Unit scheduling method, system and equipment based on multi-dimensional fatigue prediction and TEM optimization

Abstract

The application discloses a crew scheduling method, system and equipment based on multi-dimensional fatigue prediction and TEM optimization, and relates to the technical field of aviation operation control. Multi-dimensional data of a flight task is acquired, time alignment and associated integration are performed, and crew member capability parameters are generated. A demand summary result is obtained based on a TEM threat knowledge base mapping, a coupling increment is calculated in combination with a situation coupling threat graph, a situation enhanced demand summary result is formed, basic operation demand and situation enhanced demand are summarized to form comprehensive capability demand, and dynamic adjustment is performed according to a phased dynamic amplification coefficient under an unexpected event. A vulnerability correction parameter is generated based on crew operation adaptation data, situation effective operating capability and cognitive capability are calculated, an energy state parameter is obtained through a fatigue prediction model, a resilience margin index is defined based on situation effective capability and comprehensive capability demand, capability matching constraints are constructed, a scheduling scheme is generated, and parameters are updated when data changes are monitored, and rescheduling and decision chain output are performed.

Description

Technical Field

[0001] This invention belongs to the field of aviation operation control technology, and specifically discloses a crew scheduling method, system and equipment based on multi-dimensional fatigue prediction and TEM optimization. Background Technology

[0002] In the civil aviation flight operation control system, crew scheduling is a core link connecting flight plans, crew resources, and operational safety. With the increasing complexity of route networks, higher flight density, and stricter operational management requirements, the inventors discovered at least the following problems in the prior art during the development of this invention:

[0003] Fatigue assessment is based on a single dimension, often using fixed duty / rest duration thresholds, without comprehensively considering individual sleep rhythms, sleep onset time on rest days, sleep quality, duration of exceeding rest rhythm, and dynamic threat load. This makes it impossible to accurately quantify the real-time energy status of the crew, which can easily lead to safety hazards due to a mismatch between energy and tasks.

[0004] Operational threats are not quantified. Threats such as weather, fault holdout (MEL), route / airport characteristics, and aircraft type differences rely on manual judgment. A standardized TEM threat mapping system has not been formed, and it is difficult to accurately match capability requirements with the actual capabilities of the crew.

[0005] The scheduling plan is static and fixed, which leads to a lag in response to unexpected events such as flight delays, sudden failures, drastic weather changes, and air traffic control, and makes dynamic rescheduling impossible.

[0006] Compliance verification relies on manual review, and compliance checks such as qualification matching, duty periods, and rest periods are inefficient, prone to omissions, and difficult to meet the requirements of automated and traceable operation and management. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention proposes a unit scheduling method, system, and equipment based on multi-dimensional fatigue prediction and TEM optimization. Through multi-source data fusion, threat quantification, accurate fatigue prediction, dynamic demand adjustment, and multi-objective optimization, it improves the adaptability of unit scheduling to fatigue status and operational threats, and enhances dynamic adjustment capabilities.

[0008] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0009] In a first aspect, the present invention provides a unit scheduling method based on multi-dimensional fatigue prediction and TEM optimization, the method comprising:

[0010] Acquire multi-dimensional data related to flight missions, perform time alignment and correlation integration of the multi-dimensional data using flight missions as an index, and generate crew member capability parameters based on the integrated data;

[0011] Based on a pre-established TEM threat knowledge base, flight mission-related threat information is mapped to capability requirement parameters to generate a requirement summary result; a pre-established context-coupled threat map is called to obtain context-coupled increments, which are then merged with the requirement summary result to form a context-enhanced requirement summary result.

[0012] The basic operational requirements and the scenario-enhanced requirements are combined to form comprehensive capability requirements. When unexpected events are detected, the comprehensive capability requirements are dynamically adjusted in stages according to the phased dynamic amplification coefficient.

[0013] Based on the operational adaptation data of crew members, individual vulnerability profiles and vulnerability correction parameters are generated to calculate the effective situational manipulation capability and the effective situational cognition capability.

[0014] Multidimensional operational and physiological data are input into the fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters.

[0015] Based on the contextual effective capability and comprehensive capability requirements, a resilience margin index is defined, and capability matching constraints are constructed. A two-level optimization mechanism is adopted to generate a scheduling plan under the conditions of meeting capability matching constraints, energy state parameters and regulatory constraints.

[0016] When changes in flight mission-related data are detected, the summary results of enhanced scenario requirements, comprehensive capability requirements, resilience margin indicators, and energy status parameters are updated. The scheduling plan is then re-optimized and rescheduling results are output. When outputting scheduling or rescheduling plans, a self-explanatory decision chain is generated.

[0017] Optionally, the multi-dimensional data includes: crew qualification and training assessment data, route and operating environment threat data, fault retention data, meteorological data, sleep schedule and sleep quality data reported by crew members, and regulatory constraint data; wherein, the sleep schedule data includes sleep time on rest days;

[0018] The regulatory constraint data includes: duty period constraints, rest period constraints, and qualification matching constraints as defined in civil aviation operation regulations and flight operation specifications;

[0019] The acquisition of multi-dimensional data related to flight missions includes: acquiring crew qualification and training assessment data from the crew management system, acquiring route and operational environment threat data from the route and airport operation database, acquiring meteorological data from the meteorological platform, acquiring fault retention data from the maintenance system, acquiring sleep schedule data through crew mobile terminals, and acquiring regulatory constraint data from the regulatory database; and using flight missions as an index, performing time alignment and correlation integration on the above data.

[0020] Optionally, the generation of crew member capability parameters based on integrated data includes: establishing a fixed set of benchmark values ​​for operational and cognitive abilities based on qualification levels; generating initial values ​​for individual capabilities based on management adjustments; and, for candidate crews, determining overall operational and cognitive capabilities based on the initial operational and cognitive values ​​corresponding to the crew members, thereby forming crew member capability parameters.

[0021] Optionally, it also includes periodically updating the capability parameters of the crew members: calibrating the initial values ​​of individuals based on the retraining assessment results, breaking down the retraining results into operational and cognitive dimension indicators, comparing them with the statistical benchmark values ​​of the same qualification level group and adjusting them according to the degree of deviation, and updating them to the individual capability profile.

[0022] Optionally, the pre-establishment of the TEM threat knowledge base includes: analyzing historical operational data, regulatory requirements and preset threat assessment rules, associating weather phenomena, fault retention types, route and terminal operational characteristics, and aircraft type differences with flight phases, and configuring corresponding operational requirement increments and cognitive requirement increments for each association to form a threat requirement mapping table;

[0023] Among them, additional operational and cognitive requirements are required for long-fuselage aircraft.

[0024] The preset threat assessment rules include: setting four basic levels for weather-related, fault-preservation-related, route / airport characteristic-related, and operational status-related threats, with each level configured with corresponding incremental control and cognitive requirements; adjustments based on the adaptation coefficient of the flight phase application phase; and additional increments configured for long-fuselage aircraft.

[0025] Optionally, in the step-by-step dynamic adjustment of the comprehensive capability requirements according to the step-by-step dynamic amplification factor, the step-by-step dynamic amplification factor satisfies:

[0026] λc(t)=1+α Es(t)+β Ed(t)+γ C(t);

[0027] Where Es(t) is the event severity quantification value, Ed(t) is the event duration quantification value, C(t) is the coupling threat gain value, and α, β, and γ are preset weight coefficients.

[0028] Optionally, the resilience margin index includes at least a cognitive resilience margin, which is calculated using the following formula:

[0029] Rc=Ac -Dc;

[0030] Among them, Ac Dc represents context-effective cognitive ability; Dc represents comprehensive cognitive needs after context enhancement.

[0031] The effective cognitive ability in the given context satisfies:

[0032] Ac =Ac (1-vc);

[0033] Where Ac represents the candidate unit's original cognitive ability, and vc represents the cognitive vulnerability correction parameter.

[0034] Optionally, the energy state parameter is calculated using the following formula:

[0035] Pt = Pt-1 - Lt + Qt - Mt - It;

[0036] Wherein, Pt-1 is the energy value at the end of the previous task segment, Lt is the energy consumption value of the current task segment, Qt is the current rest recovery value, Mt is the continuous task load memory item, and It is the residual loss value of the energy recovery inertia item.

[0037] Secondly, the present invention provides a unit scheduling system based on multi-dimensional fatigue prediction and TEM optimization, the system comprising:

[0038] The data acquisition module is used to acquire multi-dimensional data related to flight missions, perform time alignment and correlation integration of the multi-dimensional data with flight missions as the index, and generate crew member capability parameters based on the integrated data.

[0039] The parameter mapping module is used to map flight mission-related threat information into capability requirement parameters based on a pre-established TEM threat knowledge base, and generate a requirement summary result; it calls a pre-established context coupling threat map to obtain context coupling increments, and merges them with the requirement summary result to form a context-enhanced requirement summary result;

[0040] The dynamic adjustment module is used to synthesize the results of basic operational requirements and scenario enhancement requirements into comprehensive capability requirements. When unexpected events are detected, the comprehensive capability requirements are dynamically adjusted in stages according to the phased dynamic amplification coefficient.

[0041] The vulnerability profiling module is used to generate individual vulnerability profiles and vulnerability correction parameters based on crew member operational adaptation data, and to calculate situation-effective manipulation capability and situation-effective cognitive capability.

[0042] The prediction module is used to input multi-dimensional operational and physiological data into the fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters; it defines resilience margin indicators based on situational effective capability and comprehensive capability requirements, and constructs capability matching constraints; it adopts a two-level optimization mechanism to generate a shift plan under the conditions of meeting capability matching constraints, energy status parameters and regulatory constraints.

[0043] The optimization scheduling module is used to update the summary results of scenario enhancement requirements, comprehensive capability requirements, resilience margin indicators and energy status parameters when changes in flight mission-related data are detected, and to re-optimize the scheduling plan and output the rescheduling results; when outputting the scheduling or rescheduling plan, a self-explanatory decision chain is generated.

[0044] Thirdly, the present invention provides an electronic device, the electronic device comprising:

[0045] At least one processor; and a memory communicatively connected to said at least one processor; wherein,

[0046] The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method described in any one of the first aspects.

[0047] Compared with the closest existing technology, the present invention has the following advantages:

[0048] This invention provides a crew scheduling method, system, and equipment based on multi-dimensional fatigue prediction and TEM optimization. It generates crew capability parameters through multi-dimensional data, enabling precise quantification of crew operational and cognitive abilities, thus better matching crew capabilities with flight missions and improving operational safety. Based on a TEM threat knowledge base, it standardizes and maps operational threats into capability requirements, replacing manual judgment and improving the objectivity and consistency of threat assessment.

[0049] This invention enhances the modeling of the interplay between weather, MEL (Mean Energy Requirement), flight / airport environment, aircraft characteristics, delay chains, and diurnal rhythm shifts by adding a context-coupled threat map. This transforms the modeling from a simple superposition of individual factors to a combined model, enabling the identification of complex risk scenarios that are difficult to recognize using traditional linear summation methods. Consequently, the system obtains not just static values ​​of "comprehensive operational requirements" and "comprehensive cognitive requirements," but dynamic requirement vectors tailored to specific scenarios, significantly improving the realism and resolution of capability requirement modeling.

[0050] By introducing unit vulnerability correction parameters and resilience margin indicators, this solution upgrades the focus from "whether the unit's capacity meets the standards" to "whether the unit's capacity has sufficient safety margin." Even if different units meet the requirements in terms of paper capability parameters, the system can still further distinguish their stress resistance differences in specific high-altitude, night flight, continuous delay, or complex MEL scenarios, thereby reducing the risk of mismatch caused by simple matching and improving the safety redundancy of scheduling.

[0051] In terms of fatigue prediction, by incorporating energy recovery inertia, continuous task load memory, and rhythm misalignment penalty, the system can more accurately reflect the actual energy evolution of the crew during continuous duty, multi-segment flights, and cross-day / night missions, avoiding assessment distortions caused by relying solely on current rest duration or single sleep quality. This improvement is particularly effective in identifying crews that are "superficially compliant but lack sufficient recovery," resulting in a significant safety enhancement.

[0052] Regarding scheduling optimization, this solution employs a two-tiered optimization mechanism combining risk screening and stability optimization. This enables the system to not only quickly complete rescheduling after dynamic events occur but also control the scope of changes to existing schedules, reducing large-scale chain reshuffling and operational disruptions. Simultaneously, by leveraging a self-explanatory decision chain to output the key triggering factors, risk contribution sources, and constraint fulfillment criteria for each scheduling or reshuffling, the solution's auditability and management acceptability are enhanced. Attached Figure Description

[0053] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0054] Figure 1 This is a flowchart of a unit scheduling method based on multi-dimensional fatigue prediction and TEM optimization provided by the present invention;

[0055] Figure 2 This is a schematic diagram of the unit scheduling system structure based on multi-dimensional fatigue prediction and TEM optimization provided by the present invention;

[0056] Figure 3 This is an internal structure diagram of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0057] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore merely examples, and should not be construed as limiting the scope of protection of the present invention.

[0058] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0059] This invention provides a crew scheduling method and system based on multi-dimensional fatigue prediction and TEM (Threat and Error Management) dynamic optimization, which is particularly suitable for civil aviation transportation enterprises to achieve automated, intelligent, and dynamic crew scheduling and operational risk control. The embodiments of this invention are described below with reference to the accompanying drawings.

[0060] Example 1: As Figure 1 As shown, Embodiment 1 of the present invention provides a unit scheduling method based on multi-dimensional fatigue prediction and TEM optimization, specifically including the following steps:

[0061] S101 acquires multi-dimensional data related to flight missions, performs time alignment and correlation integration on the multi-dimensional data using flight missions as an index, and generates crew member capability parameters based on the integrated data.

[0062] Based on a pre-established TEM threat knowledge base, S102 maps flight mission-related threat information into capability requirement parameters and generates a requirement summary result; it calls a pre-established context-coupled threat map to obtain context-coupled increments, and merges them with the requirement summary result to form a context-enhanced requirement summary result.

[0063] S103 combines the basic operational requirements and the scenario enhancement requirements into a comprehensive capability requirement. When unexpected events are detected, the comprehensive capability requirement is dynamically adjusted in stages according to a phased dynamic amplification factor.

[0064] S104 generates individual vulnerability profiles and vulnerability correction parameters based on crew member operational adaptation data, and calculates situation-effective maneuverability and situation-effective cognitive ability.

[0065] S105 inputs multi-dimensional operational and physiological data into a fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters; it defines resilience margin indicators based on situational effective capability and comprehensive capability requirements to construct capability matching constraints; and it adopts a two-layer optimization mechanism to generate a shift schedule under the conditions of meeting capability matching constraints, energy status parameters and regulatory constraints.

[0066] When S106 detects changes in flight mission-related data, it updates the summary results of enhanced scenario requirements, comprehensive capability requirements, resilience margin indicators, and energy status parameters, re-optimizes the scheduling plan, and outputs rescheduling results; when outputting scheduling or rescheduling plans, it generates a self-explanatory decision chain.

[0067] In step S101 above, the multi-dimensional data includes: crew qualification and training assessment data, route and operating environment threat data, fault retention data, meteorological data, sleep schedule data and sleep quality data reported by crew members, and regulatory constraint data; wherein, the sleep schedule data includes sleep time on rest days;

[0068] The regulatory constraint data includes: duty period constraints, rest period constraints, and qualification matching constraints as defined in civil aviation operation regulations and flight operation specifications;

[0069] The acquisition of multi-dimensional data related to flight missions includes: acquiring crew qualification and training assessment data from the crew management system, acquiring route and operational environment threat data from the route and airport operation database, acquiring meteorological data from the meteorological platform, acquiring fault retention data from the maintenance system, acquiring sleep schedule data through crew mobile terminals, and acquiring regulatory constraint data from the regulatory database; and using flight missions as an index, performing time alignment and correlation integration on the above data.

[0070] In one implementation, this embodiment provides that the generation of crew member capability parameters based on integrated data includes: establishing a fixed set of benchmark values ​​for operational and cognitive abilities based on qualification levels; generating initial values ​​for individual capabilities by adjusting operations according to management settings; and, for candidate crews, determining overall operational and cognitive capabilities based on the initial operational and cognitive values ​​corresponding to the crew members, thereby forming crew member capability parameters.

[0071] Furthermore, it also includes periodically updating the capability parameters of the crew members: calibrating the initial values ​​of individuals based on the retraining assessment results, breaking down the retraining results into operational and cognitive dimension indicators, comparing them with the statistical benchmark values ​​of the same qualification level group and adjusting them according to the degree of deviation, and updating them to the individual capability profile.

[0072] In step S102 above, the pre-establishment of the TEM threat knowledge base includes: analyzing historical operational data, regulatory requirements and preset threat assessment rules, associating weather phenomena, fault retention types, route and terminal operation characteristics, and aircraft type differences with flight phases respectively, and configuring corresponding operational requirement increments and cognitive requirement increments for each association to form a threat requirement mapping table;

[0073] Among them, additional operational and cognitive requirements are required for long-fuselage aircraft.

[0074] In step S103 above, during the phased dynamic adjustment of the comprehensive capability requirements according to the phased dynamic amplification factor, the phased dynamic amplification factor satisfies:

[0075] λc(t)=1+α Es(t)+β Ed(t)+γ C(t);

[0076] Where Es(t) is the event severity quantification value, Ed(t) is the event duration quantification value, C(t) is the coupling threat gain value, and α, β, and γ are preset weighting coefficients;

[0077] In one embodiment, the method for determining the event severity quantification Es(t) and the event duration quantification Ed(t) includes:

[0078] Es(t) is mapped to the event type and level according to the following table:

[0079]

[0080] If multiple events occur simultaneously, the maximum value is taken (without accumulating).

[0081] Ed(t) is based on the expected duration of the event's impact:

[0082]

[0083] That is, it increases by 0.5 for every 60 minutes of duration, up to a maximum of 2.0. For example: 30 minutes → 0.25, 120 minutes → 1.0, 240 minutes → 2.0. For events that have ended, 0.

[0084] In one embodiment, for the current flight mission, the system first identifies the set of all active threat factors. Then scan all areas in the spectrum that are completely contained within Calculate the coupling gain for the edge or hyperedge:

[0085] C )

[0086] Where e represents a coupling edge, The preset coupling strength, The phase effectiveness coefficient is set to 1.0 if the current flight phase falls within the defined phase mask of edge e, and 0 (or a partial value, such as 0.3) otherwise. The coupling strength... The determination is based on regression analysis of historical operating data or the Delphi method by experts, and is dynamically adjusted according to flight phase, crew fatigue status and crew experience.

[0087] In one embodiment, the resilience margin metric includes at least a cognitive resilience margin, which is calculated using the following formula:

[0088] Rc=Ac -Dc;

[0089] Among them, Ac Dc represents context-effective cognitive ability; Dc represents comprehensive cognitive needs after context enhancement.

[0090] In the above embodiments, the effective situational cognition ability satisfies:

[0091] Ac =Ac (1-vc);

[0092] Where Ac represents the candidate crew's original cognitive ability, and vc represents the cognitive vulnerability correction parameter. The cognitive vulnerability correction parameter vc is generated based on a weighted calculation of the crew's historical cognitive error rate, recent retraining cognitive score, the rhythm offset of reporting time relative to the midpoint of sleep, and current energy state parameters, and its value ranges from 0 to 0.3.

[0093] In one embodiment, the ability to effectively manipulate a situation must satisfy a basic matching constraint:

[0094] Effective manipulation capability in a given situation is greater than or equal to the comprehensive manipulation requirements after the situation is enhanced.

[0095] In another embodiment, the ability to effectively manipulate a situation must simultaneously satisfy the following constraints:

[0096] Basic matching constraint: The effective manipulation capability of the situation is greater than or equal to the comprehensive manipulation requirement after situation enhancement;

[0097] Resilience margin constraint: The difference between the effective control capability under the situation and the comprehensive control requirements shall not be less than the preset control margin threshold, so as to ensure that the unit has the control capability margin to cope with sudden disturbances.

[0098] Phased Adaptation Constraints: During critical flight phases, the effective control capability must meet the control requirements after phased dynamic amplification in order to adapt to changes in phased operational risks.

[0099] The above constraints together constitute the hard constraints for capability matching in the manipulation dimension, effectively solving the technical problem that traditional technologies cannot identify insufficient margin of manipulation capability under coupled threats.

[0100] In one embodiment, the energy state parameter is calculated using the following formula:

[0101] Pt = Pt-1 - Lt + Qt - Mt - It;

[0102] Wherein, Pt-1 is the energy value at the end of the previous task segment, Lt is the energy consumption value of the current task segment, Qt is the current rest recovery value, Mt is the continuous task load memory item, which is calculated by exponentially decaying and accumulating the energy consumption values ​​of past continuous task segments; It is the residual loss value of the energy recovery inertia item, which is calculated based on the sleep quality score, the time interval from wakefulness to reporting, and the diurnal rhythm offset.

[0103] The aforementioned continuous workload memory term Mt reflects the memory effect of the accumulated psychological and physiological workload of crew members across multiple consecutive task segments (such as consecutive duty days or multiple flight segments). This memory is not completely erased by short rests, and is particularly significant after sleep deprivation, circadian rhythm deviation, or continuous high-load operation. Its dimensions are consistent with the energy value (typical range 0-20).

[0104] The residual loss value It, representing the energy recovery inertia, reflects the remaining energy loss experienced by the crew after the rest period due to physiological inertia (such as sleep inertia and incomplete circadian rhythm synchronization). This parameter is primarily related to sleep quality, wake-up time, and reporting time relative to the diurnal rhythm. Its dimensions are consistent with the energy value (typical range 0-12).

[0105] Step S105 above calculates a resilience margin index based on situational effective capability and comprehensive capability requirements, and constructs capability matching constraints based on the resilience margin index; a two-layer optimization mechanism including risk access screening and stability scheduling optimization is adopted to generate a scheduling plan under the conditions of satisfying capability matching constraints, energy state parameters, and regulatory constraints. Specifically, the risk access screening is used to screen candidate unit combinations that meet safety access conditions, and the stability scheduling optimization is used to generate a scheduling plan based on a preset stability target from the screened candidate unit combinations.

[0106] Based on the situational effective manipulation ability, situational effective cognitive ability, and corresponding comprehensive ability requirements, manipulation margin, cognitive margin, and recovery margin are calculated respectively to form a resilience margin index.

[0107] The ability to perform in a given situation must be no less than the overall capability requirements, the resilience margin must be no less than the preset threshold, the energy state parameters must be at a safe level, and the requirements of civil aviation regulations must be met, all of which will serve as capability matching constraints.

[0108] A two-tier optimization mechanism is used to generate the scheduling plan:

[0109] The upper-level optimization is a risk access screening: with hard constraints such as the unit's effective capability in the scenario meeting the standard, the resilience margin meeting the safety margin, and the fatigue state being controllable, all candidate units are filtered out, and unit combinations that do not meet the safety access conditions are eliminated to ensure that the scheduling plan has a safety redundancy to cope with disturbances.

[0110] The lower-level optimization is a stable scheduling optimization: among the feasible unit combinations that have passed the upper-level screening, the optimization objectives are to minimize rescheduling disturbances, maintain the highest task continuity, optimize unit fairness, and optimize operating efficiency. Under the premise of meeting all safety and compliance constraints, the impact on the published scheduling plan is minimized, and large-scale chain adjustments and operational organization shocks are avoided.

[0111] Through the above-mentioned two-layer optimization structure, this method simultaneously solves the two major technical problems of insufficient safety margin and poor stability of rescheduling in complex scenarios, making the scheduling scheme not only more resilient to safety, but also more feasible to execute on site.

[0112] In step S106, when outputting the scheduling or rescheduling scheme, decision description data containing triggering factors, constraint satisfaction status and scheduling reasons is generated based on the threat mapping results, fatigue prediction results, capability matching constraint satisfaction results and regulatory verification results.

[0113] Example 2: Based on the same technical concept, Example 2 of this invention also provides a unit scheduling system based on multi-dimensional fatigue prediction and TEM optimization, such as... Figure 2 As shown, it includes: a data acquisition module 210, a parameter mapping module 220, a dynamic adjustment module 230, a vulnerability profiling module 240, a prediction module 250, and an optimization scheduling module 260, wherein:

[0114] The data acquisition module 210 is used to acquire multi-dimensional data related to flight missions, perform time alignment and correlation integration of the multi-dimensional data with flight missions as indexes, and generate crew member capability parameters based on the integrated data.

[0115] The parameter mapping module 220 is used to map flight mission-related threat information into capability requirement parameters based on a pre-established TEM threat knowledge base, and generate a requirement summary result; it calls a pre-established context coupling threat map to obtain context coupling increments, and merges them with the requirement summary result to form a context-enhanced requirement summary result;

[0116] The dynamic adjustment module 230 is used to synthesize the comprehensive capability requirements by summarizing the basic operational requirements and the scenario enhancement requirements, and to dynamically adjust the comprehensive capability requirements in stages according to the phased dynamic amplification coefficient when unexpected events are detected.

[0117] The vulnerability profiling module 240 is used to generate individual vulnerability profiles and vulnerability correction parameters based on the operational adaptation data of crew members, and to calculate the situation-effective manipulation capability and situation-effective cognitive capability.

[0118] The prediction module 250 is used to input multi-dimensional operational and physiological data into the fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters; it defines resilience margin indicators based on situational effective capability and comprehensive capability requirements, and constructs capability matching constraints; it adopts a two-layer optimization mechanism to generate a shift plan under the conditions of meeting capability matching constraints, energy status parameters and regulatory constraints.

[0119] The optimization scheduling module 260 is used to update the summary results of situational enhancement requirements, comprehensive capability requirements, resilience margin indicators and energy status parameters when changes in flight mission-related data are detected, and to re-optimize the scheduling plan and output the rescheduling results; when outputting the scheduling or rescheduling plan, a self-explanatory decision chain is generated.

[0120] The present invention will be further described in detail below with reference to specific embodiments, but the scope of protection of the present invention is not limited to the following embodiments.

[0121] Example 3: This example uses the application of an airline's AOC scheduling platform during the summer / autumn flight season to illustrate a crew scheduling implementation process under complex operational scenarios. The system is deployed on the operations control center server and connects to the crew management system, meteorological platform, aircraft maintenance system, airport operations database, regulatory database, and crew mobile terminals. A flight involving preceding delays, thunderstorms, a complex terminal area, and long-fuselage aircraft is used as an example to demonstrate the complete execution process of this solution.

[0122] First, the data acquisition module establishes an index based on the flight mission number, extracting the planned departure time, estimated pushback time, arrival time of preceding flights, route weather, destination airport terminal weather, MEL (Member of the Electoral Expertise) data, route type, aircraft type identifier, crew member qualifications, recent refresher training results, duty records for the most recent three consecutive days, sleep reporting data, and regulatory constraint data for each flight. A unified timeline is used to align different data sources; for example, weather forecasts are sliced ​​to check-in time, pushback time, key flight phases after takeoff, and estimated approach phases. Sleep patterns are then correlated with check-in time and mission duration.

[0123] Subsequently, the system generates the crew member capability parameters. Taking a captain and a first officer as examples, the captain's qualification benchmark is 100 points in both operation and cognition, while the first officer's qualification benchmark is 60 points in both operation and cognition. The management end adjusts these parameters based on the technical assessment, standard announcement execution rate, and performance on the retraining simulator within the most recent preset period, resulting in a captain's operation capability score of 112 points and cognition capability score of 105 points, and a first officer's operation capability score of 68 points and cognition capability score of 71 points. The candidate crew's original capability vector is thus determined to be 112 points in operation and 176 points in cognition.

[0124] Next, the parameter mapping module maps individual threats based on the TEM threat knowledge base. Assuming the flight faces moderate thunderstorms, crosswinds at the destination, terminal area congestion, and a MEL (Mission, Error, and Response) item affecting abnormal handling procedures during takeoff and approach, these are mapped to maneuvering requirement increments and cognitive requirement increments, respectively, according to the knowledge base. For example, thunderstorms are mapped to Maneuvering +3 and Cognition +5, crosswinds to Maneuvering +2 and Cognition +1, terminal area congestion to Maneuvering +0 and Cognition +4, the MEL item to Maneuvering +1 and Cognition +3, and long-fuselage aircraft to Maneuvering +2 and Cognition +2. After summing these individual threats, a total maneuvering threat requirement of 10 points and a total cognitive threat requirement of 15 points are obtained.

[0125] The system invokes the context-coupled threat map to continue calculating the threat combination increment. The system identifies "thunderstorm + terminal area congestion + delayed nighttime approach" as a high cognitive coupling combination, and "crosswinds + long-fuselage aircraft + complex MEL" as a high maneuvering coupling combination. Based on the map configuration, the aforementioned cognitive coupling combination generates an additional 6 points of cognitive coupling increment, and the subsequent maneuvering coupling combination generates an additional 4 points of maneuvering coupling increment. Following the mission phase propagation rules, 80% of the coupling impact is retained for subsequent approach phases. This results in a summary of context-enhanced requirements: maneuvering 14 points, cognitive 21 points.

[0126] The dynamic adjustment module overlays the summarized results of contextualized enhanced requirements onto the basic operational requirements. If the basic operational requirements are weighted at 30 points for manipulation and 60 points for cognition, then the comprehensive manipulation requirements are:

[0127] Dp =30+14=44

[0128] The overall cognitive needs are:

[0129] Dc = 60 + 21 = 81

[0130] If the system detects that the preceding flight delay has increased to 110 minutes and confirms that this delay will cause the actual approach time to fall within the crew's physiological low point window, then the phased dynamic cognitive amplification factor is calculated based on the event severity, duration, and coupling gain. Let α = 0.12, β = 0.08, γ = 0.15, and during the approach phase Es(t) = 1.5, Ed(t) = 1.1, C(t) = 1.3, then:

[0131] λc (t)=1+0.12×1.5+0.08×1.1+0.15×1.3=1.463

[0132] Therefore, the cognitive needs in the next stage are amplified to:

[0133] DC,app =81 × 1.463 = 118.503

[0134] The vulnerability profiling module assesses the individual situational adaptability of candidate crews. Historical records show that the captain performed consistently well during night flights and complex terminal operations, but showed only moderate recovery speed in scenarios involving rhythm deviations caused by continuous delays. Therefore, a cognitive vulnerability correction parameter vc = 0.06 was assigned to him. The co-pilot exhibited greater fluctuations in long-fuselage aircraft and crosswind landing training, resulting in a control vulnerability correction parameter vp = 0.08 and a cognitive vulnerability correction parameter vc = 0.05. The control vulnerability correction parameter vp is generated based on a weighted calculation of the crew's historical control deviation rate, recent retraining control score, aircraft type compatibility characteristics, and meteorological environmental factors, with a value ranging from 0 to 0.3.

[0135] Meanwhile, based on the two individuals' joint operation records over the past six months, the consistency between the CRM collaboration score and the standard procedures, the system provides unit collaboration compensation coefficients ηc=1.07 and ηp=1.05.

[0136] For the cognitive dimension, the system first adjusts the original cognitive ability score of 176 for vulnerability, and then combines this with collaborative compensation to obtain the situationally effective cognitive ability. If the comprehensive cognitive vulnerability adjustment is adjusted to vc=0.055, then:

[0137] Ac =176×(1-0.055)×1.07=177.8576

[0138] Similarly, for the manipulation dimension, if the original manipulation capability is 112 points, the comprehensive manipulation vulnerability is converted to vp=0.07, and the collaborative compensation coefficient ηp=1.05, then:

[0139] Ap =112×(1-0.07)×1.05=109.368

[0140] The prediction module performs continuous fatigue prediction. The system reads the crew's three most recent mission records and finds that there were two high-load flights the previous day, with the second landing late. The crew's mobile terminal reported that their sleep time on the previous rest day was 22:40, their current reporting time was 05:30, and the estimated landing time was 11:10. The system calculates the over-rest rhythm time and identifies that this reporting time is near the physiological trough. Combining the residual effects of the previous high load, a continuous mission load memory term Mt=7.5 is constructed; combining insufficient rest integrity and inadequate rhythm regression, a recovery inertia term residual loss It=5.8 is constructed. If the current mission segment's energy consumption value Lt=18.2, rest recovery value Qt=9.4, and the previous mission's end energy value Pt-1=62, then:

[0141] Pt = 62 - 18.2 + 9.4 - 7.5 - 5.8 = 39.9

[0142] Based on this, the system outputs an energy value of 39.9, which corresponds to an energy level that meets the standard but is close to the warning threshold.

[0143] Based on this, the resilience margin is calculated. The cognitive margin is:

[0144] Rc=Ac DC,app =177.8576 - 118.503 = 59.3546

[0145] The control margin is:

[0146] Rp=Ap 44 = 65.368

[0147] Taking into account the expected recovery capacity after the mission, the next day's scheduling occupancy, and backup resources, the system calculates a recovery margin of 12.4 points. If the system's preset cognitive margin threshold is 25, operational margin threshold is 20, and recovery margin threshold is 10, then this candidate unit still meets the resilience admission criteria in the current scenario.

[0148] The scheduling optimization module performs a two-layer optimization. The upper-layer optimization first filters out all candidate crew combinations that do not meet the situational effectiveness capability, energy level, duty rest regulations, or resilience thresholds, retaining only 3 feasible crews. The lower-layer optimization, based on this, comprehensively considers the stability of the published schedule, the cascading impact on subsequent flights, crew fairness, and utilization rate, prioritizing the crew that causes the least disruption to the existing schedule for this flight.

[0149] If a MEL upgrade and an increase in the destination's thunderstorm level are detected again just before launch, the system will automatically update the summary results of scenario enhancement requirements, the phased dynamic amplification coefficient, and the scenario's effectiveness and resilience margin. If the recovery margin drops to 8.6 points after recalculation, below the threshold of 10, the regroup decision matrix will immediately trigger a restrictive release strategy. The system will prioritize searching for alternative crews; if no suitable alternative crew is found within 30 minutes, it will automatically output differentiated mitigation strategies, including delaying the takeoff window, increasing dispatch monitoring frequency, requiring the cancellation of subsequent additional tasks after landing, and extending the rest period. This action chain will be recorded in the self-explanatory decision chain.

[0150] Finally, the scheduling conclusion is generated, including the triggering cause, threat contribution, coupling gain, main cause of energy decline, capability matching result, resilience margin result, regulatory verification result, and handling suggestions. Those skilled in the art can reproduce this invention based on the above steps, data items, model parameters, and solution process.

[0151] In one embodiment, the present invention also provides an electronic device, which may be a terminal, and its internal structure diagram may be as follows. Figure 3 As shown. The electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the unit scheduling method based on multi-dimensional fatigue prediction and TEM optimization as described in any one of steps S101 to S106. The display screen can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the device's casing, or an external keyboard, touchpad, or mouse.

[0152] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0153] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0154] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0155] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0156] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0157] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A unit scheduling method based on multi-dimensional fatigue prediction and TEM optimization, characterized in that, The method includes: Acquire multi-dimensional data related to flight missions, perform time alignment and correlation integration of the multi-dimensional data using flight missions as an index, and generate crew member capability parameters based on the integrated data; Based on a pre-established TEM threat knowledge base, flight mission-related threat information is mapped to capability requirement parameters to generate a requirement summary result; a pre-established context-coupled threat map is called to obtain context-coupled increments, which are then merged with the requirement summary result to form a context-enhanced requirement summary result. The basic operational requirements and the scenario-enhanced requirements are combined to form comprehensive capability requirements. When unexpected events are detected, the comprehensive capability requirements are dynamically adjusted in stages according to a phased dynamic amplification factor. Based on the operational adaptation data of crew members, individual vulnerability profiles and vulnerability correction parameters are generated to calculate the situation-effective manipulation capability and situation-effective cognitive capability. Multi-dimensional operational and physiological data are input into the fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters; resilience margin index is defined based on situational effective capability and comprehensive capability requirements to construct capability matching constraints; a two-level optimization mechanism is adopted to generate a shift plan under the conditions of meeting capability matching constraints, energy status parameters and regulatory constraints. When changes in flight mission-related data are detected, the summary results of enhanced scenario requirements, comprehensive capability requirements, resilience margin indicators, and energy status parameters are updated. The scheduling plan is then re-optimized and rescheduling results are output. When outputting scheduling or rescheduling plans, a self-explanatory decision chain is generated. The resilience margin index includes at least the cognitive resilience margin, which is calculated using the following formula: Rc=Ac -Dc; Among them, Ac Dc represents context-effective cognitive ability; Dc represents comprehensive cognitive needs after context enhancement. The effective cognitive ability in the given context satisfies: And =Ac (1-vc); Where Ac represents the candidate unit's original cognitive ability, and vc represents the cognitive vulnerability correction parameter.

2. The method according to claim 1, characterized in that, The multi-dimensional data includes: crew qualification and training assessment data, route and operating environment threat data, fault retention data, meteorological data, sleep schedule and sleep quality data reported by crew members, and regulatory constraint data; among which, the sleep schedule data includes sleep time on rest days; The regulatory constraint data includes: duty period constraints, rest period constraints, and qualification matching constraints as defined in civil aviation operation regulations and flight operation specifications; The acquisition of multi-dimensional data related to flight missions includes: acquiring crew qualification and training assessment data from the crew management system, acquiring route and operational environment threat data from the route and airport operation database, acquiring meteorological data from the meteorological platform, acquiring fault retention data from the maintenance system, acquiring sleep schedule data through crew mobile terminals, and acquiring regulatory constraint data from the regulatory database; and using flight missions as an index, performing time alignment and correlation integration on the above data.

3. The method according to claim 1, characterized in that, The process of generating crew member capability parameters based on integrated data includes: establishing a fixed set of benchmark values ​​for operational and cognitive abilities based on qualification levels; generating initial values ​​for individual capabilities based on management adjustments; and, for candidate crews, determining overall operational and cognitive capabilities based on the initial operational and cognitive values ​​corresponding to the crew members, thus forming crew member capability parameters.

4. The method according to claim 3, characterized in that, It also includes periodically updating the capability parameters of the crew members: calibrating the initial values ​​of individuals based on the retraining assessment results, breaking down the retraining results into operational and cognitive dimension indicators, comparing them with the statistical benchmark values ​​of the same qualification level group and adjusting them according to the degree of deviation, and updating them to the individual capability profile.

5. The method according to claim 1, characterized in that, The pre-establishment of the TEM threat knowledge base includes: analyzing historical operational data, regulatory requirements, and preset threat assessment rules, associating weather phenomena, fault retention types, route and terminal operational characteristics, and aircraft type differences with flight phases, and configuring corresponding operational requirement increments and cognitive requirement increments for each association to form a threat requirement mapping table; Among them, additional operational and cognitive requirements are required for long-fuselage aircraft.

6. The method according to claim 1, characterized in that, In the process of dynamically adjusting the comprehensive capability requirements in stages according to the staged dynamic amplification factor, the staged dynamic amplification factor satisfies: λc (t)=1+α It is (t)+β Ed (t)+γ C (t); Wherein, λc(t) is the phased dynamic amplification coefficient, Es(t) is the event severity quantification value, Ed(t) is the event duration quantification value, C(t) is the coupling threat gain value, and α, β, and γ are preset weight coefficients.

7. The method according to claim 1, characterized in that, The energy state parameter is calculated using the following formula: Pt = Pt-1 - Lt + Qt - Mt - It; Wherein, Pt-1 is the energy value at the end of the previous task segment, Lt is the energy consumption value of the current task segment, Qt is the current rest recovery value, Mt is the continuous task load memory item, and It is the residual loss value of the energy recovery inertia item.

8. A unit scheduling system based on multi-dimensional fatigue prediction and TEM optimization, characterized in that, The system includes: The data acquisition module is used to acquire multi-dimensional data related to flight missions, perform time alignment and correlation integration of the multi-dimensional data with flight missions as the index, and generate crew member capability parameters based on the integrated data. The parameter mapping module is used to map flight mission-related threat information into capability requirement parameters based on a pre-established TEM threat knowledge base, and generate a requirement summary result; it calls a pre-established context coupling threat map to obtain context coupling increments, and merges them with the requirement summary result to form a context-enhanced requirement summary result; The dynamic adjustment module is used to synthesize the results of basic operational requirements and scenario enhancement requirements into comprehensive capability requirements. When unexpected events are detected, the comprehensive capability requirements are dynamically adjusted in stages according to the phased dynamic amplification coefficient. The vulnerability profiling module is used to generate individual vulnerability profiles and vulnerability correction parameters based on crew member operational adaptation data, and to calculate situation-effective manipulation capability and situation-effective cognitive capability. The prediction module is used to input multi-dimensional operational and physiological data into the fatigue prediction model to predict the fatigue status of crew members and obtain energy status parameters; it defines resilience margin indicators based on situational effective capability and comprehensive capability requirements, and constructs capability matching constraints; it adopts a two-level optimization mechanism to generate a shift plan under the conditions of meeting capability matching constraints, energy status parameters and regulatory constraints. The optimization scheduling module is used to update the summary results of scenario enhancement requirements, comprehensive capability requirements, resilience margin indicators and energy status parameters when changes in flight mission-related data are detected, and to re-optimize the scheduling plan and output the rescheduling results; when outputting the scheduling or rescheduling plan, a self-explanatory decision chain is generated. The resilience margin index includes at least the cognitive resilience margin, which is calculated using the following formula: Rc=Ac -Dc; Among them, Ac Dc represents context-effective cognitive ability; Dc represents comprehensive cognitive needs after context enhancement. The effective cognitive ability in the given context satisfies: And =Ac (1-vc); Where Ac represents the candidate unit's original cognitive ability, and vc represents the cognitive vulnerability correction parameter.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1-7.

Images