Air-to-air transit linked transportation collaborative stowage method considering multiple models

Abstract

The invention relates to the technical field of air freight stowage optimization, in particular to an air-to-air transit linked transportation collaborative stowage method considering multiple aircrafts, which standardizes the physical structures and operation characteristics of various aircrafts of different models into a unified parameter set, and constructs a universal stowage optimization model suitable for any aircrafts. And mapping the six-dimensional decision space through a self-adaptive hybrid coding strategy, and outputting a globally optimal linked transportation collaborative stowage scheme. According to the invention, by constructing the parameterized configuration library, the compatibility of various cargo aircrafts is realized, and the universality and efficiency of stowage operation are improved; by establishing a cross-leg collaborative constraint mechanism, the dynamic gravity center perspective management and control and transfer congestion optimization of the whole process of linked transportation are realized, and the transportation safety and the transfer efficiency are remarkably improved; and based on a multi-target intelligent optimization algorithm, systematic collaborative optimization is carried out on global business load, gravity center stability and operation cost, so that a safe, economical and efficient global optimal stowage decision is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of air cargo load planning optimization, and specifically provides a collaborative load planning method for air-to-air transfer interline transportation considering multiple aircraft types. Background Technique

[0002] As an important part of the modern logistics system, the operation efficiency and safety of air cargo directly affect the stability and timeliness of the global supply chain. In air cargo operations, load planning is a crucial ground support process. Its core task is to reasonably arrange the position of goods in the cargo hold according to the aircraft type structure, weight balance limit, cargo characteristics, and flight segment plan, ensuring that the aircraft is in a safe and balanced state during takeoff, landing, and flight, while maximizing the utilization rate of payload and transportation economic benefits. Currently, the load planning of narrow-body all-cargo aircraft mainly relies on manual experience or basic computer-aided tools, which have problems such as low efficiency, poor consistency, and dependence on personnel experience.

[0003] Currently, the load planning operations of all-cargo aircraft, especially narrow-body cargo aircraft, mainly rely on load planners' manual experience or basic computer-aided tools, with the following significant defects: Poor adaptability to multiple aircraft types: Airlines generally operate heterogeneous fleets to meet different market demands. However, existing technologies require frequent manual switching between operation manuals and performance parameters of different aircraft types, with complex operations, high workloads, and prone to calculation errors, threatening flight safety; Insufficient interline collaboration: Air-to-air transfer has become the mainstream cargo transportation mode. However, existing load planning technologies mostly optimize for single flight segments and fail to handle the collaborative load planning problems of unit load devices (ULD) across flight segments, making it difficult to balance multiple objectives such as payload utilization rate, dynamic center of gravity control, and special cargo isolation, resulting in compliance risks and efficiency losses; Single optimization objective: Traditional methods only focus on single aircraft types, single flight segments, and single indicators, unable to fully utilize the potential of heterogeneous fleets and difficult to generate a globally optimal interline transportation load planning scheme. Summary of the Invention

[0004] The purpose of the present invention is to provide a collaborative load planning method for air-to-air transfer interline transportation considering multiple aircraft types to solve the problems raised in the above background technique.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A collaborative load planning method for air-to-air transfer interline transportation considering multiple aircraft types, including the following steps: S1. Construct a parametric configuration library (PCL): Standardize the physical structures and operation characteristics of multiple different models of cargo aircraft into a unified parameter set, decouple the aircraft type characteristics from the load planning decision-making, and provide general parameter support for multi-aircraft type collaborative load planning; S2. Establish a mixed integer programming model: Based on the parameterized configuration library, construct a general load optimization model applicable to any aircraft type. The model takes the maximum overall load of the multi-segment system, the minimum center of gravity offset, and the minimum number of additional loading and unloading times at transit airports as the objective function, while satisfying the uniqueness constraint of main cargo hold loading, load limit constraint, special cargo isolation constraint, center of gravity envelope constraint, and cross-segment cargo consistency constraint. S3. The model is solved by using the adaptive hybrid coding multi-objective optimization algorithm (MAH-MOGWO): the six-dimensional decision space is mapped by the adaptive hybrid coding strategy, the algorithm's exploration and development capabilities are balanced by the phased constraint processing strategy, the core objective of the problem-oriented exploration operator is designed, and the globally optimal intermodal transport collaborative loading scheme is output.

[0006] Preferably, the parameterized configuration library in step S1 is in the form of a quintuple, defined as: ; Among them, the structural parameter set , N For the number of cargo holds (including the number of main cargo hold spaces) N A Number of cargo holds N B W represents the load limit parameter (including the maximum load per single cabin). Combination cabin weight limit Maximum load capacity of the lower cargo hold Joint weight limit for upper and lower cabins T represents the ULD version compatibility parameter (including the compatibility relationship between A type, AKE type, and LD3 type). Operation parameter set , For the hatch position parameters, For loading and unloading sequence parameters, For relay blocking parameters; Decision variable set These correspond to the location relationships of ULD loading, bulk cargo loading, and transshipment cargo, respectively. Mapping function set This enables the mapping of specific parameters of a machine model to constraints, and the automatic association between parameters and specific constraints. Constraint rule set It covers basic constraints, load and balance limits, and transfer operation limits.

[0007] Preferably, the structural parameter set in step S1 In the middle, quantity parameters , A collection of models, Indicates the number of main cargo hold seats for aircraft type k. Indicates the number of cargo holds; load limit parameters. These correspond to the weight limits for single cargo hold, combined cargo hold, single cargo hold, and combined upper and lower holds, respectively. Compatibility parameters , These correspond to ULDs of type A, type AKE, and type LD3, respectively, describing the compatibility between the aircraft type k cabin and the ULD version.

[0008] Preferably, the set of operating parameters in step S1 ,in: hatch position parameters ,in Indicates model k The j Each cabin is the location of the cabin door; Loading and unloading sequence parameters Including loading and unloading sequence parameters and uninstallation order parameters The principle of loading first and unloading later must be followed for any two goods. i and i’， like but ; Relay blocking parameters These are used to determine the additional loading and unloading status, relocation feasibility, and loading congestion of L3U category cargo at transit airports, specifically: Indicates L3U cargo i Whether additional loading or unloading occurs at the transit airport is used to determine unloading congestion in the first segment of the flight. Auxiliary variables Indicates L3U cargo i Whether the second leg can be postponed; This indicates whether the L3U remaining on board needs additional loading and unloading during the second leg of the flight, and is used to determine the loading congestion during the second leg.

[0009] Preferably, in step S2: Uniqueness constraint for loading in the main cargo hold: ; ; Load limit constraints: ; ; Special cargo isolation constraints: ; Lower cargo hold loading constraints: ; ; ; ; Load and balance constraints for narrow-body freighters: ; ; ; Centroid envelope constraint: ; ; ; Cross-segment cargo consistency constraints: ; ; ; .

[0010] Preferably, the adaptive hybrid coding strategy in step S3 includes a positional coding layer. Selecting the coding layer and loading decision layer ,in: The position coding layer uses continuous variables This indicates the relative position of cargo i on flight segment l, via Mapped to specific cabin class, The number of available cargo slots for cargo i in aircraft type k; The encoding layer is selected using binary variables. Indicates whether cargo i is loaded; The loading decision-making layer determines the final loading plan through priority strategies and selection variables.

[0011] Preferably, the phased constraint processing strategy in step S3 includes: Exploration phase :use The constraint relaxation method has a penalty function of: ,in Indicates the first i The degree of violation of a constraint, For relaxation parameters, For penalty weighting; Convergence phase Employing a precise penalty function, Where M is the maximum penalty coefficient, It is an exponential function.

[0012] Preferably, the problem-oriented exploration operator includes a center-of-gravity-oriented search operator, a load maximization search operator, and a loading / unloading sequence optimization operator, which respectively target the center-of-gravity shift, load enhancement, and transfer efficiency optimization objectives, and the activation probability is proportional to the normalized loss of the corresponding objective.

[0013] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves compatibility of a single model with multiple cargo aircraft by constructing a parameterized configuration library for multiple aircraft types, significantly improving the versatility and efficiency of loading operations. By establishing a cross-segment collaborative constraint mechanism, it realizes dynamic forward-looking control of the center of gravity and optimization of transit congestion throughout the entire process of connecting transport, significantly improving transport safety and transit efficiency. Based on a multi-objective intelligent optimization algorithm, it systematically and collaboratively optimizes global load, center of gravity stability, and operating costs under the premise of satisfying multiple complex rules, thereby achieving a globally optimal loading decision that balances safety, economy, and efficiency. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the overall structure of a collaborative loading method for air-to-air transit transportation considering multiple aircraft types, as described in this invention. Figure 2 This is a schematic diagram illustrating the operational results of a collaborative loading method for air-to-air transit transportation considering multiple aircraft types, as described in this invention. Figure 3 This is a schematic diagram illustrating the center of gravity envelope verification result of a collaborative loading method for air-to-air transfer transportation considering multiple aircraft types according to the present invention. Figure 4 This is a schematic diagram illustrating the example of the linear load verification result of the main cargo hold in a collaborative loading method for air-to-air transit considering multiple aircraft types, as presented in this invention. Figure 5 This is a schematic diagram illustrating the joint weight limit verification results of the upper and lower cabins of a collaborative loading method for air-to-air transit considering multiple aircraft types, according to the present invention. Figure 6 This is a schematic diagram of air-to-air transit transportation. Detailed Implementation

[0015] 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.

[0016] Existing technologies for handling multi-aircraft loads typically require frequent manual switching of operation manuals and performance parameters for different aircraft models. This not only significantly increases the complexity and workload of operations but also easily introduces calculation errors and misjudgments, posing a potential threat to flight safety.

[0017] Furthermore, air-to-air transshipment operations have become a core strategy for mainstream air cargo companies, requiring unit load devices (ULDs) to undergo at least two flight segments and be distributed at transit airports. However, existing loading technologies and methods are mostly optimized for single-segment (point-to-point) transport, failing to effectively address the collaborative loading issues of ULDs when transiting through hubs and crossing multiple segments. This makes it difficult to systematically and collaboratively optimize multiple conflicting objectives during intermodal transport loading, such as load utilization, dynamic center of gravity control across multiple segments, special cargo segregation rules, and transit operation convenience, leading to risks of non-compliant loading and losses in transport efficiency. Moreover, different aircraft types exhibit significant differences in load capacity, cabin layout, center of gravity envelope, and operational restrictions, resulting in varying strengths and weaknesses in addressing different optimization objectives in intermodal transport. Therefore, traditional loading methods targeting a single aircraft type, single segment, and single metric cannot fully leverage the operational potential of heterogeneous fleets and struggle to generate globally optimal collaborative loading solutions for intermodal transport.

[0018] like Figure 6 As shown, in an air-to-air transit network, cargo transportation requires at least two flight segments. Considering a typical two-segment transport link O→T→D, cargo exhibits three different loading and unloading modes: the initial segment cargo (L1U type ULD and L1C type bulk cargo) only serves the OT segment; the subsequent segment cargo (L2U type ULD and L2C type bulk cargo) is loaded from point T and serves the TD segment; the full-journey cargo (L3U type ULD and L3C type bulk cargo) runs through the entire transport link. Although it remains in aircraft status at the transit station, it may face reconfiguration due to the flight balance requirements of subsequent segments. This collaborative loading of multiple types of cargo involves a six-dimensional decision space: ULD / bulk cargo, cargo space, flight segment, cargo type, ULD form factor, and aircraft type. These dimensions are coupled with each other, forming a highly complex optimization problem.

[0019] Please see Figure 1-5 This invention provides a technical solution: a method for coordinated loading of air-to-air transit transportation considering multiple aircraft types, comprising the following steps: S1. Constructing a Parameterized Configuration Library (PCL): To uniformly handle heterogeneous constraints across multiple aircraft models, drawing on the concept of parameterized modeling for heterogeneous systems, a parameterized configuration library (PCL) for mainstream narrow-body freighters was developed. Specifically, this standardizes the physical structure and operational characteristics of various freighter models into a unified parameter set, decoupling aircraft characteristics from loading decisions and providing general parameter support for collaborative loading of multiple aircraft models. The parameterized configuration library is in 5-tuple form, defined as: ; Among them, the structural parameter set ; Quantity parameters , A collection of models, Indicates the number of main cargo hold seats for aircraft type k. This indicates the number of cargo hold slots; see Table 1 for the specific mapping.

[0020] Table 1 Aircraft Type-Cabin Quantity Mapping

[0021] Load limit parameters These correspond to the weight limits for single cargo hold compartments, combined cargo hold compartments, single cargo hold compartments, and combined upper and lower cargo hold compartments, respectively. W is the load limit parameter (including the maximum load per single cargo hold). Combination cabin weight limit Maximum load capacity of the lower cargo hold Joint weight limit for upper and lower cabins ),; Compatibility parameters , Table 2 describes the compatibility relationship between the aircraft type k cabin and the ULD version, corresponding to type A, type AKE, and type LD3 ULD respectively. The specific mapping is shown in Table 2.

[0022] Table 2 Aircraft Type-Cabin Class-Pattern Compatibility Mapping

[0023] Operation parameter set ,in: hatch position parameters ,in Indicates model k The j Each cabin is the location of the cabin door; Loading and unloading sequence parameters Including loading and unloading sequence parameters and uninstallation order parameters The principle of loading first and unloading later must be followed for any two goods. i andi’， like but ; Relay blocking parameters These are used to determine the additional loading and unloading status, relocation feasibility, and loading congestion of L3U category cargo at transit airports, specifically: Indicates L3U cargo i Whether additional loading or unloading occurs at the transit airport is used to determine unloading congestion in the first segment of the flight. Auxiliary variables Indicates L3U cargo i Whether the second leg can be postponed; This indicates whether the L3U left on board needs additional loading and unloading in the second segment, and is used to determine the loading blockage in the second segment; (1) (2) (3) (4) Equations (1)-(4) constitute the loading and unloading operation rules, which meet the loading and unloading requirements of most narrow-body cargo aircraft and can be placed in the parameterized configuration library to avoid redundant modeling. Among them, Equation (1) represents the loading and unloading operation rules of L3U cargo. i Loaded in the compartment j And there are L1U goods. i’ Loaded in a more aft compartment j’ At that time, if L3U is located between the hatch and L1U, that is Then the L3U must be additionally loaded and unloaded. Equations (2) and (3) indicate that when the L3U cargo... i Loaded in the cargo hold during the first leg of the voyage j And when there are other L3Us in the climate cabin, This indicates that it cannot be moved. , In equation (4), This indicates that the L3U was not unloaded during transit and remains on the machine; This refers to the L3U assembly that underwent additional loading and unloading at the transit airport. Decision variable set The decision variable set contains three types of variables: ; in For ULD configuration: n For the type of goods, i For goods numbering, j Main cargo hold space, lFor the flight segment, For ULD pattern, k Model; In the l Flight segment, if using k Type of aircraft, No. n Class, No. i The number of ULDs awaiting installation was assigned to the [number missing]. j If the main cargo hold location is determined, the decision variable is 1; otherwise, it is 0. ; in, For bulk cargo loading: Number the bulk cargo For unloading cargo hold space, For the flight segment, For the model, when hour, Indicates bulk goods; when hour, And by default it is LD3 type ULD.

[0024] ; To calculate and reduce the number of additional loading and unloading operations at transit airports, auxiliary variables are introduced. The default orientation is the nose forward and the tail backward, during the flight segment. l If it belongs to L3U ULD i 1 in i The value after 2 is 1, otherwise it is 0; Mapping function set This enables the mapping of specific parameters of a machine model to constraints, and the automatic association between parameters and specific constraints. For example: Example 1: Load Limitation Parameters (like → Load limit constraint, indicating that it passes The single-cabin weight limit parameter constraint specifies the threshold to the right of "≤", enabling automatic binding of aircraft load parameters and constraints. Example 2: Relay blocking parameters (like → Transit operation restrictions and constraints, indicating that through Additional loading and unloading judgment parameters dynamically trigger the relevant constraints for "L3U category cargo transshipment loading and unloading"; Constraint rule set It covers basic constraints, load and balance limits, and transfer operation limits.

[0025] The hierarchical structure of the parameterized configuration library satisfies: Basic Parameter Layer: Defines an abstract parameter template common to all cargo aircraft.

[0026]

[0027] Model-specific layer: Instantiation of parameters for a specific model k: ; ; For each model The parameter values ​​are obtained by inheriting the base template and assigning specific numerical values.

[0028] Constraint mapping layer: maps parameters to specific constraints. ; The parameterized configuration library supports mainstream narrow-body freighters, including the B757-200PCF, B737-400F, B737-800BCF, and A321P2F, and uses parameter inheritance functions. Supports rapid expansion for new models, and defines parameter inheritance functions. This makes the new model Able to build upon existing models Extend the parameter template:

[0029] in This indicates adding or modifying parameters. For example, when introducing a new model, A320-P2F, the parameter template of A321-P2F can be used, and only the number of cabins and the corresponding load limit parameters need to be modified.

[0030] S2. Establish a mixed integer programming model: Based on the parameterized configuration library, construct a general load optimization model applicable to any aircraft type. The model can automatically adapt to the constraint characteristics of different aircraft types. The model takes the maximum overall load of the multi-segment system, the minimum center of gravity offset, and the minimum number of additional loading and unloading times at transit airports as the objective function, while satisfying the uniqueness constraint of main cargo hold loading, load limit constraint, special cargo isolation constraint, center of gravity envelope constraint, and cross-segment cargo consistency constraint. Maximum objective function for load: (5) The center of gravity deviates from the minimum objective function: (6) Objective function: Minimize additional loading and unloading operations (7) Uniqueness constraint for loading in the main cargo hold: (8) (9) Load limit constraints: (10) (11) Special cargo isolation constraints: (12) Lower cargo hold loading constraints: (13) (14) (15) Load and balance constraints for narrow-body freighters: (16) (17) (18) Center of gravity calculation method: (19)

[0031] (20) Centroid envelope constraint: (twenty one) (twenty two) (twenty three) (twenty four) Cross-segment cargo consistency constraints: (25) (26) (27) Equations (5)-(7) constitute the objective function. Equation (5) represents maximizing the load of the objective function. For all ULD cargo sets, the corresponding parameter set The decision variables are adapted to the target object, and the comprehensive load of the two flight segments is measured by calculation; Equation (6) represents the objective function with the minimum centroid offset. For model-based segment The optimal center of gravity position is taken from the parameterized configuration library. The center of gravity envelope constraint in the formula is measured by calculating the difference between the actual center of gravity and the target center of gravity for each segment; Equation (7) represents the objective function of minimizing the number of additional loading and unloading operations. The first term counts the additional loading and unloading caused by blocking L1U during transit, and the second term counts the additional loading and unloading caused by blocking L2U loading and preventing it from being moved.

[0032] Equations (8)-(11) constitute the basic constraints of the main cargo hold. Among them, equation (8) specifies the aircraft type to be used. k In any flight segment Each ULD can only be loaded into one compartment that matches its type; Equation (9) stipulates that in any flight segment Each cargo hold can only carry a maximum of one ULD compatible with its type, thus ensuring the uniqueness of loading in the main cargo hold; Formula (10) specifies the aircraft type to be used. k In any flight segment The weight of the ULDs loaded on board must not exceed the maximum load capacity of the cargo hold in which they are located. Indicates the first The weight of the ULD number Indicates the first... The maximum load of the cargo hold is determined to ensure that any main cargo hold meets the load limit; Equation (11) specifies the cumulative weight limit of the cargo hold, that is, using aircraft type k, on any flight segment The above, the assembled cabin The total weight of the ULDs must not exceed the specified maximum load limit. , For model k The combination of cumulative weight limits ensures that any main cargo hold compartment meets load limits.

[0033] Equation (12) represents the isolation constraint for special goods, with subscripts 2 and 3 representing special goods with mutually exclusive properties. u and v , j and j +1 indicates two physically adjacent cargo bays in the main cargo hold, ensuring the transport of special cargo. u and v They cannot be adjacent in the main cargo hold.

[0034] Equations (13)-(15) constitute the basic constraints of the lower cargo hold. Among them, equation (13) specifies the use of Boeing series aircraft models, on any flight segment l On each cargo b(Here, loose cargo) can only be loaded into one lower cargo hold that matches its aircraft type. Multiple loose cargoes can be loaded into the same lower cargo hold. Since the lower cargo hold layout of the A321-P2F aircraft (Airbus series) is similar to that of the main cargo hold and is used to load ULDs, the loading uniqueness must be met. Therefore, formula (14) stipulates that in any flight segment l On each cargo b (ULD) can only be loaded into one lower cargo hold that matches its aircraft type, and each lower cargo hold can only hold one ULD. Formula (15) specifies the aircraft type to be used. k In any flight segment l The maximum weight limit that can be loaded in each of the upper and lower cargo holds. For the first b The weight of the goods. For model k No. h The maximum weight that the lower cargo hold can withstand is determined to ensure that any lower cargo hold meets the load limits.

[0035] Equations (16)-(24) constitute the load and balance limits. Equation (16) represents any flight segment. l To ensure flight safety, load limits must be implemented for each flight segment. The total cargo load in the main cargo hold and lower cargo hold cannot exceed the maximum load limit for the aircraft type. Equation (17) defines the mechanism for determining the maximum payload. By converting the constraints of maximum empty weight, maximum takeoff weight, and maximum landing weight into corresponding payload limits, this equation uses the minimization principle to determine the effective payload capacity, ensuring operational safety margin throughout the entire flight. Equation (18) ensures that within a given position range... Inside, the combined weight limit of the upper and lower cargo holds is less than the maximum weight that the main cargo hold and lower cargo hold can carry together. This weight is generally less than the sum of the individual load limits for the main cargo hold and the lower cargo hold. Indicates when the first j The main cargo hold compartment was loaded with cargo, and the cargo was the first... i When the number ULD is used, j The weight of the cargo loaded in compartment number 1; Indicates when the first When cargo is loaded into a lower cargo hold, the weight of the cargo loaded in that lower cargo hold; the specified position range is defined based on the main cargo hold position. This is the weight-sharing coefficient for the interval.

[0036] Equation (19) specifies the method for calculating the aircraft's center of gravity. In air transport, the position of the aircraft's center of gravity is generally represented by the meaerodynamic chord (MAC). Equation (20) defines the lever arm after the aircraft is loaded. The calculation method. In a given flight segment, the conditions for generating torque include the empty aircraft weight. fuel weight for flight segment Container / Bulk Cargo Weight.

[0037] Equations (21) to (23) ensure that the aircraft's operational balance meets the center of gravity envelope requirements under different weights and arbitrary flight segments. The center of gravity envelope of the aircraft is a range bounded by the aircraft's index and weight. The aircraft's index is a scaling and coordinate transformation of various moments of the aircraft. Among them, and These represent the flight segments when using aircraft type k. l The leading and trailing limits of the center of gravity envelope at takeoff weight (TOW), zero fuel weight (ZFW), and landing weight (LDW). This indicates that when using aircraft type k, the aircraft is in the flight segment l The exponential expression of the center of gravity corresponding to the loading results under different weights is shown in equation (24), where... The empty center of gravity index of aircraft model k. For model k Different weights W Different segments l The fuel index corresponding to the fuel weight can be obtained by looking up a table in the aircraft manual; and E k are constants, representing the aircraft's reference balance torque and reduction factor, respectively.

[0038] In the air-to-air transit network, although L3 category cargo does not require unloading at transit airports, the linear cargo layout of narrow-body freighters may result in additional operations due to obstructing the loading and unloading paths of other cargo. To save transit time and operating costs, ULD cargo in connecting transport should be kept as far away from the cargo door as possible, while ULD cargo in direct transport should be kept as close to the cargo door as possible. Equations (25)-(27) constitute additional operational constraints at transit airports. Among them, Equation (25) ensures the use of the appropriate aircraft type. k At the same time, in the same main cargo hold Goods i The aircraft is on board for both the preceding and following segments, ensuring consistency of cargo in the main cargo hold across segments; Formula (26) guarantees the use of the appropriate aircraft type. k At the same time, in the same lower cargo hold and The cargo is on the aircraft in both the preceding and following segments, thus ensuring the consistency of cargo in the lower cargo hold across segments; Equation (27) stipulates that the relative position of the transit ULD remains unchanged, thereby reducing the number of additional loading and unloading operations.

[0039] S3. The model is solved by using the adaptive hybrid coding multi-objective optimization algorithm (MAH-MOGWO): the six-dimensional decision space is mapped by the adaptive hybrid coding strategy, the algorithm exploration and development capabilities are balanced by the phased constraint processing strategy, the core objective of the problem-oriented exploration operator is designed, and the globally optimal intermodal transport collaborative loading scheme is output. Considering the structural differences between different machine models and the complexity of the six-dimensional decision space, the algorithm adopts an adaptive hybrid coding strategy to map the decision space into an operable chromosome structure. For different machine models... k The adaptive hybrid coding strategy includes a position coding layer. Selecting the coding layer and loading decision layer Three levels, among which: The position coding layer uses continuous variables Mapping the "cabin-segment" dimension represents the relative position of cargo i in segment l, through... Mapped to specific cabin class, The number of available cargo slots for cargo i in aircraft type k is taken into account, considering ULD form compatibility constraints; The encoding layer is selected using binary variables. Map the "Cargo Type - UL / Bulk Cargo" dimension to indicate whether cargo i is loaded; Loading decision layer The final loading scheme is determined by prioritization strategies and selection variables; This coding structure achieves hierarchical decoupling of the six-bit decision space. The first two layers display the coding position and selection decision, while the model, version, cargo type, and ULD / bulk cargo attributes are implicitly embedded through PCL parameters and decoding rules. By employing a phased constraint handling strategy, a dynamic balance is achieved between exploration and development. In the exploration phase... , Represents the maximum number of iterations, using Constraint relaxation methods allow solutions to violate constraints to a certain extent: (28) in, Indicates the first i The degree of violation of a constraint, For relaxation parameters, The penalty weights are applied. The relaxation parameters are dynamically adjusted as the iteration progresses. (29) During the convergence phase This is transformed into an exact penalty function to ensure the feasibility of the final solution: (30) Where M is the maximum penalty coefficient. It is an exponential function, equal to 1 when the constraint is violated, and 0 otherwise. This phased strategy ensures both the algorithm's exploratory capability and the feasibility of the final solution; Problem-oriented exploration operators include a center-of-gravity search operator, a load maximization search operator, and a loading / unloading sequence optimization operator. The center-of-gravity search operator addresses the center-of-gravity deviation problem by identifying key cargo based on torque contribution and adjusting its position to achieve center-of-gravity optimization. The load maximization search operator optimizes the load through weight descending order and filling remaining space. The loading / unloading sequence optimization operator specifically handles the transfer efficiency problem by analyzing the relative positional relationship between L3 category cargo and L1 / L2 category cargo, identifying potential congestion situations, and reducing additional loading / unloading operations through local exchange operations. The three types of operators are adaptively activated during iteration, and the activation probability is proportional to the normalized loss of the corresponding target, thus realizing dynamic search of the problem.

[0040] Figure 2 The numbers in the main cargo hold above each sub-graph represent the ULD number. The shaded slots in the main cargo hold are loaded with transit ULDs. "▲" indicates that the ULD contains special cargo u, "■" indicates that the ULD contains special cargo v, and "×" indicates that the slot does not contain a ULD. The numbers in the bulk cargo holds below represent the "quantity / weight of bulk cargo". The shaded boxes below are used to indicate the center of gravity deviation and load level of the current segment, where "△" indicates center of gravity deviation and "W" indicates load.

[0041] Specifically Figure 2 Figures (a) and (b) show the allocation results of the B757-200PCF aircraft under the center of gravity preference strategy. The center of gravity deviated by 0.62% MAC in segment 1 and 0.43% MAC in segment 2, demonstrating excellent center of gravity control performance. The average center of gravity deviation (the sum of the absolute values ​​of the two deviations) of the 60 experimental systems was only 1.06% MAC, which is significantly better than the 5.2% MAC of traditional manual load planning.

[0042] Figure 2 Figures (c) and (d) show the allocation results of the B737-800BCF model under the loading and unloading preference strategy. It can be seen that the transit ULDs (serial numbers 1-5) are located in front of the hatch and at the rear of the main cargo hold, respectively. They do not obstruct other ULDs during transit loading and unloading, and the number of additional loading and unloading operations is 0. By statistically analyzing the probability of additional loading and unloading (number of transit ULDs requiring additional loading and unloading / total number of transit ULDs), 78% of the experiments achieved the ideal state of zero additional loading and unloading for transit cargo, and the average probability of additional loading and unloading was only 3.93%, which is far lower than the 32% of the manual loading and unloading scheme.

[0043] Figure 2Figures (e) and (f) show the allocation results of the A321B2F aircraft under the load preference strategy. The load factors of the two segments reached 96.9% and 94.2% respectively. The center deviation of the two segments (2.1% MAC and 1.67% MAC) and the additional loading and unloading (1 time) also ensured the better results. Compared with the manual scheme, the average load increased by 1010 kg, and the improvement effect was significant.

[0044] Figure 3 The result is the center of gravity envelope verification. The actual center of gravity index (black dot in the figure) is strictly within the range of takeoff weight, no-fuel weight, and landing weight envelopes, and the verification is passed.

[0045] Figure 4 The results show that the actual load of each cargo hold is strictly less than the maximum load, and the verification is successful.

[0046] Figure 5 The results show that the actual load in each upper and lower compartment is strictly less than the maximum load, and the verification is successful.

[0047] This invention standardizes and encapsulates heterogeneous parameters such as cabin layout, load limits, and ULD compatibility for different aircraft models by constructing a Parametric Configuration Library (PCL). The loading system can automatically identify the aircraft model and call the corresponding parameter set, eliminating the need for manual manual switching or repeated input. This significantly improves the versatility, operational efficiency, and system maintainability of loading operations, fundamentally reducing the risk of human error.

[0048] By establishing a cross-segment collaborative constraint mechanism, the system coordinates cargo layout and center of gravity changes across the entire segment, achieving proactive control of the dynamic center of gravity. Simultaneously, the system can automatically identify and avoid transshipment and unloading congestion caused by unreasonable cargo placement, significantly shortening transit time, improving transshipment efficiency, and ensuring that each segment meets the center of gravity envelope requirements, thus enhancing the overall safety of intermodal transport.

[0049] By using a mixed integer programming model and a multi-objective adaptive optimization algorithm (MAH-MOGWO), the system can intelligently and collaboratively optimize load utilization, center of gravity stability, and transfer operation efficiency while satisfying multiple constraints such as load, balance, isolation, and transfer. This overcomes the limitations of manual compromise and local optimization, and provides airlines with a global optimal solution that balances safety, economy, and efficiency.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0051] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for collaborative stowage of air-air transshipment intermodal transport considering multi-aircraft types, characterized in that: The method comprises the following steps: S1, constructing a parameterized configuration library: standardizing the physical structures and operating characteristics of multiple different types of cargo aircraft into a unified parameter set, decoupling the aircraft type characteristics and loading decision, and providing general parameter support for multi-aircraft collaborative loading; S2, establishing a mixed integer programming model: based on the parameterized configuration library, a general loading optimization model suitable for any aircraft type is constructed, the model takes the maximum total cargo load, the minimum center of gravity deviation, and the minimum additional loading and unloading times at transfer airports as the objective function, while meeting the unique loading constraints of main cargo compartments, load weight constraints, special cargo isolation constraints, lower cargo compartment loading constraints, narrow-body all-cargo aircraft load and balance constraints, center of gravity envelope constraints, and cross-segment cargo consistency constraints; S3, solving the model using an adaptive hybrid coding multi-objective optimization algorithm: mapping a six-dimensional decision space through an adaptive hybrid coding strategy, balancing algorithm exploration and development capability through a phased constraint processing strategy, optimizing the core objective by designing a problem-oriented exploration operator, and outputting a globally optimal interline transportation collaborative loading scheme. 2.The method according to claim 1, wherein: The parameterized configuration library in step S1 is in the form of a five-tuple, defined as: ； wherein the set of structure parameters , N is a number of bays parameter, W is a weight limit parameter, and T is a ULD version compatibility parameter. set of operating parameters , for hatch position parameters, for stowage sequence parameters, for transit blockage parameters; set of decision variables , respectively, correspond to ULD stowage, bulk stowage and transshipment cargo location relationship identification; Mapping function set to implement mapping of model-specific parameters to constraints, to implement automatic association of parameters with specific constraints; Constraint rules set covering basic constraint limits, load and balance limits, transit operation limits.

3. The method of claim 2, wherein the method further comprises: The structure parameter set in the step S1 The quantity parameter , is a machine type set, represents the number of main cargo hold bays of the machine type k, represents the number of lower cargo hold bays; the load limit parameter corresponds to the single-bay weight limit of the main cargo hold, the combined-bay weight limit, the single-bay weight limit of the lower cargo hold, and the combined weight limit of the upper and lower holds, respectively; Compatibility parameters , Corresponding to A type, AKE type, LD3 type ULD respectively, describe the compatibility relationship between the cabin of the model k and the ULD version.

4. The method of claim 3, wherein the method further comprises: The set of operating parameters in said step S1 wherein: Cabin door position parameter wherein indicates a model k of the first j cabin position is a cabin door position; Loading and unloading sequence parameters Including loading and unloading sequence parameters And unloading sequence parameters Satisfy the principle of loading first and unloading last, for any two goods i And i’， If Then ; Transfer congestion parameters respectively used to determine the additional loading and unloading state, the transfer feasibility and the loading congestion of the L3U type cargo at the transfer airport, specifically: L3U goods i whether the transit airport is additionally handled for determining the unloading blockage of the first flight segment; auxiliary variable representing L3U cargo i whether the second leg can be pushed back; L3U indicates whether additional loading and unloading is required for the L3U remaining on the aircraft on the second leg, to determine loading congestion for the second leg.

5. The method of claim 4, wherein the method further comprises: The objective function in step S2 includes: Maximum cargo load objective function: ； Minimum center of gravity deviation objective function: ； Minimum additional loading and unloading times objective function: 。 6. The method of claim 5, wherein the method further comprises: In step S2: Unique loading constraints of main cargo compartments: ； ； Load weight constraints: ； ； Special cargo isolation constraints: ； Lower cargo compartment loading constraints: ； ； ； ； Narrow-body all-cargo aircraft load and balance constraints: ； ； ； Center of gravity envelope constraints: ； ； ； Cross-segment cargo consistency constraints: ； ； ； 。 7. The method of claim 6, wherein the method further comprises: The adaptive hybrid coding strategy in step S3 comprises a position coding layer , a selection coding layer and a loading decision layer , wherein: The position encoding layer employs continuous variables representing the relative position of cargo i on leg l, is mapped to a specific stowage position, is the number of available stowage positions for cargo i in aircraft model k;​ selecting coding layers using binary variables denotes whether the cargo i is loaded or not; The loading decision layer determines the final loading scheme through priority strategies and selection variables.

8. The method of claim 7, wherein the method further comprises: The phased constraint processing strategy in step S3 includes: Exploration phase : using a constraint relaxation method with a penalty function where represents the degree of violation of the i th constraint, is a relaxation parameter, is a penalty weight; Convergence phase : with an exact penalty function, where M is a maximum penalty coefficient, is an exponential function.

9. The method of claim 8, wherein the method further comprises: The problem-oriented exploration operator includes a center of gravity-oriented search operator, a maximum cargo load search operator, and a loading and unloading sequence optimization operator, which are respectively aimed at the center of gravity deviation, cargo load improvement, and transfer efficiency optimization objectives, and the activation probability is proportional to the normalized loss of the corresponding objective.

Images