Capital International Airport Research Articles

Airport Express Passenger Flow Prediction Based on “Time-Feature” Cooperative Attention

Accurate prediction of airport express passenger flow is the basis for realizing intelligent, refined and efficient management and control of the airport rail transit system, and is of great significance to improving airport service levels and operational efficiency. Due to the numerous influencing factors that overlap with each other, and the complex mechanism of factors affecting passenger flow timing, accurate prediction of airport express passenger flow is extremely challenging. This paper proposes an airport express rail passenger flow prediction model based on the "time-feature" collaborative attention mechanism, which achieves accurate capture of the impact of multi-dimensional factors on airport express rail passenger flow in different time series. Experiments were conducted based on actual passenger flow data of the Beijing Capital International Airport Express Rail Link, and the results showed the effectiveness of the proposed method.

Machine Learning-Based Prediction of Aircraft Taxi-In Time: A Case Study at Beijing Capital International Airport

Accurate prediction of flight taxi-in time has a significant meaning in allocating aircraft support resources reasonably and improving airport surface movement efficiency. It can effectively overcome the deficiency of extensive aircraft arrival time prediction in major airports currently. Taking Beijing Capital International Airport as the research object, we analyzed the influence factors of taxi-in time and constructed the feature set; Then we applied linear regression, K-nearest neighbor, support vector regression, decision tree, random forest and gradient boosting regression tree, which were widely used in the prediction of taxi-out time, to predict the taxi-in time. The results show that the prediction accuracy of the six machine learning models is over 90%within \pm3minutes, it means that the construction of feature set and the selection of models are effective; The gradient boosting regression tree model has the best performance based on the prediction results and model fitting evaluation results; The surface traffic flow features have the largest contribution to the prediction model, and the newly introduced cross-regional feature has more contribution to the prediction model than most traditional features, according to the prediction results of gradient boosting regression tree.

Flight Schedule Optimization Considering Fine-Grained Configuration of Slot Coordination Parameters

In response to the rapid growth of air passenger and cargo transportation services and the sharp increase in congestion at various airports, it is necessary to optimize the allocation of flight schedules. On the basis of reducing the total airport delay time and ensuring the total deviation of flight schedules applied by airlines, it is necessary to consider finely configuring flight schedules with slot coordination parameters, introducing a 5 min slot coordination parameter, and optimizing airport flight schedules in different time periods. This article considers factors such as flight schedule uniqueness, corridor flow restrictions, and time adjustment range limitations to establish a three-objective flight-schedule refinement configuration model, which is solved using the NSGA-II algorithm based on the entropy weight method. Taking Beijing Capital International Airport as an example, the optimized results show that the total flight delay was reduced from 4130 min to 1142 min, and the original delay of 389 flights was reduced to 283 flights. Therefore, flight schedule optimization considering the fine-grained configuration of slot coordination parameters can effectively reduce airport delays, fully utilize time resources, and reduce waste of time slot resources.

Enhancing airline connectivity: An optimisation approach for flight scheduling in multi-hub networks with bank structures

While one salient characteristic of hub airports lies in connecting passengers, the full-service airlines in North America concentrate their networks spatially over a number of hubs. Having witnessed the emerging multi-hub network, this paper investigated the flight scheduling problem under a multi-hub configuration, taking a well-defined bank structure and airport operational restrictions into account. An integrated non-convex Mixed-Integer Nonlinear Programming (MINLP) approach was proposed to enhance airline connectivity, considering different combinations of traffic flow direction and connecting times. To verify the scalability and effectiveness of the proposed model, a comprehensive case study has been undertaken with real-world scheduling data from Air China, which was solved by a novel problem-specific Selective Simulated Annealing (SSA) algorithm. Substantial improvements were achieved without sacrificing the scheduling efficiency. Precisely, the programme adjusted the flights during a typical operational day in a timely manner. The post-optimisation outcomes have witnessed its effectiveness with a 17.97%, 17.06%, 22.41% and 53.86% increase in airline connectivity at its four major hub airports (Chengdu Shangliu, Beijing Capital, Shanghai Pudong and Hongqiao) in China, respectively. A clear pattern of the bank structure also confirms its positive impact on airline connectivity under the multi-hub network configuration. Lastly, a comparative analysis for the distribution of all feasible connections further highlights the critical challenge concerning the role of the hubs in a multi-hub network. More specifically, Air China’s multi-hub network systematically performs better on Domestic-International routes, due to flight schedule, frequencies, geographical placements and detours. Among the four hub airports, Beijing Capital International Airport stands out as a dominant one, which implies its potential to serve as a robust international hub airport.

Multi-Task Dynamic Spatio-Temporal Graph Attention Network: A Variable Taxi Time Prediction Model for Airport Surface Operation

Variable taxi time prediction is the core of the Airport Collaborative Decision Making (A-CDM) system. An accurate taxi time prediction contributes to enhancing airport operational efficiency, safety and predictability. The deep dynamic spatio-temporal correlation inherent in airport traffic data is critical for taxi time prediction. However, existing machine learning (deep learning) methods have been unable to thoroughly exploit these correlations. To address this issue, we propose a deep learning-based model called the multi-task dynamic spatio-temporal graph attention network (MT-DSTGAN). Our model also predicts future entire airport traffic flow and taxiing segment traffic flow as auxiliary tasks, with the goal of enhancing the accuracy of aircrafts’ taxi time prediction. The proposed MT-DSTGAN model is implemented and assessed through a case study of Beijing Capital International Airport with a real-world dataset. The advantage of the proposed model, which shows better performance in various evaluation metrics, is demonstrated in a comparative study with other baseline works. In summary, the proposed MT-DSTGAN exhibits promising capabilities in perceiving the dynamic changes in the taxiing process of aircraft and demonstrates the ability to capture complex spatio-temporal correlations in airport traffic data.

Proactive real-time scheduling method for apron service vehicles based on mixed strategies

Due to the uncertainty of flight arrival and departure times, the scheduling of apron vehicles in airports has always faced significant challenges. This paper considers the real-time scheduling problem of apron vehicles under dynamic requests, establishing a new parking rule to allow apron vehicles to dynamically adjust parking areas. In order to maximize service level and minimize operational costs, this paper proposes a proactive real-time scheduling method for apron vehicles based on mixed strategies. Through strategies such as flexible waiting and dynamic relocation, this paper responds proactively to future requests. The proposed strategies are validated through implementation in large and medium-sized scale cases at Beijing Capital International Airport and Nanjing Lukou International Airport in China, respectively. Numerical results indicate that the proposed scheduling method is effective and applicable. The flexible waiting strategy significantly reduces vehicle travel distance, while the dynamic relocation strategy is significant in shortening request response times.

Airport–Industry–City Integrated Accessibility Using Big Data Case View from Beijing Capital International Airport

Big data technology can help collect, analyze, and process large-scale transportation, population, economic, and other relevant data, which has been proven to be more accurate in assessing the accessibility of different areas. An aerotropolis is composed of the airport as the core, airport-related industry as the link, and the airport-oriented city as the carrier, airport–industry–city (AIC) system. To better grasp the development quality of aerotropolises, an AIC integrated accessibility measurement model is constructed comprising the accessibility of airport route network, airport ground traffic, and airport information coverage. Using big data of traffic, network and information from 2001 to 2020, we apply the model to the Beijing Capital International Airport (PEK) aerotropolis. The results show that PEK’s AIC connectivity exceeded 1 for the first time in 2008, reaching 1.0890, which the range of the scale is [0, +∞]. It fluctuated significantly from 2017 to 2020, showing a double life-cycle S-curve. So, the PEK aerotropolis’ development can be divided into four stages: germination (before 2001), high-speed development (2002–2011), gradual formation (2012–2019), and “seeking change in the process of change” (after 2020). Our model offers a new tool for identifying the development stages of aerotropolises and for improving connectivity—namely, improving the accessibility of the airport node network, the layout of airside-preferring enterprises, and the network structure of new Internet-based airport cities. Obviously, the AIC integrated accessibility can indeed guide the scientific development of an aerotropolis.

Dynamic departure pushback control at airports: Part A—Linear penalty‐based algorithms and policies

AbstractAirport surface congestion can lead to significantly long taxi‐out times, thus resulting in increased fuel‐burn costs as well as excessive emissions of greenhouse gases. To curtail this undesirable syndrome, in this article, we propose a new penalty‐based dynamic departure pushback control (PDPC) strategy, which employs a linear penalty function dependent not only on the taxiway queue limit but also on the current queue length to ration the pushback frequency at airports, and trades taxiway queueing times with gate‐hold delays to minimize the total operational cost (fuel‐burn and gate‐hold costs). Using data from Beijing Capital International (PEK) airport, four different departure pushback control policies, namely: (i) no‐control (baseline case); (ii) traditional ‐control; (iii) PDPC with a constant taxiway limit; and (iv) PDPC with varying taxiway limits; are compared. Detailed Monte Carlo simulations, which showcase the sensitivity of the total cost function to various problem parameters are presented, and our results indicate that deploying the PDPC policy results in a 42% reduction in total operational costs and a 68% reduction in fuel‐burn (kg) as compared to the baseline case. To analytically reinforce these simulation results, an iterative Markov chain‐based optimization algorithm is also developed to estimate the optimal values of the pushback rate and taxiway queue limit that minimize the total cost function. Such an analytical framework is very useful in the absence of reliable airport data as it only requires estimates of the historical pushback request rates and service times at the taxiway, while yet retaining the capability to closely mirror the simulation results. Our Monte Carlo simulations as well as the Markov chain optimization model validate the strength and impact of the proposed PDPC policy, and demonstrate its practical efficacy in reducing airport surface congestion when applied using data from PEK airport.

Long-Term SAR Data Analysis for Subsidence Monitoring and Correlation Study at Beijing Capital Airport

Land subsidence, resulting from natural or human activities, is a global environmental geological disaster. The Interferometric Synthetic Aperture Radar (InSAR) time-series analysis technique offers high spatial and continuous temporal resolution, providing data and a foundation for investigating regional land subsidence and its evolution mechanism. Beijing Capital International Airport (BCIA) has experienced uneven land subsidence since 1935, together with severe fissures significantly affecting its normal operations. In this study, the time-series InSAR method was successfully applied to monitor the gradual increase in uneven local subsidence and ground fissures activity at BCIA from June 2003 to March 2023. Initially, ENVISAT-ASAR, Cosmo-SkyMed, and Sentinel-1 data were processed by time-series InSAR techniques to generate deformation rate maps and time series for the airport area. Subsequently, a comparison was made between the displacement time series from InSAR and ground leveling measurements to assess the accuracy of InSAR-derived measurements. Through a comprehensive analysis of the distribution characteristics of land subsidence at the airport, a long-standing ground fault was located within the airport was identified. A preliminary discussion on the development status of this ground fissure was carried out based on the visual interpretation of optical images. Lastly, the inducing factors and evolutionary conditions of land subsidence were discussed. This case demonstrates the applicability of InSAR technology in identifying and monitoring geological processes such as land subsidence and ground fissure activities. It provides a scientific approach to exploring and studying the causes and formation mechanisms of land subsidence and ground fissures in the Beijing Capital Airport area.

Field‐tested signal controller to mitigate spillover using trajectory data

AbstractTrajectory data from connected vehicles (CVs) provide a continuous and reliable means of obtaining information that can be leveraged to optimize traffic signals. This paper proposes a real‐time traffic signal control method using CV trajectory data as the sole input. The primary goal of the proposed signal control method is to prevent queue spillover, which may significantly decrease the traffic efficiency on urban networks and induce high delays to the travelers. The proposed method formulates the signal control problem via a linear quadratic optimization model, considering the constraints related to the duration and variability of green lights in practical traffic signal control systems. Compared to conventional max‐pressure‐based methods, the optimization model offers enhanced efficiency in handling these constraints, making it highly suitable for real‐life implementation. The proposed method has undergone testing in both simulated environments and real‐world applications. In the simulation experiments, the proposed method has been demonstrated to effectively reduce spillover risks and outperform a conventional max‐pressure‐based approach even when the CV penetration rate is as low as 5%. In the real‐world experiment, the proposed method had been tested in a traffic network around the Beijing Capital International Airport for several months. The severity of spillovers, which was represented by two performance indicators, had been significantly reduced after implementing the proposed method.

Quick taxi route assignment via real-time intersection state prediction with a spatial-temporal graph neural network

The optimization of airport taxi routes is crucial for increasing airport capacity and resource utilization. However, existing methods for preplanning schedules are often time-consuming. To address this, we propose a new approach that reduces the problem scale by predicting the intersection state in the airport taxiway network using a spatial–temporal graph neural network (STGNN) that incorporates a graph convolutional network and recursive neural network with an attention mechanism. A mixed-integer linear problem (MILP) model for taxi route assignment is established based on a subgraph obtained from the prediction results. Our approach treats time as a continuous variable without discretization, allowing aircraft to stop and wait at any node to avoid conflict. Finally, we demonstrate the effectiveness of our approach by testing it on actual operation data from Beijing Capital International Airport (PEK). Compared with several baseline algorithms, the method developed in this study has a significant impact on improving airport transportation efficiency. The results indicate that our method not only produces high-quality solutions but also operates in real time.

A data-driven approach to the strength evaluation of airfield rigid pavement using in situ instrumentation data

The pavement strength evaluation plays a vital role in the airfield operation and management. Traditionally, the evaluation has relied on the Heavy Weight Deflectometer (HWD) test. This method encounters challenges, including interruptions in airfield operations, limited coverage of inspection locations, extensive time required for data collection and analysis. This paper introduces a methodology for evaluating the structural strength of airfield rigid pavement using instrumentation data, aiming to enhance both the efficiency and accuracy of such evaluations. Strain responses, collected using the instrumentation system at Chengdu Tianfu International Airport under HWD loading, are extensively analyzed. A three-dimensional numerical simulation model encompassing nine slabs was formulated, leveraging finite element analysis to replicate pavement working conditions that were not covered by the instrumentation samples. The merging of the instrumentation and simulation samples resulted in a comprehensive database for diverse pavement working conditions. This facilitated the establishment of a data-driven framework for the fusion of mechanical models. From this database, a neural network mapping model was developed to correlate the instrumentation data with the pavement slab modulus and subgrade reaction modulus. These correlations were subsequently employed for the calculation of the pavement classification number (PCN). The primary advantages of the proposed method include real-time monitoring and evaluation of airfield rigid pavement strength without interference to airfield operations. The developed method has been deployed at seven hub airports in China, encompassing Shanghai Pudong International Airport (PVG), Beijing Capital International Airport (PEK), and Chengdu Tianfu International Airport (TFU), etc.

Emerging and legacy perfluoroalkyl and polyfluoroalkyl substances (PFAS) in surface water around three international airports in China

Perfluoroalkyl and polyfluoroalkyl substances (PFAS) are a large category of crucial environmental contaminants of global concerns. There are limited data on PFAS in surface water around international airports in China. The present study investigated the concentrations, distributions, and sources of emerging and legacy PFAS in surface waters around Beijing Capital International Airport (BC), Shanghai Pudong International Airport (SP), and Guangzhou Baiyun International Airport (GB) in China. Twenty-seven target compounds were quantified. The Σ27PFAS concentrations ranged from 19.0 to 62.8 ng/L (mean 36.1 ng/L) in BC, 25.6–342 ng/L (mean 76.0 ng/L) in SP, 7.35–72.7 ng/L (mean 21.6 ng/L) in GB. The dominant compound was perfluorooctanoic acid (PFOA), which accounted for an average of 27% (5%–65%) of the Σ27PFAS concentrations. The alternatives with –C6F12- group had detection frequencies ranging from 72% to 100%. The partition coefficient results indicate that the longer chain PFAS (C > 8) tend to be more distributed in the particle phase. Fifty suspect and nontarget PFAS were identified. In GB, 44 PFAS were identified, more than SP of 39 and BC of 38. An ultra short-chain (C = 2) precursor, N-methylperfluoroethanesulfonamido acetic acid (MeFEtSAA), was identified and semi-quantified. Domestic wastewater discharges might be the main sources around BC, while industrial and aviation activities might be the main sources around SP and GB.

Optimising the flight turnaround schedules: An improved sliding time windows approach based on MILP and CP models

This paper addresses the scheduling of turnaround operations for flights in civil airports. A series of complex issues, including the heterogeneity of flights and operators, resource constraints on turnaround operations, temporal restrictions and alternative service patterns of turnaround operations, multi-functional operators, multi-trip services and multi-to-one services, are jointedly considered. We propose two formulations for the problem: A mixed-integer linear programming (MILP) model, and a constraint programming (CP) model. For comparison purpose, we also present simplified versions of the models in which the multi-to-one services are prohibited. The uncertainty are incorporated by modifying the models based on robust optimisation principles. The numerical experiments are conducted using instances generated based on real flight data at Beijing Capital International Airport. It is shown that under current computer configurations, small-scale instances can be solved to the optimum by CPLEX within an acceptable period of time using the MILP model, while moderate-scale instances can be tackled using the CP formulation. To cope with large-scale instances, an improved sliding time window method is developed, and evaluated with different parameter combinations, showing that practical turnaround schedule would best be updated two to three times per hour. Besides, after a case study in both static and stochastic scenarios, we illustrate that multi-to-one services should be specifically considered and highlighted, for their existence could significantly improve the operator utilisation rate and reduce flight delay, especially when the uncertainties are taken into account.

A Methodology for Allocating Incremental Resources in Single-Airport Time Slots

Air carriers shall not readily relinquish their held flight slots. In cases where the historical flight slot pool cannot be easily altered, a pressing need arises for an allocation method that can efficiently utilize the incremental resources of these time slots. This paper presents an integer planning model to address the efficient allocation of incremental airport time slot resources. The model considers the capacity of key resource nodes and flight waveforms as constraints to maximize the total incremental slots. Moreover, it considers the adaptation of strategic and tactical optimization. After conducting a case study using Beijing Capital International Airport for verification, the proposed model effectively reduces potential operational delays by 66.27% while adding 366 to 397-time slots. Notably, the model demonstrates remarkable delay reduction capabilities and can serve as a valuable decision-support tool for the incremental allocation of time slots.

Improving the performance of airport shuttle through demand-responsive service with dynamic fare strategy considering mixed demand

Given the booming aviation industry, airport landside transportation systems have been facing severe challenges. With the development of communication technologies, the conventional airport shuttle—one of the major components of landside transportation—is expected to operate more flexibly to meet diverse travel demands. Therefore, this study aims to explore how the demand-responsive airport shuttle might work and support the decision-making for airport management in the aspects of flexible stop location and diversified service design. The mixed demand, consisting of reservation and real-time demand, is considered. First, we propose an improved clustering method that takes into account the spatio-temporal similarity of reservation demand to identify demand-driven fixed stops, and the batch response strategy for the selection of demand-responsive stops meeting real-time demand. The dynamic fare strategy is further presented to discuss the effects of fare incentives for spatio-temporal service adjustments. Moreover, we develop a two-stage service optimization approach considering the cost and benefits from the perspective of passengers and operators. Trips from the Huitian area (located to the north of Beijing, China) and Beijing Capital International Airport are used as case studies. The results show that the design of 5 routes with 14 fixed stops effectively covers the areas of dense travel demand. Without affecting the on-time arrival of the demand-responsive shuttle at the airport, the load factor for all arriving vehicles exceeds 80% when demand-responsive stops are also served. The study thus provides deep insights into the service design of the demand-responsive airport shuttle and the development of differentiated operating strategies.

A Long-Term Traffic Flow Prediction Model Based on Variational Mode Decomposition and Auto-Correlation Mechanism

Traffic flow forecasting, as an integral part of intelligent transportation systems, plays a critical part in traffic planning. Previous studies have primarily focused on short-term traffic flow prediction, paying insufficient attention to long-term prediction. In this study, we propose a hybrid model that utilizes variational mode decomposition (VMD) and the auto-correlation mechanism for long-term prediction. In view of the periodic and stochastic characteristics of traffic flow, VMD is able to decompose the data into intrinsic mode functions with different frequencies, which in turn helps the model extract the internal features of the data and better capture the changes of traffic flow data in the cycle. Additionally, we improve the residual structure by adding a convolutional layer to propose a correction module and use it together with the auto-correlation mechanism to jointly build an encoder and decoder to extract features from different data components (intrinsic mode functions) and fuse the extracted features for output. To meet the requirements of long-term forecasting, we set the traffic flow forecast length to 4 levels: 96, 192, 336, and 720. We validated our model using the departure statistics dataset of a taxi parking lot at Beijing Capital International Airport and achieved the best prediction performance in terms of mean squared error and mean absolute error, compared to the baseline model.

Simulation-based optimization of bus departure scheme for large airport considering passenger evacuation

Purpose This study aims to optimize the traffic capacity allocation to solve the problem of low share of public transit in the landside system so as to get rid of the congestion trouble in landside traffic. The optimal timetable for airport buses can be searched by changing the departure interval of each line and evaluating the corresponding performance continuously. Design/methodology/approach This paper constructs a simulation model based on the real-world situation in Beijing Capital International Airport (BCIA), which simulates the whole process of airport bus schedules and analyzes the connections among multiple steps for transferring. The evaluation system is constructed by considering the benefits of passengers, airports and companies comprehensively. The optimal timetable for airport buses can be searched by changing the departure interval of each line and evaluating the corresponding performance continuously. Findings According to the experimental results, an excellent evacuation effect can only be achieved when the majority of departure intervals of airport buses are shortened to 50% of their original values, and some busy routes such as the Beijing Station line are supposed to be reduced to one-third of their original fixed intervals. As the airport bus passenger flow presents an obviously periodic variation over days, the timetable of the airport bus is supposed to be redesigned every day. A flexible bus timetable can not only meet the dynamic passenger flow but also enhance the attractiveness of public transit. Originality/value This paper constructs a simulation model based on the real-world situation in BCIA, which can not only model the complex scenes in the whole process of airport bus schedules but also reflect the intricate interaction between transferring passengers and vehicles caused by dense streamlines.

Airport Capacity Prediction With Multisource Features: A Temporal Deep Learning Approach

Accurate airport capacity estimation is crucial for the secure and orderly operation of the aviation system. However, such estimation is a non-trivial task as capacity depends on various meteorological and operational features. The complex coupling characteristics among these multi-source features have proved to be challenging for most of the traditional regression models. Recently, enhanced by its excellent ability to mine nonlinear relationships, the machine learning methods trigger widely applications. However, due to the imbalance of features scatter and the neglect of temporal dependences in aviation systems, existing machine learning methods for airport capacity prediction still have room for improvement. In light of these, this paper presents a novel airport capacity prediction method based on the multi-channel fusion Transformer model (MF-Transformer). Besides the commonly used aviation features, we unprecedentedly harness the power of the high-dimensional meteorological feature for accurate prediction. As to the model, we construct a multi-channel feature fusion structure, which includes a three-channel network for multi-source features extraction and an attention-based feature fusion module between channels. In each channel, the Transformer-based model is utilized to capture the temporal dependences of features. We conduct experiments on the capacity prediction tasks of the Beijing Capital International Airport which is the largest airport in China and verify that the proposed MF-Transformer outperforms benchmarks under different prediction horizons.

Cloud Occlusion Probability Calculation Jointly Using Himawari-8 and CloudSat Satellite Data

Cloud occlusion is an important factor affecting flight safety and scientific observation. The calculation of Cloud Occlusion Probability (COP) is significant for the planning of the flight time and route of aircraft. Based on Himawari-8 and CloudSat satellite data, we propose a method to calculate the COP. The COP statistics were carried out on different distances in 12 directions 6 km above Beijing Capital International Airport (BCIA), at different heights and directions in the Haiyang aerostat production base, and at different times and seasons in Mount Qomolangma. It was found that the COP going in the southern direction from BCIA was greater than that in the northern direction by 0.67–3.12%, which is consistent with the climate conditions of Beijing. In Haiyang, the COP for several seasons in the direction of land was higher than in the direction of the ocean. The maximum COP for the 6 km altitude is 29.63% (summer) and the minimum COP is 7.59% (winter). The aerostat flight test can be conducted in the morning of winter and the direction of the ocean. The best scientific observation time for Mount Qomolangma is between 02:00 and 05:00 UTC in spring. With the increase in altitude, the COP gradually decreases. The research in this paper provides essential support for flight planning.