Dr. M.A. Mitici

profiel

Bio

Mihaela Mitici has an MSc degree in Operations Research from University of Amsterdam and a PhD in Stochastic Operations Research, Department of Applied Mathematics, from University of Twente. Since 2016 she is an assistant professor in the Air Transport & Operations section, Faculty of Aerospace Engineering, TU Delft. She specializes in Operations Research, with a focus on stochastic processes, decision-making under uncertainty, applied probability theory. Her main application domains are predictive aircraft maintenance, airport operations, and urban air mobility.

Mihaela has been part of several research projects such as HO2020-SESAR ADAPT, HO2020 ReMAP, EFRO Airport Technology Project. She is currently supervising 3 PhD students, several MSc students, and teaches the MSc course “Stochastic processes and simulation”. 


Link to Google Scholar:    scholar.google.com/citations

Expertise

·       Operations Research

·       Stochastic Processes and Simulation

·       Stochastic Optimization

·       Machine Learning, supervised and unsupervised learning

Application domains:

Predictive Aircraft Maintenance

Airport Operations Scheduling

Operations of Urban Air Mobility

Projects

Title: Models and optimization approaches for predictive aircraft maintenance (2020-present)

Description: This project aims to develop innovative scheduling models for aircraft maintenance that use i) data-driven prognostics about the condition of aircraft components and ii) stock levels for aircraft components. The scope of the project is scheduling of maintenance tasks at the fleet level, for a short to medium planning horizon. The optimization models are expected to take into account prognostics on the remaining useful life of components. The project will also contribute to the development of prognostics algorithms for the remaining useful life of components.

PhD Student: Ingeborg de Pater

 

Title: HO2020 Real-time Condition-based Maintenance for Adaptive Aircraft Maintenance Planning (2018-present)

Website: h2020-remap.eu

Description: The ReMAP project proposes methods and tools to support the implementation of condition-based maintenance for aircraft. Among others, the project aims to develop health diagnostics and prognostics of aircraft systems and structures, using innovative data-driven machine learning techniques and physics models; to develop efficient maintenance optimization models; and to evaluate maintenance strategies by means of end-to-end (rare event) Monte Carlo simulation.

PhD student: Juseong Lee


Title: EFRO Airport Technology Lab – Airside Operations (2019-present)

Website: www.stichtingrhia.nl/portfolio/airport-technology-lab/

Description: To support an efficient planning of airside operations at airports, there is a need for optimisation models that take into account potential flight delays and cancellations. This project aims to develop machine learning algorithms to predict flight delays and cancellations. These results are further considered in optimisation models for airport operations. The goal of the Airport Technology Lab is to support early identification and monitoring of critical flights, and to efficiently plan airside operations.

PhD student: Mike Zoutendijk


(COMPLETED) HO2020 ADAPT (2017-2019)

Website: visualization.adapt-h2020.eu

Description:The ADAPT project proposed a set of flight scheduling methods that assume the concept of en-route time-window flying, i.e., a temporal interval to which flights are recommended to adhere to so that sector capacities are met. Several scheduling models are proposed at a strategic level (to address demand-capacity imbalances), as well as at a pre-tactical level (to account for weather uncertainties). With these models, the ADAPT project aims to provide methods to increase flight flexibility and predictability.

Selected papers

Title: An integrated assessment of safety and efficiency of aircraft maintenance strategies using agent-based modelling and stochastic Petri nets, J. Lee, M. Mitici, Reliability Engineering and System Safety, 2020

DOI: doi.org/10.1016/j.ress.2020.107052

Description: We model an end-to-end aircraft maintenance processes, from task generating to task planning to task execution to aircraft operation, using stochastically and dynamically coloured Petri nets. Next, making use of Monte Carlo simulation, we assess both safety and efficiency indicators of condition-based maintenance strategies. We demonstrate our framework for the case of aircraft landing gear brakes, where we model the degradation of the brakes by means of Gamma processes.

 

Title: Assessing strategic flight schedules at an airport using machine learning-based flight delay and cancellation predictions, Journal of Air Transport Management
DOI:
doi.org/10.1016/j.jairtraman.2019.101737

Description:We propose classification algorithms to predict whether flights scheduled in the strategic phase (6 months prior to the day of the execution) are subject to arrival/departure delays and cancellations during execution. Using the obtained flight delay and cancellation predictions, we propose a generic methodology to rank strategic flight schedules at an airport. We demonstrate our methodology using strategic flight schedules at London Heathrow Airport.

 

Title:  Rolling-Horizon Electric Vertical Takeoff and Landing Arrival Scheduling for On-Demand Urban Air Mobility, Journal of Aerospace Information Systems

DOI: doi.org/10.2514/1.I010776

Description:This paper investigates the throughput of a double-landing-pad vertiport by proposing a new vertiport terminal area airspace design and a novel rolling-horizon scheduling algorithm with route selection capability to compute the optimal required time of arrival for eVTOLs. A case study on arrivals in a hexagonal vertiport network is performed to show the algorithm performance with different configurations.

vakken
2016 - Monte Carlo Simulation of Stochastic Processes 1
2017 - Monte Carlo Simulations of Stochastic Processes II
2016 - Monte Carlo Simulations of Stochastic Processes II
2018 - Monte Carlo Simulation of Stochastic Processes 1
2017 - Monte Carlo Simulation of Stochastic Processes 1
2018 - Operations optimisation
2019 - Stochastic Processes and Simulation
2020 - Stochastic Processes and Simulation
2018 - Stochastische Processen en Simulatie
nevenwerkzaamheden
Geen nevenwerkzaamheden -

2020-03-01 - 2022-03-01