CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Enhancing Airspace Capacity Assessment in CAPAN Studies by Clustering ATC Complexity Regimes

Estudiants que han llegit aquest projecte:

Director/a: MELGOSA FARRÉS, MARC

Departament: FIS

Títol: Enhancing Airspace Capacity Assessment in CAPAN Studies by Clustering ATC Complexity Regimes

Data inici oferta: 26-02-2026     Data finalització oferta: 26-02-2026


Estudis d'assignació del projecte:
    GR ENG SIST AEROESP
Tipus: Individual
 
Lloc de realització: ERASMUS
 
Segon director/a extern: Raffaele Russo / Christian Lorenzo
 
Paraules clau:
Capacity, Air Traffic Management, Workload, Fast Time Simulations, Clustering
 
Descripció del contingut i pla d'activitats:
 
Overview (resum en anglès):
Airspace capacity refers to the maximum number of aircraft that an airspace volume can safely accommodate without causing excessive controller workload. The EUROCONTROL CAPAN methodology assesses airspace capacity by employing Fast Time Simulations, and relating air traffic demand to controller workload through regression analysis.

The problem arises because relying on a single traffic count to define capacity implicitly assumes a stable workload-demand relationship. In practice, controller workload is strongly affected by Air Traffic Control (ATC) complexity, and a similar number of aircraft may generate different workload levels depending on the operational situation. This introduces dispersion in the workload-demand regression, increasing uncertainty in the estimated capacity.

This thesis proposes a data-driven methodology to improve the interpretation and granularity of capacity assessments performed in CAPAN studies. ATC complexity is modeled through the ratio between controller working time and air traffic demand. Sectors with heterogeneous ATC complexity are identified, and clustering is applied using the K-means algorithm to extract different complexity regimes, enabling a separate capacity assessment for each cluster. In addition, new reporting tools are proposed leveraging clustering information, including hourly capacity profiles and a visualization that links complexity regimes to specific subsets of demand.

The methodology was validated using four previous CAPAN studies covering both Area Control Centre (ACC) and Terminal Manoeuvring Area (TMA) environments. Results show that clustering is particularly relevant in occupancy assessments and effective in both environments, especially in TMAs. Whenever clustering was applied, regression error was consistently reduced, indicating lower uncertainty in the resulting capacity assessment.

Overall, this thesis incorporates ATC complexity into capacity assessments to provide more detailed results, while remaining compatible with the CAPAN methodology and current operational practices, which rely mostly on traffic counts.

© CBLTIC Campus del Baix Llobregat - UPC