be allocated by a maximum of two.

[0085] (5) Optimize the range rules

[0086] In this embodiment of the invention, the first preset time range is weekly, and the second preset time range is monthly, i.e., setting an optimized time range of weekly / monthly. The data acquisition time range is one day longer than the boundary of the optimization end time to handle the identification problem of the last flight pair.

[0087] (6) Model Change Rules

[0088] The revenue and costs of flights operated by different aircraft types are different. The marginal contribution of each flight is calculated as follows: Marginal contribution = Revenue from operating a certain flight using a certain aircraft type - Cost of operating a certain flight using a certain aircraft type;

[0089] By continuously adjusting the aircraft type used for flights, the marginal contribution of flights can be maximized. However, the more flights that change aircraft types, the more unstable the flight schedule will become. To maintain consistency in flight schedules before and after optimization, a balance needs to be struck between optimizing the marginal contribution and limiting the proportion of aircraft type changes. A penalty threshold for aircraft type change is set for each flight to be changed. The change is only allowed when the change in marginal contribution before and after the change exceeds this threshold. In other words, for each flight, it is necessary to determine whether the marginal contribution value after the change in aircraft type exceeds the penalty threshold.

[0090] (7) Model-priority cancellation rule

[0091] Based on the aircraft type switching rules, weighting coefficients affecting the marginal contribution value are set according to different flight types. The formula for calculating the decision marginal contribution value is as follows:

[0092] Marginal contribution value of decision = Marginal contribution * Weighting coefficient

[0093] Set the weight coefficient for regular flights to 1. For more important international flights, wide-body aircraft flights, VIP flights, etc., set the weight coefficient to greater than 1, such as 1.5 or 2. For flights of type N that can be canceled with priority, set the weight coefficient to less than 1, such as 0.5.

[0094] (8) Flight schedule rules

[0095] Fixed flight scheduling means that, provided connections are met, the aircraft type originally scheduled for a flight will not be cancelled or changed. All flight types except J / G / N are fixed. This includes international flights with any of their departure, arrival, or transit points at an international airport; and single-segment flights not identified as flight pairs.

[0096] Secondly, the network flow model based on the spatiotemporal network graph and the flow balance checkpoints involved in configuring the constraint set for the aircraft type allocation model in the embodiments of the present invention will be described in detail below:

[0097] (1) Network flow model based on spatiotemporal network graph

[0098] For each machine type k, a spatiotemporal network is established, such as... Figure 2 As shown, the horizontal axis represents time, and the vertical axis represents space; each horizontal axis represents an airport, and time increases from left to right.

[0099] Let a directed arc Gk represent a flight, and the direction of the arc represent the flight direction. The start time of the arc is the flight departure time. Because after a flight lands, it cannot immediately take off with the next flight. It must wait for the minimum turnaround time of the corresponding aircraft type and airport before it can take off with the next flight. Therefore, the end time of the arc is the flight landing time plus the minimum turnaround time.

[0100] A flight can be broken down into a takeoff and a landing operation, with nodes in the spatiotemporal network diagram representing either a takeoff or a landing operation. In short-term flight optimization, in addition to regular flight missions, there are also scheduled maintenance missions. Since the timing and aircraft type of the scheduled maintenance plan are already determined, maintenance flights are treated as a special type of flight, with the takeoff and landing airports being the same, and the starting and ending points of their corresponding directed arcs being the same airport. Because no layover time is required after maintenance, the flight can directly connect to the next flight, so the start and end times of the scheduled maintenance flight arc directly correspond to its original start and end times. Thus, after establishing the spatiotemporal network model, each arc can immediately connect to the next arc after its completion.

[0101] (2) Flow balance checkpoint

[0102] First, each flight mission is broken down into a takeoff maneuver and a landing maneuver, with each maneuver corresponding to a specific point in time. Takeoff and landing maneuvers at the same airport are then sorted by their occurrence time. For multiple maneuvers occurring at the same time, they are ordered in the order of landing first, followed by takeoff.

[0103] Secondly, a flow balance checkpoint is established at the start of the optimization process, and a daily fixed flow balance checkpoint is also established at 4:02 AM. A flow balance checkpoint is established after each takeoff maneuver. If there are multiple consecutive takeoff maneuvers without any landing maneuvers in between, a checkpoint only needs to be established after the last takeoff maneuver.

[0104] Finally, traverse all flow balance checkpoints at an airport in the order they were established, recording all arriving and departing flights between the current checkpoint n and the previous checkpoint n-1. According to the rules for establishing flow balance checkpoints, all arrival actions between n-1 and n occur before any takeoff action. Let the set of departing flights be the outbound flight set O(k,n), and the set of arriving flights be the inbound flight set I(k,n).

[0105] A schematic diagram of flow balance points established for any aircraft type at any airport, as shown below. Figure 3 As shown, taking node_3 as an example, its inbound flight set includes one arriving flight between node_2 and its outbound flight set includes two departing flights between node_2.

[0106] Figure 4The presentation also showcased the construction process of the spatiotemporal network model and flow balancing, primarily utilizing initial route adjustment data, aircraft data, revenue data by flight ID / aircraft type, turnaround times differentiated by airport and aircraft type, and configuration data. The spatiotemporal network model models aircraft flow from both temporal and spatial perspectives based on takeoff and landing times between different airports. Flow balancing further refines the spatiotemporal network model, establishing relationships between aircraft types and quantities. This means that for each airport and each aircraft type, when a flight of that type takes off, the number of that aircraft type at that airport decreases by one; when it lands, the number increases by one. At different times, it is required to ensure that there are sufficient aircraft of that type when a flight takes off.

[0107] The embodiments provided by this invention are as follows:

[0108] See Figure 1 This is a flowchart of a flight aircraft type allocation method provided in the first aspect embodiment of the present invention.

[0109] A method for flight aircraft type allocation provided in a first aspect embodiment of the present invention includes steps S11 to S16:

[0110] Step S11: Collect flight information, aircraft type information, airport information, and economic data of the target airline;

[0111] Step S12: Extract the first set of flights within the first preset time range from the flight information;

[0112] Step S13: Set the same aircraft type change penalty threshold for each flight in the first flight set; and establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; the objective function is used to obtain the target aircraft type change penalty threshold;

[0113] Step S14: Based on the flight information, aircraft type information, airport information, and economic data, construct an aircraft type allocation model with multi-level optimization objectives;

[0114] Step S15: Iteratively search the objective function, and obtain the candidate model allocation scheme and the number of candidate replacement models output by the model allocation model according to the penalty threshold of each candidate replacement model in the iterative search process; update the objective function value according to the number of candidate replacement models.

[0115] Step S16: When the preset number of iterations is reached, or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold, the target aircraft type change penalty threshold and the target flight aircraft type allocation scheme within the first preset time range are obtained.

[0116] It should be noted that the key data collected in step S11 from the target airline is shown in Table 1, including flight information, aircraft type information, airport information, and economic data.

[0117] Table 1. Key Data for the Target Airlines

[0118]

[0119] In this embodiment of the invention, a common aircraft type change penalty threshold is set for each flight. This threshold is used to determine whether a flight is allowed to change aircraft type based on whether the change in marginal contribution before and after changing aircraft type exceeds the penalty threshold. If the change in marginal contribution before and after changing aircraft type exceeds the penalty threshold, the flight is allowed to change aircraft type. Under each aircraft type change penalty threshold condition, a Fleet Assignment Model (FAM) can be used to obtain candidate aircraft type allocation schemes that satisfy multi-level optimization objectives and the corresponding number of candidate aircraft types to change. An objective function for the aircraft type change penalty threshold is established, which aims to minimize the deviation between the number of candidate aircraft types to change and the preset optimal number of aircraft types to change, and is used to obtain the target aircraft type change penalty threshold. The objective function is iteratively searched. When a preset number of iterations is reached, or the iteration change rate of the candidate aircraft type change penalty threshold is less than a preset threshold, the target aircraft type change penalty threshold and the target flight aircraft type allocation scheme within a first preset time range under the target aircraft type change penalty threshold are obtained.

[0120] In practice, the target airline can set different multi-level optimization objectives for the FAM model according to actual needs. In this embodiment of the invention, the multi-level optimization objectives are set as follows: the maximum number of actual aircraft scheduling, the minimum deviation of aircraft layout before and after optimization at each airport, and the maximum total revenue of the first flight set.

[0121] In an optional embodiment, the objective function for establishing the model replacement penalty threshold includes:

[0122] The penalty threshold for switching models is randomly assigned multiple times;

[0123] After each random assignment, it is determined whether the change in marginal contribution before and after changing the aircraft type for each flight is greater than the aircraft type change penalty threshold; if so, the flight that is greater than the aircraft type change penalty threshold is allowed to change the aircraft type; and the number of all flights that are allowed to change the aircraft type and the sum of the corresponding changes in marginal contribution are counted; wherein, the number of flights that are allowed to change the aircraft type is used to represent the number of aircraft type changes.

[0124] Obtain several data pairs between the number of replacement models and the cumulative sum, and establish a relationship function between the number of replacement models and the cumulative sum;

[0125] Based on the aforementioned relationship function, determine the optimal number of replacement models;

[0126] An objective function for the device replacement penalty threshold is established, with the goal of minimizing the deviation between the number of device replacements corresponding to the device replacement penalty threshold and the optimal number of device replacements.

[0127] It should be noted that, generally speaking, in the Fleet Assignment Model (FAM), the more aircraft type changes (chg) there are (the number of changes corresponds to the number of flights allowed to change aircraft type, excluding the specific aircraft type), the larger the cumulative sum of the marginal contribution changes from these changes (Sum_delta_profit). Based on the least squares method, data fitting is performed on Sum_delta_profit and chg, as follows... Figure 5 As shown, through calculation and analysis, it is found that within a certain error range, the coefficient of determination R-squared is 0.9951, indicating that there is a good power-law distribution characteristic between the change in flight marginal contribution value Sum_delta_profit and the number of aircraft type changes chg. The fitting relationship between the two is as follows:

[0128] Sum_delta_profit=-3.921e+09*chg^(-1.265)+3.266e+07 (1-1)

[0129] Figure 5 The data points are obtained by randomly assigning values ​​to the replacement penalty threshold multiple times. By traversing different replacement penalty thresholds, several sets of data pairs of "cumulative sum of marginal contribution change values ​​- number of replacement models" are obtained. The relationship function between the number of replacement models and the cumulative sum is obtained by fitting multiple sets of data pairs, as shown in equation (1-1).

[0130] Based on this, the present invention further fits the change in the border tribute difference (for illustrative purposes only), such as... Figure 6 As shown, the calculated R-squared value is 0.9901, and a good power-law distribution relationship exists between the two. Further analysis... Figure 5 and 6It is known that in the initial stage of the change in the number of aircraft types, the increase in side contribution is relatively large. As the number of aircraft types changed increases, the change in side contribution gradually slows down, forming a long tail. Based on the 80 / 20 rule of power-law distribution, this invention adopts the number of aircraft types with a side contribution value at the 80th percentile as the expected optimal number of aircraft types to change. This maintains the benefit of 80% of aircraft types changing while reducing the impact on flight schedule stability caused by a surge in the number of aircraft types changing due to an excessively long tail. The number of aircraft types changing corresponding to the 80th percentile of Sum_delta_profit is selected as a fixed proportion, not a constant value. That is, the number of aircraft types changing corresponding to the 80th percentile is determined as the optimal number of aircraft types changing through the relationship function between the number of aircraft types changing and the cumulative sum. Based on this, within the first preset time (week) of precise optimization, a convex function at the 80th percentile is constructed as the objective function for the iterative optimization of the Adaptive Weight Particle Swarm Optimization (AW-PSO) algorithm, which is used to further accurately solve for the aircraft type change penalty threshold. Let the number of replacement models at the 80th percentile be l. 80％ (Optimal number of replacement models), the constructed objective convex function is:

[0131] Obj weekly :Max(-chg 2 +2l 80％ ·chg)(1-2)

[0132] Specifically, the relationship between the number of replacement models and the cumulative sum of the side contribution changes is obtained by fitting the values ​​after traversing the partial replacement model penalty thresholds, as shown in Equation (1-1); the functional relationship between the two can be obtained (e.g. Figure 5 As shown, there is an upper limit to the number of machine models that can be changed and the side contribution adjustment. Exceeding this limit, no matter how the machine models are adjusted, the side contribution will not increase or will increase very little. This embodiment of the invention will obtain the number of machine models where "80% of the maximum value of Sum_delta_profit" is located, i.e., l, based on the fitted curve. 80％ The optimal number of replacement models is set as the target number. Since the optimal number of replacement models is calculated using a fitted curve, and the optimal replacement model penalty threshold is uncertain, it is necessary to continuously make chg approach l based on the objective function (e.g., equation (1-2)). 80％ This allows us to deduce the exact solution for the replacement model penalty threshold. It is equivalent to establishing an objective function for the replacement model penalty threshold with the goal of minimizing the deviation between the number of replacement models corresponding to the replacement model penalty threshold and the optimal number of replacement models, and then solving for the target replacement model penalty threshold.

[0133] The above-mentioned optimal number of replacement models and objective function settings can find a balance saddle point between the number (proportion) of replacement models and the degree of marginal contribution optimization.

[0134] In an optional embodiment, the aircraft type allocation model is further configured with a first set of constraints; wherein the first set of constraints includes flow balance constraints, aircraft type number constraints, flight coverage constraints, aircraft type number constraints, canceled flight constraints, and aircraft type connection number constraints.

[0135] In an optional embodiment, the flow balance constraint is used to constrain the number of aircraft of each type at each flow balance checkpoint in the spatiotemporal network model in the candidate flight aircraft type allocation scheme.

[0136] The aircraft type change quantity constraint is used to restrict the number of aircraft type changes to not exceed the preset upper limit of the number of flights that are allowed to change aircraft types.

[0137] The flight coverage constraint is used to restrict that the same flight can only be operated by a maximum of one aircraft;

[0138] The aircraft type quantity constraint is used to ensure that the number of aircraft of the same type used in the candidate flight aircraft type allocation scheme each day does not exceed the number of available aircraft of that type.

[0139] The flight cancellation constraint is used in the candidate flight aircraft type allocation scheme to ensure that the total number of aircraft with all cancelled flights on the ground at each airport is not less than the minimum number of aircraft required to cover the preset set of cancelled flights.

[0140] The aircraft type connection quantity constraint is used to constrain the conservation of the number of aircraft of each type at each airport in the candidate flight aircraft type allocation scheme.

[0141] It should be noted that the first set of constraints configured for the first-level optimization objective function in the FAM model requires the construction of a network flow model based on a spatiotemporal network graph. Specifically, a spatiotemporal network model is constructed for each aircraft type to set flow balance checkpoints and add flow balance constraints at fixed times each day. Furthermore, the first set of constraints also includes constraints on the number of aircraft type changes, flight coverage, number of aircraft types, canceled flights, and the number of aircraft type connections. Detailed mathematical expressions and parameters are shown in Tables 2-4 and 6.

[0142] In an optional embodiment, the step of constructing an aircraft allocation model with multi-level optimization objectives based on the flight information, the aircraft type information, the airport information, and the economic data includes:

[0143] Based on the flight information and the aircraft type information, allocation decision variables between aircraft type and flight are constructed, and a first-level optimization objective function is established with the maximum actual number of aircraft scheduled as the objective; wherein, the constraints of the first-level optimization objective function are the first set of constraints.

[0144] The optimal solution of the first-level optimization objective function is added as a constraint condition to the first constraint condition set to obtain the second constraint condition set;

[0145] Based on the airport information and the aircraft type information, decision variables for the number of new aircraft of different types and the number of aircraft of different types are constructed for each airport, respectively. A secondary optimization objective function is established with the goal of minimizing the deviation of aircraft layout before and after optimization at each airport. The constraints of the secondary optimization objective function are the second set of constraints.

[0146] The optimal solution of the second-level optimization objective function is added as a constraint condition to the second constraint condition set to obtain the third constraint condition set;

[0147] Based on the aforementioned economic data, obtain revenue data for each flight operated by each aircraft type;

[0148] Based on the allocation decision variables and the revenue data, and with the goal of maximizing the total revenue of the first flight set, a three-level optimization objective function is established; wherein the constraints of the three-level optimization objective function are the third set of constraints.

[0149] It should be noted that because some flights cannot be covered in the original flight information, and the given aircraft type data may not be consistent with the initial aircraft type distribution across airports, a multi-objective, priority-based solution approach is adopted. After solving the current objective function, the current objective value is added as a constraint to the model to solve the next objective function. The multi-objective calculation process is as follows: Figure 7 As shown.

[0150] In this embodiment of the invention, the first optimization objective is set as the maximum number of actual aircraft scheduling (maximum number of flights covered by actual aircraft), the second optimization objective is set as minimizing the aircraft layout deviation before and after optimization at each airport (maintaining consistency with the original flight schedule to the greatest extent), and the third optimization objective is set as maximizing the total revenue of the first flight set; wherein, the above three optimization objectives are all carried out within a first preset time range, and multi-objective sequential solutions are performed using solver software; and Figure 7The fourth optimization objective is to maximize the total revenue of the second flight set. This is carried out within the second preset time range. Specifically, based on the aircraft type change ratio obtained from the weekly optimization, the aircraft type change quantity constraints in the first constraint set are updated within the monthly optimization range, and the flight plan scheme with the fourth-level optimization objective function aimed at maximizing the total revenue of the second flight set is solved.

[0151] Figure 7 middle The formula for calculating the actual number of aircraft scheduled (see Tables 2-6 for formulas and detailed parameters) is as follows: V1 represents the objective function value obtained by solving the first-level optimization objective function under the first set of constraints. After calculating V1, it is added to the constraint set of the next-level optimization objective function to obtain the second set of constraints, which is the candidate aircraft allocation scheme output by the next-level optimization objective function. It must satisfy both the first set of constraints and the maximum actual number of aircraft scheduled. The formula for calculating the aircraft layout deviations before and after optimization for each airport is given, and V2 represents the objective function value obtained by solving the secondary optimization objective function under the second set of constraints. That is, in this embodiment of the invention, the solver is invoked to perform multi-objective iterative solutions based on the priority requirements of business optimization, such as... Figure 8 As shown. Among these, minimizing the maximum actual number of flights covered by aircraft and minimizing aircraft layout deviation are particularly important. These represent minimizing cancellations and changes in aircraft type, and minimizing changes compared to existing flight schedules, respectively, facilitating business continuity and stability. Solving the four-level optimization objective function set in this embodiment of the invention involves calling a complete optimization execution process, including adding constraints, setting optimization objectives, solving, post-solution processing, and adding constraints for subsequent solutions.

[0152] The parameter set, decision variables, input data, configurable multi-level optimization objective function, and first constraint set required for constructing a multi-level optimization objective model for aircraft allocation in this embodiment of the invention are shown in Tables 2, 3, 4, 5, and 6, respectively.

[0153] Table 2. Parameter Set

[0154]

[0155] Table 3. Decision Variables

[0156]

[0157] Table 4. Input Data

[0158]

[0159] Table 5. Configurable Multi-Level Optimization Objective Functions

[0160]

[0161] Table 6. First Set of Constraints

[0162]

[0163]

[0164] Formulas (1)-(4) represent the multi-level optimization objective functions that can be set for the FAM model as follows:

[0165] Equation (1) represents the maximum number of actual aircraft scheduling (the maximum number of actual flights covered by aircraft);

[0166] Equation (2) indicates that the deviation of aircraft layout before and after optimization of each airport is minimized (maintaining consistency with the original flight schedule to the greatest extent); Equation (3) indicates the maximum benefit; Equation (4) indicates the highest balance between benefit and flight schedule stability. It should be noted that the universal aircraft (virtual aircraft) in Table 2 is a virtual machine type, which is not subject to any operating rules. If a flight is covered by a virtual aircraft, no benefit will be generated, and no cost will be calculated; it can be assumed that no real aircraft is operating this flight (i.e., the flight is considered to be canceled).

[0167] Formulas (5)-(10) represent the constraints in the first set of constraints as follows: Formula (5) represents the flow balance constraint, which is used to constrain the number of aircraft of each type at each flow balance checkpoint in the spatiotemporal network model in the candidate flight aircraft type allocation scheme; Formula (6) represents the aircraft type change number constraint, which is used to constrain the number of aircraft type changes to not exceed the preset upper limit of the number of flights that can change aircraft types; Formula (7) represents the flight coverage constraint, which is used to constrain that the same flight can only be operated by one aircraft; Formula (8) represents the aircraft type number constraint, which is used to constrain the number of aircraft of the same type used in the candidate flight aircraft type allocation scheme to not exceed the number of available aircraft of that type per day; Formula (9) represents the flight cancellation constraint, which is used to constrain the total number of aircraft of all canceled flights on the ground of each airport in the candidate flight aircraft type allocation scheme to not be less than the minimum number of aircraft required to cover the preset set of canceled flights; Formula (10) represents the aircraft type connection number constraint, which is used to constrain the number of aircraft of each type at each airport in the candidate flight aircraft type allocation scheme.

[0168] Table 7 Statistical Report of Optimization Results

[0169]

[0170]

[0171] Furthermore, in this embodiment of the invention, to enable flight planners to more intuitively understand key core data such as capacity resources, aircraft type adjustment plans, differences between optimized results and the original flight plans, and high-load-rate flights requiring attention over a future period, a statistical data report is generated based on the optimization results, as shown in Table 7, using weekly data as an example.

[0172] The optimization results are analyzed as follows:

[0173] (1) Flight schedule consistency analysis

[0174] To maintain consistency with the original flight schedule before algorithm optimization and minimize the layout deviation of each aircraft type at each airport, thereby reducing instability caused by aircraft type changes, the distribution of flight schedule adjustment types after algorithm optimization is shown below, using flight data from Guangzhou base over 7 days as an example. Figure 9 As shown.

[0175] After algorithm optimization, approximately 90.3% of flights maintained their original flight schedules without requiring aircraft type changes; approximately 7.8% of flights underwent aircraft type changes; and approximately 1.9% of flights were cancelled. The purpose of these cancellations or adjustments was to balance excess capacity, optimize marginal contributions, and improve connections between flights.

[0176] (2) Marginal contribution change analysis

[0177] For flights without aircraft type changes, the marginal contribution remains unchanged before and after optimization, and will not be elaborated further here. For flights with aircraft type changes and cancellations, the changes in marginal contribution before and after optimization are as follows: Figure 10 As shown in the image, the optimization of aircraft types reduced the negative side severity of most negative-side flights and eliminated some connecting negative-side flights. Statistics show that the algorithm optimization effectively saved approximately 15,206,920 yuan in costs over seven days.

[0178] (3) Aircraft utilization rate analysis

[0179] Daily utilization of each device model after algorithm optimization is as follows: Figure 11 As shown, all aircraft types are free of over-allocation and have met the balance requirements. Statistics show that after algorithm optimization, the average daily utilization rate of all aircraft types reached 96.38%, with 85% of the aircraft types achieving 100% daily utilization. The remaining underutilized aircraft were unable to be assigned flight missions due to aircraft registration restrictions, air traffic control restrictions, and other connection rules.

[0180] (4) Connection Time Analysis

[0181] at last, Figure 12 (a) and Figure 12(b) This section compares the average number of tasks on flight sequences and the average connection time between flight tasks (excluding overnight tasks) before and after optimization. It can be seen that the flight scheduling adjustment algorithm effectively increases the average number of tasks on flight sequences while shortening the average connection time between flight tasks (excluding overnight tasks). Statistical results show that the average number of tasks on flight sequences was 14 before optimization, which increased to 17 after optimization. The average connection time was reduced from 121.111 minutes to 96.5 minutes, allowing each aircraft to perform more flight tasks.

[0182] In an optional embodiment, the method further includes:

[0183] Based on the target flight aircraft type allocation scheme within the first preset time range, the number of target aircraft type changes within the first preset time range is obtained; and the aircraft type change ratio within the first preset time range is calculated.

[0184] Extract the second set of flights within the second preset time range from the flight information, and count the number of flights within the second preset time range;

[0185] Based on the aircraft type change ratio and the number of flights within the second preset time range, the upper limit of the number of flights that can be changed to different aircraft types within the second preset time range is calculated.

[0186] Based on the upper limit of the number of flights that can change aircraft types within the second preset range, update the constraints on the number of flights that can change aircraft types in the first constraint set to obtain the fourth constraint set.

[0187] With the goal of maximizing the total revenue of the second flight set, a four-level optimization objective function is established to obtain the target flight aircraft type allocation scheme within the second preset time range; wherein, the constraints of the four-level optimization objective function are the fourth set of constraints.

[0188] It should be noted that, depending on the time range of the input data, this invention can efficiently formulate weekly and monthly flight plans, improving airline operational efficiency and reducing costs. The target aircraft type allocation scheme within a first preset time range determines the target number of aircraft type changes, and the aircraft type change ratio is calculated based on the number of flights within the first preset time range. Then, based on the aircraft type change ratio and the number of flights within a second preset time range, the upper limit of the number of flights allowed to change aircraft type within a second preset time range is obtained, which is used to update the aircraft type change number constraints included in the first constraint set, resulting in a fourth constraint set. Finally, with the goal of maximizing the total revenue of the second flight set, a four-level optimization objective function is established, and the target aircraft type allocation scheme within the second preset time range is solved to obtain the solution. That is, this embodiment of the invention can transform the target number of aircraft type changes obtained from the weekly target aircraft type allocation scheme into a constraint on the number of aircraft type changes, which is then added to the solution of the monthly flight aircraft type allocation scheme, achieving rolling optimization of the monthly flight aircraft type allocation scheme.

[0189] In an optional embodiment, the updated third constraint set includes the following constraints on the number of replacement models:

[0190] Update the constraint on the number of replacement models included in the third constraint set using the following formula:

[0191]

[0192] in, To assign decision variables, if aircraft type k is used to operate flight f, then otherwise K is the set of all aircraft types; F′ is the second set of flights; k f The aircraft type originally scheduled to operate flight f; weekly The number of flights within the first preset time range; f monthly chg_obj represents the number of flights within the second preset time range, and chg_obj represents the number of target aircraft type changes within the first preset time range.

[0193] It should be noted that within the first preset time range (e.g., weekly), the target replacement model threshold will be precisely calculated based on the objective function, and the target replacement model quantity chg_obj will be determined, where the target replacement model quantity chg_obj is equivalent to the optimal replacement model quantity l. 80％Furthermore, based on the target number of aircraft type changes and the number of flights within the first preset time range, the aircraft type change ratio within the first preset time range is calculated; and based on the aircraft type change ratio and the number of flights within the second preset time range, the upper limit of the number of flights allowed to change aircraft types within the second preset time range is calculated. This is equivalent to converting the target number of aircraft type changes within the first preset time range into an aircraft type change ratio, and then directly multiplying it by the number of flights within the second preset time range, as a constraint on the number of aircraft type changes within the second preset time range (e.g., monthly), achieving a balance between side-tribute optimization and flight schedule stability. The constraint expression is shown in equation (1-3).

[0194]

[0195] Among them, f weekly and f monthly These are the number of flights per week and per month, respectively.

[0196] In an optional embodiment, the iterative search of the objective function specifically includes:

[0197] A particle swarm is generated to represent candidate device replacement penalty thresholds, and the position and velocity of the particle swarm are randomly initialized; wherein each particle in the particle swarm represents a candidate device replacement penalty threshold.

[0198] The position and velocity of each particle in the particle swarm are repeatedly iterated and updated according to the following equation:

[0199] The velocity update equation is

[0200] The position update equation is

[0201] The weight update equation is

[0202] The equation for the degree of stability loss is:

[0203] Among them, the and It represents the velocity of the i-th particle during the it-th and it+1-th iterations; gbest is the penalty threshold for candidate device switching based on the historical best of particle i at the it-th iteration; it The threshold value for penalizing candidate device switching is the globally optimal threshold value at the it-th iteration. and These are the candidate model switching penalty thresholds for the i-th particle at the it-th and it+1-th iterations, respectively; w it and w it+1These are the inertia weights at the it-th and it+1-th iterations, respectively; w max and w min These are the preset maximum and minimum values ​​of the inertia weight, respectively; r1 and r2 are random numbers within the interval (0,1); This is the business acceptability value at the it-th iteration, chg it The number of candidate replacement models output by the model allocation model at the it-th iteration.

[0204] It should be noted that this invention establishes an adaptive model allocation model (AP-FAM) with a model switching penalty threshold within a first preset time range (weekly optimization interval) to improve optimization accuracy. The algorithm flow is as follows: Figure 9 As shown. The model is mainly divided into two parts: the first part is to establish an adaptive weighted particle swarm optimization algorithm (AW-PSO), and the second part is the FAM model. The AW-PSO algorithm can adaptively adjust the optimization inertia coefficient based on the number of candidate replacement models given by the FAM model, thereby determining the direction and trend of algorithm iteration, so as to achieve the purpose of quickly solving the target replacement penalty threshold. The termination condition of the AP-FAM model is to reach the maximum number of iterations, or the gap is less than 1%, that is, the iteration result of the replacement penalty value in this round differs from the result of the previous round by less than 1%. The objective function of AW-PSO is shown in equation (1-4):

[0205] Obj:Max(-chg 2 +2l 80％ ·chg)(1-4)

[0206] The optimal number of replacement models was calculated based on the fitted curve. 80％ The particle swarm optimization algorithm is used to adjust the device switching penalty threshold, so that the number of device switching models corresponding to the penalty threshold continuously approaches the optimal number of device switching models. 80％ Specifically, in AW-PSO, each particle represents a candidate aircraft type replacement penalty threshold. The candidate aircraft type replacement penalty threshold obtained in each iteration is input into the FAM model to output the corresponding candidate aircraft type allocation scheme and the number of candidate aircraft types to be replaced. That is, the FAM model can output the optimal aircraft type allocation scheme under the penalty threshold condition for each candidate aircraft type replacement as a candidate aircraft type allocation scheme, and statistically analyze this allocation scheme to obtain the corresponding number of candidate aircraft types to be replaced (chg). Then, the objective function value is updated based on the number of candidate aircraft types to be replaced (chg), such as... Figure 13 As shown.

[0207] In AW-PSO, the particle velocity and position update formulas are shown in equations (1-5) and (1-6). Where, w it+1 The inertial weight at the (it+1)th iteration is given by the adaptive weight dynamic formula (1-7); wmax and w min These are the preset maximum and minimum inertia weights, w. max It can be set to 0.9, w min It can be set to 0.4; r1 and r2 are random numbers in the interval (0,1); It is the business acceptability value at the it-th iteration.

[0208] Velocity update equation:

[0209] Position update equation:

[0210] Weight update equation:

[0211] Business Acceptability Equation:

[0212] The curve showing how business acceptability changes with the degree of stability loss is as follows: Figure 14 As shown, generally speaking, the maximum acceptable flight schedule adjustment ratio for business personnel is 30%. When the number of aircraft type changes is small, the degree of stable loss is low, and the acceptable level F for business personnel is [not specified]. loss Higher, F loss When the value approaches 1, the weight is adaptively adjusted to a larger value, which is beneficial for conducting a thorough global search. Similarly, as the number of device models increases, the stability loss increases, and F... loss As the weights gradually decrease to 0, they are adaptively adjusted to a smaller value, which helps to converge to the optimum within a small range. Furthermore, the change in weights affects the velocity and direction of the particle motion in equation (1-5).

[0213] See Figure 15 This is a structural block diagram of a flight aircraft type allocation device provided in a second aspect embodiment of the present invention.

[0214] A second aspect of the present invention provides a flight aircraft type allocation apparatus for implementing a flight aircraft type allocation method as described in any of the first aspects of the present invention, the apparatus comprising:

[0215] The information and data acquisition module is used to collect flight information, aircraft type information, airport information, and economic data of the target airline.

[0216] The first flight set acquisition module is used to extract the first flight set within a first preset time range from the flight information;

[0217] The objective function establishment module is used to set the same aircraft type change penalty threshold for each flight in the first flight set; and to establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; and the objective function is used to obtain the target aircraft type change penalty threshold;

[0218] The aircraft type allocation model construction module is used to construct an aircraft type allocation model with multi-level optimization objectives based on the flight information, the aircraft type information, the airport information, and the economic data.

[0219] The objective function solving module is used to iteratively search the objective function and obtain the candidate model allocation scheme and the number of candidate replacement models output by the model allocation model according to the penalty threshold of each candidate replacement model in the iterative search process; and update the objective function value according to the number of candidate replacement models.

[0220] The aircraft type allocation scheme output module is used to obtain the target aircraft type penalty threshold and the target flight aircraft type allocation scheme within a first preset time range when the preset number of iterations is reached or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold.

[0221] It should be noted that the flight type allocation device provided in the second aspect embodiment of the present invention can realize all the processes of the flight type allocation method described in any of the first aspect embodiments. The functions and technical effects of each module and unit in the device are the same as the functions and technical effects of the flight type allocation method described in the first aspect embodiments, and will not be repeated here.

[0222] See Figure 16 This is a structural block diagram of an electronic device provided in a third aspect embodiment of the present invention. The electronic device includes a processor 21, a memory 22, and a computer program stored in the memory 22 and configured to be executed by the processor 21. When the processor 21 executes the computer program, it implements the flight type allocation method as described in any of the first aspects embodiments above.

[0223] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0224] The processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor 21 may be any conventional processor. The processor 21 is the control center of the electronic device, connecting various parts of the electronic device through various interfaces and lines.

[0225] The memory 22 mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 22 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart media card (SMC), a secure digital card (SD), and a flash card, etc., or the memory 22 can also be other volatile solid-state storage devices.

[0226] It should be noted that the aforementioned electronic devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 16 The structural block diagram shown is merely a structural example of the above-described electronic device and does not constitute a limitation on the structure of the above-described electronic device. The above-described electronic device may include more or fewer components than shown, or combine certain components, or different components.

[0227] In summary, this invention provides a method, apparatus, and equipment for flight type allocation, enabling airlines to achieve short-term rolling optimization of flights within weekly or monthly timeframes based on the division of optimization time. An adaptive aircraft type allocation model (AP-FAM) with an adaptive aircraft type change penalty threshold is designed based on the improved particle swarm optimization (AW-PSO) algorithm and optimized using solver software. The aircraft type allocation model within a first preset timeframe is configured with multi-level optimization objective functions, namely, maximizing the actual number of aircraft scheduled, minimizing the aircraft layout deviation before and after optimization at each airport, and maximizing the total revenue of the first flight set. This invention can capture real-time revenue trends in the passenger flight market, adaptively adjust the aircraft type change penalty threshold to achieve flight and capacity balance for airlines, and find a balance saddle point in terms of aircraft type change ratio and marginal contribution optimization, assisting operational personnel in making real-time adjustments to flight plans for each season on a weekly and monthly basis.

[0228] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary hardware platforms, and of course, it can also be implemented entirely by hardware. Based on this understanding, all or part of the technical solution of the present invention that contributes to the background art can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0229] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for allocating aircraft types for flights, characterized in that, include: Collect flight information, aircraft type information, airport information, and economic data of the target airline; Extract a first set of flights within a first preset time range from the flight information; Set the same aircraft type change penalty threshold for each flight in the first flight set; and establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; and the objective function is used to obtain the target aircraft type change penalty threshold; Based on the flight information, aircraft type information, airport information, and economic data, an aircraft type allocation model with multi-level optimization objectives is constructed. The multi-level optimization objectives are as follows: the primary objective is to maximize the actual number of aircraft scheduled, the secondary objective is to minimize the aircraft layout deviation before and after optimization at each airport, and the tertiary objective is to maximize the total revenue of the first flight set. The objective function is iteratively searched, and the candidate model allocation scheme and the number of candidate replacement models are obtained according to the penalty threshold for each candidate replacement model in the iterative search process; the objective function value is updated according to the number of candidate replacement models. When the preset number of iterations is reached, or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold, the target aircraft type change penalty threshold and the target flight aircraft type allocation scheme within the first preset time range are obtained; wherein, the target flight aircraft type allocation scheme can shorten the connection time of aircraft to perform flight tasks and improve aircraft utilization while maintaining the stability of flight schedules, so as to achieve balance between flights and capacity; The aircraft type allocation model is configured with a first set of constraints; the first set of constraints includes flow balance constraints, aircraft type number constraints, flight coverage constraints, aircraft type number constraints, canceled flight constraints, and aircraft type connection number constraints. The flow balance constraint is used to constrain the number of aircraft of each aircraft type at each flow balance checkpoint in the spatiotemporal network model in the candidate flight aircraft type allocation scheme. The spatiotemporal network model is a directed network graph constructed for the corresponding aircraft type based on the time dimension and airport dimension, used to describe the flow process of the corresponding aircraft type in flight operation. The conservation of the number of aircraft at each flow balance checkpoint means that the number of ground aircraft of the corresponding aircraft type at the current flow balance checkpoint time on the time axis of the corresponding airport is determined by the number of ground aircraft at the previous flow balance checkpoint time, as well as the number of arriving flights and departing flights in the corresponding time interval. The step of constructing an aircraft allocation model with multi-level optimization objectives based on the flight information, aircraft type information, airport information, and economic data includes: Based on the flight information and the aircraft type information, allocation decision variables between aircraft type and flight are constructed, and a first-level optimization objective function is established with the maximum actual number of aircraft scheduled as the objective; wherein, the constraints of the first-level optimization objective function are the first set of constraints. The optimal solution of the first-level optimization objective function is added as a constraint condition to the first constraint condition set to obtain the second constraint condition set; Based on the airport information and the aircraft type information, decision variables for the number of new aircraft of different types and the number of aircraft of different types are constructed for each airport, respectively. A secondary optimization objective function is established with the goal of minimizing the deviation of aircraft layout before and after optimization at each airport. The constraints of the secondary optimization objective function are the second set of constraints. The optimal solution of the second-level optimization objective function is added as a constraint condition to the second constraint condition set to obtain the third constraint condition set; Based on the aforementioned economic data, obtain revenue data for each flight operated by each aircraft type; Based on the allocation decision variables and the revenue data, and with the goal of maximizing the total revenue of the first flight set, a three-level optimization objective function is established; wherein the constraints of the three-level optimization objective function are the third set of constraints.

2. The method for allocating flight aircraft types as described in claim 1, characterized in that, The objective function for establishing the model replacement penalty threshold includes: The penalty threshold for switching models is randomly assigned multiple times; After each random assignment, it is determined whether the change in marginal contribution before and after changing the aircraft type for each flight is greater than the aircraft type change penalty threshold; if so, the flight that is greater than the aircraft type change penalty threshold is allowed to change the aircraft type; and the number of all flights that are allowed to change the aircraft type and the sum of the corresponding changes in marginal contribution are counted; wherein, the number of flights that are allowed to change the aircraft type is used to represent the number of aircraft type changes. Obtain several data pairs between the number of replacement models and the cumulative sum, and establish a relationship function between the number of replacement models and the cumulative sum; Based on the aforementioned relationship function, determine the optimal number of replacement models; An objective function for the device replacement penalty threshold is established, with the goal of minimizing the deviation between the number of device replacements corresponding to the device replacement penalty threshold and the optimal number of device replacements.

3. The method for allocating flight aircraft types as described in claim 1, characterized in that, The aircraft type change quantity constraint is used to restrict the number of aircraft type changes to not exceed the preset upper limit of the number of flights that are allowed to change aircraft types. The flight coverage constraint is used to restrict that the same flight can only be operated by a maximum of one aircraft; The aircraft type quantity constraint is used to ensure that the number of aircraft of the same type used in the candidate flight aircraft type allocation scheme each day does not exceed the number of available aircraft of that type. The flight cancellation constraint is used in the candidate flight aircraft type allocation scheme to ensure that the total number of aircraft with all cancelled flights on the ground at each airport is not less than the minimum number of aircraft required to cover the preset set of cancelled flights. The aircraft type connection quantity constraint is used to constrain the conservation of the number of aircraft of each type at each airport in the candidate flight aircraft type allocation scheme.

4. The flight aircraft type allocation method as described in claim 1, characterized in that, The method further includes: Based on the target flight aircraft type allocation scheme within the first preset time range, the number of target aircraft type changes within the first preset time range is obtained; and the aircraft type change ratio within the first preset time range is calculated. Extract the second set of flights within the second preset time range from the flight information, and count the number of flights within the second preset time range; Based on the aircraft type change ratio and the number of flights within the second preset time range, the upper limit of the number of flights that can be changed to different aircraft types within the second preset time range is calculated. Based on the upper limit of the number of flights that can change aircraft types within the second preset range, update the constraints on the number of flights that can change aircraft types in the first constraint set to obtain the fourth constraint set. With the goal of maximizing the total revenue of the second flight set, a four-level optimization objective function is established to obtain the target flight aircraft type allocation scheme within the second preset time range; wherein, the constraints of the four-level optimization objective function are the fourth set of constraints.

5. The flight aircraft type allocation method as described in claim 4, characterized in that, The updated first constraint set includes the following constraints on the number of replacement models: Update the constraint on the number of replacement models included in the first constraint set using the following formula: ； in, To assign decision variables, if using the model k Flights ,but ,otherwise ; This is a collection of all models; This is the second flight assembly; For flights f The aircraft type originally planned to fly; The number of flights within the first preset time range; The number of flights within the second preset time range. The target number of replacement models within the first preset time range.

6. The flight aircraft type allocation method as described in claim 1, characterized in that, The iterative search of the objective function specifically includes: A particle swarm is generated to represent candidate device replacement penalty thresholds, and the position and velocity of the particle swarm are randomly initialized; wherein each particle in the particle swarm represents a candidate device replacement penalty threshold. The position and velocity of each particle in the particle swarm are repeatedly iterated and updated according to the following equation: The velocity update equation is ; The position update equation is ; The weight update equation is ; Business acceptability equation is ; Among them, the and They are the first it Subsequent During the nth iteration i The velocity of each particle; For the first it Particles in the next iteration i The best historical threshold for penalty for switching to a different device model; For the first it The globally optimal candidate model switching penalty threshold in the next iteration; and They are the first it Subsequent During the nth iteration i The penalty threshold for candidate model switching corresponding to each particle; and The first it Subsequent Inertia weights in the next iteration; and These are the preset maximum and minimum inertia weights, respectively; and for Random numbers within the interval; It is the first it The business acceptability value at the next iteration. For the first it The number of candidate replacement models output by the model allocation model in the next iteration; This refers to the number of flights within the first preset time range.

7. A flight aircraft type allocation device, characterized in that, Applicable to a flight aircraft type allocation method as described in any one of claims 1 to 6; the apparatus comprises: The information and data acquisition module is used to collect flight information, aircraft type information, airport information, and economic data of the target airline. The first flight set acquisition module is used to extract the first flight set within a first preset time range from the flight information; The objective function establishment module is used to set the same aircraft type change penalty threshold for each flight in the first flight set; and to establish an objective function for the aircraft type change penalty threshold; wherein, the aircraft type change penalty threshold is used to determine whether each flight is allowed to change aircraft type; and the objective function is used to obtain the target aircraft type change penalty threshold; The aircraft type allocation model construction module is used to construct an aircraft type allocation model with multi-level optimization objectives based on the flight information, the aircraft type information, the airport information, and the economic data. The objective function solving module is used to iteratively search the objective function and obtain the candidate model allocation scheme and the number of candidate replacement models output by the model allocation model according to the penalty threshold of each candidate replacement model in the iterative search process; and update the objective function value according to the number of candidate replacement models. The aircraft type allocation scheme output module is used to obtain the target aircraft type penalty threshold and the target flight aircraft type allocation scheme within a first preset time range when the preset number of iterations is reached or the iteration change rate of the candidate aircraft type change penalty threshold is less than the preset threshold.

8. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a flight type allocation method as described in any one of claims 1 to 6.

Images