Multi-objective optimization approach for air traffic flow management

Rabie Fadil,Badr Abou El Majd,Hicham Rahil,Naima Kaabouch,Hassan El Ghazi,B Abou El Madj,M Bouya

doi:10.1051/matecconf/201710500005

Abstract

Air traffic has long been a generally high growth sector and all forecasts indicate that this trend will continue at a similar pace for the next twenty years. The regular traffic demand growth has led to congestionat airports and in space.In this paper, we will create in first a probabilistic model which describes the uncertainty of the aircraft’s trajectory,and its presence in a sector during a time interval. We define as result, amulti-objective optimization problem whose objective functions are the expected cost of delay and the expected cost of congestion.Then we use the Non-dominated Sorting Genetic Algorithm (NSGA-II) to solve an instance included 21 flights and 1 sector, and is able to provide a good approximation of the Pareto front.The decision-making stage was then performed with the aid of data clustering techniques to reduce the sizeof the Pareto-optimal set and obtain a smaller representation of the multi-objective design space, there by making it easier for the decision-maker to find satisfactory and meaningful trade-offs, and to select a preferred final design solution.

Highlights

The Operational Research community has studied many variants of the air traffic flow management problem since the beginning of the 90s
One of the first formulations of air traffic flow management is the ground holding problem, which minimizes the sum of airborne and ground delay costs when the demand for the runways exceeds the allowed capacities
This paper has presented a probabilistic model to handle the propagation of the uncertainty from the trajectories to the sectors

Summary

INTRODUCTION

The Operational Research community has studied many variants of the air traffic flow management problem since the beginning of the 90s. One of the first formulations of air traffic flow management is the ground holding problem, which minimizes the sum of airborne and ground delay costs when the demand for the runways exceeds the allowed capacities. The Multi-Airport Ground Holding Problem was addressed by [3] This does not take into account the sector capacities,rerouting and speed changes. Constraint programming was used by [10] and [11] The former solves the slot allocation problem with sector capacity constraints and the former minimizes an air traffic complexity metric for multiple sectors. A multi-objective optimization approach has been used in air traffic control by [11] to minimize an aggregated complexity metric, designed and validated by Eurocontrol 1, over sectors. [12] uses the multi-objective to model the trade-off between sector congestion and delays

THE MATHEMATICAL MODEL

EXPERIMENTS

Experimental Setting

Analysis

Post-Pareto Analysis

Conclusion