Mixed integer 0-1 programming for time-based metering in air traffic management

Narendra Sharma, Purdue University

Abstract

In this thesis, we present a mixed integer 0-1 programming approach to time-based metering in air traffic management. In time-based metering we assign scheduled arrival times to aircraft at predefined points in the airspace, meter fixes and runway. The problem is one of assigning scheduled times of arrival to the aircraft at above points such that minimum separation between the aircraft are followed, no overtake is issued if the aircraft belong to same stream (set of aircraft which a controller separate from each other), time window restrictions are followed, and aircraft do not exceed allowable maximum delay times (AMDT) in a sector. Under above constraints, we minimize the total delays of aircraft at the runway. The problem is solved optimally using Gams/Cplex software. We also present a mixed-integer 0-1 programming formulations of an existing sequencing technique called constrained position shifting. Other techniques such as FCFS sequencing are also discussed. Computational results of proposed approach and the existing sequencing techniques are compared for a number of test problems involving up to 35 aircraft. The computational results demonstrate that the proposed approach prevent overtaking in the same stream. The other three approaches may result in infeasible solutions (i.e., via allowing overtake) upstream of terminal area. In the proposed approach excess delays are absorbed at higher altitude and the delay get reduced as we go upstream whereas the excess delays in other approaches are not only absorbed at lower altitude but also has higher values. The proposed optimization model can be used as a benchmark to compare the quality of solutions obtained from different real-time air traffic planning tools such as Multi-center Traffic Management Advisory (McTMA). The quality of solution would be assessed in terms of how close the solutions provided by such software tools are to the optimal solution.

Degree

M.S.I.E.

Advisors

Ozsen, Purdue University.

Subject Area

Industrial engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS