17 March 2021 in The Aeronautical Journal
In this study, prediction of aircraft Estimated Time of Arrival (ETA) is proposed using machine learning algorithms. Accurate prediction of ETA is important for management of delay and air traffic flow, runway assignment, gate assignment, collaborative decision making (CDM), coordination of ground personnel and equipment, and optimisation of arrival sequence etc. Machine learning is able to learn from experience and make predictions with weak assumptions or no assumptions at all. In the proposed approach, general flight information, trajectory data and weather data were obtained from different sources in various formats. Raw data were converted to tidy data and inserted into a relational database. To obtain the features for training the machine learning models, the data were explored, cleaned and transformed into convenient features. New features were also derived from the available data. Random forests and deep neural networks were used to train the machine learning models. Both models can predict the ETA with a mean absolute error (MAE) less than 6min after departure, and less than 3min after terminal manoeuvring area (TMA) entrance. Additionally, a web application was developed to dynamically predict the ETA using proposed models.
6 October 2021 in Transportation Research Record: Journal of the Transportation Research Board
The Point Merge System (PMS) is one of the approaches that offer systematic sequencing in terminal maneuver area (TMA) which was developed at the EUROCONTROL Experimental Centre in 2006. The Parallel-Point Merge System (P-PMS) is used in TMAs with high air traffic density as a combination of two PMSs with opposite directions. However, it is difficult to optimize arrival sequences in such a system because of the additional separation requirements at common points as well as at merge points. Additionally, it is important to operate at maximum efficiency with parallel runway systems when using the runway reassignment process. Reassigning arriving/departing aircraft to a different runway, however, can significantly affect delays and fuel consumption depending on the difference in taxi-in/out times. This effect is overlooked in most aircraft sequencing and scheduling optimization studies with PMS. This study proposes single and multi-objective programming models for a TMA with P-PMS to minimize total fuel consumption, total flight time, and total delay including taxi-in/out times. The models were implemented on the current layout of Istanbul Airport having two PMS with two merge points, three common points, and five parallel runways. The considered traffic scenarios included both dependent and independent runway operations for mixed arrival–departure sequences. The results revealed that arrival–departure sequencing considering taxi-in/out times resulted in shorter delays up to 77.5% and low level of fuel consumption up to 8.7%.
21 July 2022 in Transportation Research Record: Journal of the Transportation Research Board
Aircraft landing is a critical operation for both terminal airspaces and airports. This study presents a multi-objective optimization model for this problem and aims to minimize the total flight time and emission value. A point merge system and vector maneuver techniques are implemented in aircraft to regulate air traffic. The model uses the augmented è-constraint method to reveal objective function relationships. To test the performance of the mathematical model, several realistic scenarios are generated and solved by GAMS/CPLEX solver. The algorithm can obtain Pareto-optimal solutions for each air traffic situation. In addition, noticeable reductions were observed in most of the Pareto-optimal solutions in emission values. The results showed that total emissions value and flight time were reduced up to 4.69% and 0.92%, respectively, compared with the first-come-first-served approach.
3 November 2022 in Aircraft Engineering and Aerospace Technology
Purpose: The purpose of this paper is to create a flight route optimization for all flights that aims to minimize the total cost consists of fuel cost, ground delay cost and air delay cost over the fixed route and free route airspaces. Design/methodology/approach: Efficient usage of current available airspace capacity becomes more and more important with the increasing flight demands. The efficient capacity usage of an airspace is generally in contradiction to optimum flight efficiency of a single flight. It can only be achieved with the holistic approach that focusing all flights over mixed airspaces and their routes instead of single flight route optimization for a single airspace. In the scope of this paper, optimization methods were developed to find the best route planning for all flights considering the benefits of all flights not only a single flight. This paper is searching for an optimization to reduce the total cost for all flights in mixed airspaces. With the developed optimization models, the determination of conflict-free optimum routes and delay amounts was achieved with airway capacity and separation minimum constraints in mixed airspaces. The mathematical model and the simulated annealing method were developed for these purposes. Findings: The total cost values for flights were minimized by both developed mathematical model and simulated annealing algorithm. With the mathematical model, a reduction in total route length of 4.13% and a reduction in fuel consumption of 3.95% was achieved in a mixed airspace. The optimization algorithm with simulated annealing has also 3.11% flight distance saving and 3.03% fuel consumption enhancement. Research limitations/implications: Although the wind condition can change the fuel consumption and flight durations, the paper does not include the wind condition effects. If the wind condition effect is considered, the shortest route may not always cause the least fuel consumption especially under the head wind condition. Practical implications: The results of this paper show that a flight route optimization as a holistic approach considering the all flight demand information enhances the fuel consumption and flight duration. Because of this reason, the developed optimization model can be effectively used to minimize the fuel consumption and reduce the exhaust emissions of aircraft. Originality/value: This paper develops the mathematical model and simulated annealing algorithm for the optimization of flight route over the mixed airspaces that compose of fixed and free route airspaces. Each model offers the best available and conflict-free route plan and if necessary required delay amounts for each demanded flight under the airspace capacity, airspace route structure and used separation minimum for each airspace.
20 July 2023 in The Aeronautical Journal
Global air traffic demand has shown rapid growth for the last three decades. This growth led to more delays and congestion within terminal manoeuvring areas (TMAs) around major airports. The efficient use of airport capacities through the careful planning of air traffic flows is imperative to overcome these problems. In this study, a mixed-integer nonlinear programming (MINLP) model with a multi-objective approach was developed to solve the aircraft sequencing and scheduling problem for mixed runway operations within the TMAs. The model contains fuel cost functions based on airspeed, altitude, bank angle, and the aerodynamic characteristics of the aircraft. The optimisation problem was solved by using the $\varepsilon$ -constraint method where total delay and total fuel functions were simultaneously optimised. We tested the model with different scenarios generated based on the real traffic data of Istanbul Sabiha Gökçen Airport. The results revealed that the average total delay and average total fuel were reduced by 26.4% and 6.7%, respectively.
31 December 2020 in Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi
İniş uçaklarının sıralanması problemi, tahmini operasyon zamanlarının bulunduğu bir uçak kümesi için belirli kısıtlar altında gerekli emniyet ayırmalarının sağlanarak sıralamanın yapılmasıdır. Yapay Bağışıklık Sistemi (YBS) anormallik tespiti, bilgisayar ve ağ güvenliği, çizelgeleme, eniyileme, sınıflandırma, veri madenciliği gibi birçok alanda kullanılan ve doğal bağışıklık sisteminden esinlenerek oluşturulan bir tekniktir. Bu çalışmada, tek piste iniş yapmayı planlayan uçakların sıralanması YBS algoritmalarından olan Klonal Seçim Algoritması (KSA) kullanılarak yapılmıştır. Amaç fonksiyonu olarak operasyonların tamamlanma zamanının enküçüklenmesi alınmıştır. Ayrıca çalışmada uçakların ilk gelen ilk hizmet alır (FCFS) prensibi ile belirlenen sıralamalarına göre sınırlı sayıda yer değiştirmelerine (CPS) izin verilmiştir. Farklı sayıda ve kategoride uçaklardan oluşan senaryolar için algoritma test edilmiştir. Algoritma ile elde edilen çözümler ve matematiksel model ile edilen çözümler kıyaslanmıştır ve çözüm süreleri de paylaşılmıştır.