Flight Operations Research Articles

Efficient and secure service function chain deployment method for delay optimization in air traffic information network

The air traffic information network is fundamental for ensuring the efficient and safe operation of flights. However, with the increasing demands of air traffic services, the rigidity of the traditional air traffic information network has become increasingly pronounced. Network function virtualization (NFV) technology presents an effective solution to address this issue. Within the NFV environment, the deployment of service function chains (SFC) forms the basis for achieving end-to-end air traffic services. This paper addresses the low delay and high security requirements inherent in air traffic services and proposes an efficient and secure SFC deployment for delay optimization (ESSFCD-DO) method. Primarily, the paper conducts a theoretical analysis of SFC deployment, and derives four key principles related to SFC delay, security, and resource cost. Subsequently, based on these principles, the ESSFCD-DO method is developed, which encompasses the virtual network function (VNF) aggregation algorithm for security and resources cost optimization (VNFAA-SRCO), the VNF deployment algorithm based on topology aware (VNFD-TA), and the algorithm for deploying virtual links based on k-shortest path. Experimental results demonstrate that, in comparison to other SFC deployment methods within this domain, ESSFCD-DO achieves significant optimizations in terms of end-to-end delay of SFC and long-term revenue-cost ratio, while ensuring security and service request acceptance rate.

Work of Breathing for Aviators: A Missing Link in Human Performance

In this study, we explore the work of breathing (WoB) experienced by aviators during the Anti-G Straining Maneuver (AGSM) to improve pilot safety and performance. Traditional airflow models of WoB fail to adequately distinguish between breathing rate and inspiratory frequency, leading to potentially inaccurate assessments. This mismatch can have serious implications, particularly in critical flight situations where understanding the true respiratory workload is essential for maintaining performance. To address these limitations, we used a non-sinusoidal model that captures the complexities of WoB under high inspiratory frequencies and varying dead space conditions. Our findings indicate that the classical airflow model tends to underestimate WoB, particularly at elevated inspiratory frequencies ranging from 0.5 to 2 Hz, where resistive forces play a significant role and elastic forces become negligible. Additionally, we show that an increase in dead space, coupled with high-frequency breathing, elevates WoB, heightening the risk of dyspnea among pilots. Interestingly, our analysis reveals that higher breathing rates lead to a decrease in total WoB, an unexpected finding suggesting that refining breathing patterns could help pilots optimize their energy expenditure. This research highlights the importance of examining the relationship between alveolar ventilation, breathing rate, and inspiratory frequency in greater depth within realistic flight scenarios. These insights indicate the need for targeted training programs and adaptive life-support systems to better equip pilots for managing respiratory challenges in high-stress situations. Ultimately, our research lays the groundwork for enhancing respiratory support for aviators, contributing to safer and more efficient flight operations.

Analysis of Apron Movement Control (AMC) Personnel Availability Towards Airside Supervision Operational Activities of PT Angkasa Pura I Pattimura Ambon International Airport

Apron Movement Control (AMC) is a unit that is responsible for flight operations, aircraft movement supervision, vehicle traffic, passengers and cleanliness supervision in the airside area and recording flight data on the apron. AMC officers are one of the factors that also determine the achievement of organizational goals in the airport sector, because the human resources of AMC officers are the subject of every activity in the organization, the assessment for subjects is carried out through their performance assessment. Through a combination of human resources, facilities/equipment and procedures in a series of elements that work together in carrying out the functions and tasks of supervision and service on the airside so that the goals are achieved. This study aims to determine the supervision of the AMC unit at Pattimura Ambon International Airport in terms of human resources and supporting facilities as well as obstacles in carrying out supervision in the airside area and efforts to overcome obstacles. This study uses a descriptive qualitative approach. The data used are primary and secondary data. Primary data is obtained from observation and interviews, while secondary data comes from literature studies and photographs related to the problems studied. The data analysis techniques used are data reduction, data presentation and drawing conclusions. To test the validity of the data, the triangulation technique is used. The research time was in March 2024. The results of the study showed that the supervision of the AMC unit at Pattimura Ambon International Airport was carried out in accordance with PM-IK (Quality Procedures and Work Instructions), starting from supervision of FOD (foreign object debris), fuel spills, vehicle speed according to SKEP/140/VI/1999, supervision of aircraft movements, supervision of the use of PPE according to procedures and supervision of wind running up / idle. The number of AMC unit human resources currently has 7 personnel consisting of 6 implementing personnel and 1 leader. The AMC unit has equipment that is suitable for use including Handy Talky (HT), Telescope (Binoculars), Computer, Ear Muff, Bad Marshaller, Sunglasses, Light Stick, Raincoat, Follow Me Car, Speed Trap and CCTV. The obstacle in AMC supervision is the lack of personnel so that the workload increases and there are still service users / workers who violate the rules on the airside. The AMC unit has attempted to submit additional personnel and the AMC unit strictly enforces service user discipline according to procedure.

Comparing NIRS and Pulse Oximetry for Cerebral Oxygen Saturation During Hypoxia Testing

Objective: This study evaluates the suitability of cerebral oximetry using near-infrared spectroscopy (NIRS) compared to traditional pulse oximetry (SpO2) for measuring cerebral oxygen saturation during hypoxia testing, aiming to enhance safety during flight operations and training. Material and Methods: The study included 106 participants aged 18–60 years at the Aerospace Medicine Training Center in Königsbrück. Cerebral oxygen saturation (rSO2) and peripheral oxygen saturation (SpO2) were measured using the INVOS™ 5100C cerebral oximeter and Masimo™ MS5 pulse oximeter, respectively. Measurements were taken at baseline, during hypoxia at 25,000 feet, and post recovery. Data analysis included regression analysis, Bland–Altman plots, and concordance correlation coefficients (CCC). Ethical approval was obtained from the Hannover Medical School. Data from 100 participants were analyzed. Results: Baseline SpO2 was 97.5 ± 1.5%, and baseline rSO2 was 77.25 ± 6.4%. During hypoxia, SpO2 dropped significantly, while rSO2 showed higher values. SpO2 recovered faster than rSO2. Deviations in rSO2 between the right and left sides during hypoxia were minimal. Lin’s CCC indicated moderate to substantial concordance. NIRS measurements were more stable and less prone to disturbances, with 95 disruptions in pulse oximetry, 25 of which were potentially critical. Conclusions: NIRS is a reliable method for detecting cerebral oxygen saturation, offering significant advantages over traditional pulse oximetry in stability and reliability during hypoxia testing. NIRS is less error-prone, supporting its use for continuous monitoring in aviation settings and enhancing flight safety by providing more accurate hypoxia detection.

Pollutant Dispersion of Aircraft Exhaust Gas during the Landing and Takeoff Cycle with Improved Gaussian Diffusion Model

Evaluating aviation emissions and examining the dispersion properties of contaminants are crucial for understanding atmospheric pollution. To assess the pollutant emissions and dispersion of aircraft during the landing and takeoff (LTO) cycle, and address air pollution surrounding the airport resulting from flight operations, this study evaluated emissions throughout the LTO phase based on Quick Access Recorder (QAR) data in conjunction with the first-order approximation method. An improved Gaussian diffusion model for mobile point sources was employed to examine the diffusion characteristics of contaminants. Additionally, CFD calculation outcomes for various exhaust velocities and wind speeds were utilized to validate the trustworthiness of the improved Gaussian model. The discussion also encompasses the influence of diffusion time, wind direction, wind speed, temperature gradient, and particle deposition on the concentration distribution of contaminants. The findings indicated that the Gaussian diffusion model aligned with the results of the CFD calculations. The diffusion distribution of contaminants around airports varies over time and is significantly influenced by atmospheric environmental factors, including wind direction, wind speed, and atmospheric stability. Specifically, a change in wind direction from 0° to 45° caused a shift of approximately 1000 m in the contaminant’s center. An increase in wind speed from 3 m/s to 5 m/s led to a decrease in concentration by about 15%. Furthermore, a transition in atmospheric stability from category ‘a’ (very unstable) to ‘f’ (very stable) resulted in a two-order-of-magnitude increase in contaminant concentrations.

Dynamic model of arriving passengers based on dual value drive

In order to optimize the service process of integrated transportation hubs, the dynamic characteristics of the gathering behavior of arriving passengers were analyzed, and the transformation mechanism of the gathering behavior of arriving passengers in the time dimension of inverse gamma distribution, gamma distribution and chi-square distribution was studied. It was revealed that the gathering behavior was jointly affected by the flight operation time and the utility of the transfer mode, and was converted into the time value and utility value of the arriving passengers respectively. On this basis, a dual-value-driven dynamic model of arriving passengers was established based on the three distributions, and the output was the parameter-adjustable distribution of arriving passengers. The results show that the simulation output with adjustable parameters is consistent with the real distribution, which provides a method and basis for the accurate prediction of the distribution of arriving passengers at hub airports.

Stochastic optimal power flow framework with incorporation of wind turbines and solar PVs using improved liver cancer algorithm

AbstractThe present study introduces a nature inspired improved liver cancer algorithm (ILCA) for solving the non‐convex engineering optimization issues. The traditional LCA (t‐LCA) inspires from the conduct of liver tumours and integrates biological ethics during the optimization procedure. However, t‐LCA facing stagnation issues and may trap into local optima. To avoid such issues and provide the optimal solution, there are some modifications are implemented into the internal structure of t‐LCA based on Weibull flight operator, mutation‐based approach, quasi‐opposite‐based learning and gorilla troops exploitation‐based mechanisms to enhance the overall strength of the algorithm to obtain the global solution. For validation of ILCA, the non‐parametric and the statistical analysis are performed using benchmark standard functions. Moreover, ILCA is applied to resolve the stochastic renewable‐based (wind turbines + PVs) optimal power flow problem using a modified RER‐based IEEE 57‐bus. The objective of this work is to obtain the minimum predicted power losses and enhance the predicted voltage stability. By incorporation of renewable resources into the modified IEEE57‐bus network can help the system to reduce the power losses from 5.6622 to 3.8142 MW, while the voltage stability is enhanced from 0.1700 to 0.1164 p.u.

Resilience assessment and recovery of airport departure flights under severe weather

In order to ensure the overall performance of the airport under severe weather, scientifically evaluate the resilience of the airport's flight operations, improve flight recovery capabilities, and alleviate the impact of severe weather effectively. This article first gives the definition of airport departure flight operation. Starting from the performance of the airport departure flight operation system, it analyzes flight departure delay time, total departure delay time, departure flight normality rate and departure airport flight operation system comprehensive resilience index four indicators to evaluate the resilience changes of the system under severe weather conditions; put forward the concept of the performance recovery strategy of the airport departure flight operation system, and established the corresponding resilience model, and used the genetic algorithm to optimize the sequence of the delayed departure flights; finally, take the "721" heavy rain event in Beijing Capital International Airport in 2012 as an example to analyze the data, obtain the performance index and resilience index of the Capital Airport under the influence of heavy rain, and use the optimized departure flight schedule to compare and analyze the changes in airport departure flight operating system performance and resilience level. The results indicate that: under the influence of heavy rain, the comprehensive resilience index of the airport departure flight operation system decreased from 0.4573 to 0.0628 , and increased to 0.2223 after the rainstorm decreased; after the flight optimization and sequencing was carried out in the recovery stage of the airport departure flight operation performance, the total number of departure flights decreased. The delay time is reduced by 24.85%, the airport performance recovery speed is increased by 13.89%after optimization, and the minimum resilience index of the optimized airport departure flight operation system is increased by 13.38%, and under the same heavy rain, the system performance is given priority to restore to the initial state, indicating that the evaluation Indicators are valid.

Rapid Aircraft Wake Vortex Identification Model Based on Optimized Image Object Recognition Networks

Wake vortices generated by aircraft during near-ground operations have a significant impact on airport safety during takeoffs and landings. Identifying wake vortices in complex airspaces assists air traffic controllers in making informed decisions, ensuring the safety of aircraft operations at airports, and enhancing the intelligence level of air traffic control. Unlike traditional image recognition, identifying wake vortices using airborne LiDAR data demands a higher level of accuracy. This study proposes the IRSN-WAKE network by optimizing the Inception-ResNet-v2 network. To improve the model’s feature representation capability, we introduce the SE module into the Inception-ResNet-v2 network, which adaptively weights feature channels to enhance the network’s focus on key features. Additionally, we design and incorporate a noise suppression module to mitigate noise and enhance the robustness of feature extraction. Ablation experiments demonstrate that the introduction of the noise suppression module and the SE module significantly improves the performance of the IRSN-WAKE network in wake vortex identification tasks, achieving an accuracy rate of 98.60%. Comparative experimental results indicate that the IRSN-WAKE network has higher recognition accuracy and robustness compared to common recognition networks, achieving high-accuracy aircraft wake vortex identification and providing technical support for the safe operation of flights.

Reflection of Crew Resource Management (CRM) Trainings to Real-Life Field Practice in Air Passenger Transportation: A Qualitative Research

In the air transportation sector, flight safety stands as one of the foremost values. To ensure safety, airlines rely on Crew Resource Management (CRM) training. These training programs aim to foster coordination, communication, and awareness among all personnel involved in flight operations, encouraging rational responses to potential issues as a unified team. Naturally, as is the case with any training process, the practical applicability of CRM training alongside its theoretical structure is of paramount importance. Hence, this study, conducted within the framework of qualitative research, delves into the perspectives of cabin crew members regarding CRM training by evaluating its content, benefits, and real-life field applications. Employing interview-based methodology, qualitative data were gathered from 19 cabin crew participants through purposive sampling. The collected data underwent descriptive analysis, leading to certain conclusions. The findings illustrate that CRM training has a positive impact on emotional and knowledge management, communication skills, team collaboration, and a proactive approach in matters of safety. It underscores that CRM training significantly permeates into the realm of real-world applications—namely, flight operations. However, participants highlighted certain issues and emphasized the need for more practical, interactive, and participant-centered training methods for the enhancement of CRM training, aiming to address those concerns.

Reliability of selected components of the PZL M28 Bryza aircraft

The increasing number of flight operations, and the resulting increased likelihood of adverse situations, intensifies the need for research into aspects of aircraft reliability. This article is devoted to the reliability of selected assemblies of the PZL M28 Bryza aircraft, widely used in military and civil aviation. The article presents the research object, the PZL M28 Bryza aircraft, and the research methodology to assess the reliability of its key assemblies. The research focuses on analyzing the failure rate of such assemblies as the propulsion system, airframe, radioelectronic devices and on-board systems, using in-service data. The results provide a better understanding of the critical aspects affecting the reliability of the PZL M28 Bryza aircraft, which is important for improving the safety and efficiency of flight operations conducted.

Analytical Neural Network System for the Helicopter Turboshaft Engines Operating Modes Classification

A novel neural network-based analytical system has been developed for classifying helicopter turboshaft engine operating modes during flight operations. This system utilizes a neural network architecture comprising an input layer with three neurons, two fully connected hidden layers, and an output layer with two neurons. The proposed approach demonstrates an exceptional recognition accuracy of 0.997 (99.7%) across steady-state, unsteady, and transient operating modes of helicopter turboshaft engines. A new method for training this neural network has been introduced, employing forward propagation, loss calculation, backpropagation, and weight updates, enhanced by an adaptive learning rate and the cross-entropy function as the loss criterion. The method also incorporates a novel modified Smooth ReLU activation function for hidden layer neurons. This innovation led to a near-perfect accuracy in network training and reduced the loss to 0.025 (2.5%), highlighting the high quality and reliability of the neural network in classifying engine operating modes during flight. Furthermore, it has been empirically shown that the application of this neural network significantly reduces type I errors by a factor of 2.09 to 2.14 and type II errors by 2.05 to 2.21 times compared to traditional classifiers based on ART-1 and BAM networks. This advancement marks a substantial improvement in classification accuracy and error minimization for helicopter turboshaft engine operating modes.

The Importance of Weather Factors in the Resilience of Airport Flight Operations Based on Kolmogorov–Arnold Networks (KANs)

The Importance of Weather Factors in the Resilience of Airport Flight Operations Based on Kolmogorov–Arnold Networks (KANs)

Exploring the Human-Centric Interaction Paradigm: Augmented Reality-Assisted Head-Up Display Design for Collaborative Human-Machine Interface in Cockpit

Human-cybernetic interfaces (HCI) focuses on the integration and synergy of human-centric design, human factors (HFs) and economics, and human–machine interaction (HMI). The scrutiny of single pilot operations (SPO) approval from civil aviation authorities (CAAs) is subject to the aviation safety and pilot capability during exceptional handling and flight emergency. Airliners advocate the possibility from dual pilot operations (DPO) to SPO because of the financial considerations and captain shortage in the post-pandemic era. One could expect that SPO can only realise when reduced crew operations initiative is proven to be safe and airworthy. Public concerns pilot mental workload and potentially low situational awareness (SA) when handling overwhelmed multi-source information from the HMI in a cockpit, leading to degraded skills and flying performance. Advanced head up display (HUD) design, an integrated transparent display, can assist pilots in maintaining their optimal performance when SPO exercised. In this study, we are interested to design a SPO-favoured cockpit design with HUD and investigate the potential of integrating augmented reality (AR) and HUD in SPO in flight operations. An experiment involving 21 pilots was conducted to assess the effectiveness of AR-HUD on SA and workload in DPO, SPO, and SPO with AR-HUD scenarios. The Situation Awareness Present Assessment Method (SPAM) and a self-rated workload scale were used to evaluate SA and workload levels. Eye-tracking data was also collected to analyse behavioural performance. Results indicated that the AR-HUD scenario achieved optimal workload and SA levels and demonstrated superior performance in handling emergencies. Eye-tracking data revealed that AR-HUD facilitated a more equitable distribution of gaze between instruments and external view. In conclusion, the integration of AR in cockpit design introduces a human-centric interaction paradigm, contributing to aviation safety, SPO implementation and adaptive HMI design.

Climb Performance Prediction in High Drag Configuration Middle-Class Transportation Aircraft: An Ensemble Learning Approach

This study addresses the application of machine learning and artificial neural network models for predicting the climb speed of the C-130H military transport aircraft. Random Forest, Neural Network, and Ensemble models were developed to overcome limitations of traditional chart reading and interpolation methods. Models were trained on flight manual data, considering factors such as gross weight, pressure altitude, drag index, temperature deviation, and engine efficiency. Comparative analysis revealed the Ensemble approach, combining Random Forest and Neural Network techniques, provided the highest accuracy (R² ≈ 0.4532), followed by Random Forest (R² ≈ 0.4303) and Neural Network (R² ≈ 0.3765) models. All significantly outperformed the traditional Young Method (R² = -1.2673). Feature importance analysis identified pressure altitude, gross weight, and engine efficiency as critical factors influencing climb speed. The ensemble approach demonstrated more reliable and accurate results in predicting C-130H climb rates, reducing risks associated with single-model reliance. This research highlights the potential of machine learning in aircraft performance prediction, offering possibilities for improving pre-flight preparation, reducing workload, and enhancing flight safety. Implications for the aviation industry and future research directions are discussed, emphasizing the role of advanced predictive models in shaping future flight operations and aircraft performance management.

Real-time identification of precursors in commercial aviation using multiple-instance learning

This research pioneers the application of precursor concepts to preemptively identify and prevent aviation safety incidents using Machine Learning (ML). Airlines and governing organizations, such as the Federal Aviation Administration (FAA) in the United States, have been trying to prevent safety incidents during routine operations. However, this task is challenging due to the lack of timestep-wise event annotation in flights and the complexity involved in the timely identification of incidents prior to their occurrence. To address these issues, we propose a real-time precursor identification methodology combining Multiple-Instance Learning (MIL) and feature-based Knowledge Distillation (KD) learning. Our two-stage approach, involving deep MIL for labeling and a KD-based model for real-time warnings, demonstrates state-of-the-art performance and a time delay of 2.99ms using a dataset of 23,549 real flights. Further experiments using t-distributed Stochastic Neighbor Embedding (t-SNE) and occlusion method confirm our model’s transparency, enabling the generation of reliable quantitative precursor scores and facilitating reasoning about the causes of safety incidents at the parameter level. Additionally, statistical analysis of precursors reveals varying evolution times for different safety events, which indicates that pilots have at least 8 s to react after receiving a warning. In conclusion, our research provides a theoretical foundation and technical support for the next generation of online risk warning systems, enhancing flight safety and offering a pathway towards more intelligent and secure flight operations.

UAV path planning based on improved dung beetle algorithm with multiple strategy integration

This research presents a multi-strategy augmented dung beetle algorithm (CDBO) for UAV path planning in intricate 3D environments. Its main objective is to address the path planning challenge by formulating an integrated cost function model and environment representation that aligns the optimisation task with the navigation prerequisites and safety constraints of the UAV. Firstly, the algorithm initialises the particle population based on the Lévy flight principle to improve the species diversity and adequately search the solution space. Subsequently, an exponentially decreasing inertia weighting strategy is introduced to improve the convergence speed of the algorithm. In addition, the algorithm uses an adaptive [Formula: see text]-distribution with perturbed variation of the updated positions so that the algorithm avoids falling into a local optimum. Finally, to enhance the robustness of the system, the algorithm incorporates an elite retention strategy based on the Pareto principle. In this study, rigorous tests using 12 CEC2022 test functions and Wilcoxon rank sum test are conducted as benchmarks for the performance of the algorithm. The results show that the enhanced dung beetle algorithm outperforms the traditional algorithm, highlighting its efficiency and robustness. In addition, the efficacy of the enhanced dung beetle algorithm was validated in a variety of complex 3D environments, affirming its ability to generate optimal paths. This study further highlights the importance of effective algorithms in UAV path planning, addressing the key issue of optimising flight operations in intricate spatial scenarios.

Ship-based RPA operations for cetacean research in Antarctica: progress, opportunities and challenges

Data collection facilitated by remotely piloted aircraft (RPA) has proven to be revolutionary in many disciplines including for research in extreme environments. Here we assess current use and utility of small multirotor remotely piloted aircraft (RPAs) for the challenging role of facilitating ship-based cetacean research in Antarctica. While such aircraft are now used routinely in sheltered environments in and off Antarctica, a comprehensive literature review found that RPA-mediated cetacean research conducted from ships at sea and outside of the Antarctic Peninsula region was relatively uncommon. In order to determine the potential utility of ship-based multirotor RPA operations for cetacean research, we repeatedly deployed small RPAs during a multidisciplinary research voyage in maritime East Antarctica to collect scientific data contributing to an understanding of krill and krill predator interactions. RPA flight metrics (duration, height, length, speed, distance from ship, battery drainage, satellites acquired) were compared to ship underway environmental sampling data. At a mean duration of 12 minutes, these 139 RPA flights were relatively short yet adequate to achieve the science intended, namely a range of cetacean related data streams including photogrammetry, photo identification, behavioural observations and whale blow sampling in addition to water sampling and collection of general scenic imagery. RPA flight operations were constrained by wind speed but not by air temperature with flights undertaken throughout the full range of air temperatures experienced (down to –9.5°C) but not throughout the full range of wind speeds experienced. For a 12-minute flight duration, battery drainage was around 60% indicating that the RPAs were rarely pushed to their operational limit. There was little evidence that the cold impacted RPA lithium battery performance with estimated maximum flight time within approximately 10% of expected flight time for the RPA platforms most used. Whist small multirotor RPAs are rarely applied to cetacean related research in maritime East Antarctica, we demonstrate their value and potential to deliver data critical to address knowledge gaps that challenge the effective management of both krill and their predators.

AI-Driven FlightOps: Revolutionizing Airline Operations Control with Predictive Analytics and NLP

Abstract: Flight operations control centers (OCCs) are the nerve centers of airline operations, tasked with managing a wide range of activities such as flight scheduling, crew assignments, maintenance coordination, and dealing with disruptions. While traditional approaches have relied heavily on human decision-making and manual analysis of operational data, the increasing complexity of modern airline operations demands more efficient solutions. This paper presents FlightOps AI, a cutting-edge AIpowered system that integrates advanced natural language processing (NLP), deep neural networks (DNNs), reinforcement learning (RL), and computer vision to provide OCC staff with actionable insights and predictive capabilities. The system processes massive amounts of structured and unstructured data in real time, enabling smarter decision-making and automation of routine tasks. By adopting these technologies, FlightOps AI transforms the operational landscape, optimizing airline performance while minimizing human errors

Analysis on pulse rate variability for pilot workload assessment based on wearable sensor

AbstractThe workload levels of pilots directly affect their flight performance and the safety of the whole flight. To explore the real‐time workload of pilots in different flight phases (takeoff, cruise, and landing), this paper leveraged National Aeronautics and Space Administration Task Load Index (NASA‐TLX), a subjective evaluation scale, and PhotoPlethysmoGraphy (PPG) signals of 21 participants using a flight simulator and a wearable sensor. First, the workloads of pilots under different phases were explored by the NASA‐TLX scales; secondly, the pulse rate variability (PRV) features were selected by variance analysis and random forest importance evaluation; finally, the performances of the k‐nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) were compared for workload levels identification. It is shown that the workloads are ranked as follows: landing > takeoff > cruise. SDNN, CVCD, CVNNI, LF, TP, SD2, and SD2/SD1 were used as selected features with significant differences in different flight phases. In addition, machine learning models can effectively identify pilot workloads, and feature selection enhances the performance of both KNN and RF classifiers. The best identification of workload was achieved using the selected PRV features as inputs to the KNN classifier, with an average accuracy of 88.9%. Our results indicate that the KNN classifier and PRV features are suitable for identifying pilot workload. The pilot workload is highest during the landing phase, which provides a reference for flight safety management. The findings from this research could contribute to developing a robust pilot workload detection system and improve current flight operation safety regulations.