The Dutch healthcare sector wrestles with rising costs, staff shortages, and increased demand due to healthcare centralization. This, coupled with worsening traffic congestion, underscores the need for efficient solutions like drone-based medical transport. This paper addresses the need for effective fleet management strategies tailored to the unique demands of the healthcare environment. Specifically, it seeks to develop an adaptive strategy that accommodates the network's expansion and the inherently stochastic and urgent nature of pickup and delivery orders associated with medical transport. Utilizing an agent-based model formulation, the paper introduces a novel approach combining adapted temporal sequential single-item auctions for allocation and scheduling with reinforcement learning for drone repositioning to optimize fleet management. Our findings highlight the strategy's consistent efficiency across various demand scenarios, maintaining performance within predefined limits. Notably, the repositioning module significantly enhances the fleet utility and the fraction of served orders, albeit at the expense of increased cost per delivery. Conversely, the reallocation module causes minimal performance improvement. Under heightened stochasticity introduced by urgent orders, the strategy maintains stable costs per delivery while fleet utility and order fulfillment rates decline. Additionally, our investigation underscores the increasing benefits of repositioning in more stochastic scenarios. Moreover, exploring hybrid fleets reveals that while short-range high-payload drones can reduce cost per delivery, they compromise overall fleet utility and order fulfillment rates. Furthermore, we identify the under-utilization of payload capacity in scenarios with orders weighing up to 2 kilograms for a drone with a payload of 10 kilograms.

Machine learning models have improved Prognostics and Health Management (PHM) in aviation, notably in estimating the Remaining Useful Life (RUL) of aircraft engines. However, their 'black-box' nature limits transparency, critical in safety-sensitive aviation maintenance. Explainable AI (XAI), particularly Counterfactual (CF) explanations, offers a way to explain model decisions by suggesting alternative scenarios for different outcomes. Additionally, Bayesian models enhance predictions by quantifying uncertainty, yet the combination of CF explanations and Bayesian methods is largely unexplored. This study investigates counterfactual methods within a Bayesian framework to improve the explainability of RUL estimation and improve model performance. For this, a Bayesian Long Short-Term Memory (LSTM) model was applied to the C-MAPSS data-set. This research uniquely applies CF explanations in two ways, with the goal of offering insights into how varying operational conditions could affect the RUL, and to improve the model's performance by generating additional augmented data with reduced uncertainty for the model to train on. Preliminary results show that CF explanations are able to provide insights and suggestions for RUL improvement. Also, the addition of the augmented data using the CF uncertainty reduction method has shown to improve the models predictive performance, confirming the viability of this approach as a data augmentation method.

This Master Thesis investigates the possible improvements to the Target Time Management concept to optimize the arrival flows for SWISS International Airlines. The aim is to improve operational performance based on the current model used, as well as prove that Target Time Management constitutes a valuable system to improve operations in a dynamic way. To leverage the dynamic nature of slot assignment, an environment model is created and used as training base for two Multi-Agent Reinforcement Learning algorithms. These two algorithms, Soft-Actor Critic (SAC) and Proximal Policy Optimization (PPO), are then tested against the baseline model currently used in operations at SWISS (based on Mixed-Integer Linear Programming). The four domains to measure the algorithms' performance are passenger connecting time, curfew performance, rotation delay and fairness to other airlines. The algorithms were trained in a simulation environment based on statistical representations of the dynamics of the slot allocation system of EUROCONTROL. They were then tested with new data, where they outperformed a MILP implementation in passenger connecting time and rotation delay metrics (curfew and fairness were comparable in magnitude, since the MILP was slightly unfair for SWISS and RL was slightly unfair for other airlines). PPO was then also tested on the real slot assignment environment hosted by EUROCONTROL and once again compared to a MILP approach. Here, it was found that the improvement in critical passenger connecting time was 5.0 minutes for the MILP, and 5.9 minutes for PPO. Rotation delay was improved by 0.9 minutes by the MILP, and by 4.8 minutes by PPO. PPO also made the highest delays higher and the lowest delays lower, which would require EUROCONTROL or SkyGuide representatives to interpret and make conclusions on fairness and safety. Curfew performance was optimal for both methods. In conclusion, it is proven that Reinforcement Learning techniques can aid the dynamicity of decision-making within Target Time Management. It is also proven that Target Time Management with a dynamic decision making approach can improve operational performance compared to a static one.

Punctuality is a key performance indicator for any airline. Hub-and-spoke airlines are particularly committed to on-time arrivals to guarantee passenger connections. Flights that are delayed at departure need to compensate for the lost time whilst airborne. Because fueling takes place well before scheduled departure, predicted departure delays determine the planned fuel amounts for en-route speed optimization. To prevent unnecessary fuel burn, airlines benefit from highly accurate departure delay predictions. This study aims to extend previous work on airline departure delay forecasting to a dynamic and probabilistic domain, whilst incorporating novel day-of-operations airline information to further minimize prediction errors. Random Forest, CatBoost, and Deep Neural Network models were proposed for a case study on KLM departures from Amsterdam Airport Schiphol between 1 January 2020 and 1 August 2023. The Random Forest model was selected for its superior probabilistic performance and high accuracy in predicting delays between 5 and 25 minutes, for which en-route speed optimization has the largest effect. The departure delay probability distribution forecasts are made at six distinct prediction moments: 90, 75, 60, 45, 30, and 15 minutes before scheduled departure time. At the 90-minute prediction horizon, the model reaches a Mean Absolute Error (MAE) of 8.46 minutes and a Root Mean Square Error (RMSE) of 11.91 minutes. Simultaneously, for 76% of flights, the actual delay is within the predicted probability distribution range. Considering the costs and emissions associated with the decision-making following the departure delay prediction model, this study puts strong emphasis on explainability. Flight dispatchers are therefore provided not only the predicted departure delay but also the main factors impacting the prediction, explaining the context of the flight. The versatility of the model was demonstrated in two shadow runs, where delays caused by familiar and unfamiliar factors were successfully predicted.

Fuel-efficient flight operations and improved Air Traffic Management (ATM) operations are identified as one of the main pillars in achieving net-zero CO2 emissions by 2050. While considerable research has focused on airspace management and ATM operations, flight operations as managed by airlines have received little attention.

Accurate aircraft/tail-specific performance modeling is crucial for identifying savings while existing models such as Euro-Control’s BADA and manufacturers’ book models remain too generic. Additionally, trajectories including optimal routes, altitudes, and airspeeds, must be determined to minimize fuel consumption. Emerging solutions leverage in-flight data connectivity and Machine Learning (ML) methods to provide pilots with real-time decision support. However, quantifying and validating saving potentials present challenges due to unpredictable variables and performance modeling complexities.

This thesis aims to address these challenges by developing a tail-specific performance modeling framework using high-fidelity flight data and ML methods. The framework identifies and corrects tail- and flight-specific biases from the flight data, allowing fuel savings to be identified on a per-flight basis in post-flight analysis. The tail-specific performance model shows different Maximum Range Cruise (MRC) speeds than generic values determined by the aircraft. The benefits emerging from these optimal speeds are determined by high-accuracy simulations of different cost index strategies on flight-specific and network-wide levels. Three cost index strategies are evaluated and compared to generic MRC operations. Savings in both fuel and time are observed of 75 kg and 93 s, 96 kg and 111 s, and 127 kg and 107 s, on an average per-flight basis. In conclusion, this research demonstrates the existence and magnitude of fuel and time savings by flying tail-specific cruise speeds compared to generic values determined by the aircraft.

Urbanization is a worldwide trend that drives an immense increase in air traffic demand. The worldwide aviation network makes global business possible which generates economic growth, creates jobs and facilitates international tourism and trade. Many of the world's transport hubs like The Greater London Area have allocated this demand over various airports, which forms a multi-airport region (MAR). Airport expansion plans and infrastructural investments in multi-airport regions, therefore, depend not only on the total level of air traffic growth but on its allocation. Therefore, the goal of this research is to develop an analysis framework for the market dynamics driving airport activity levels, focusing on multi-airport regions, and analyse how this provides a base for strategic decisions in the region. This goal was split into two sub-goals. First, understand the evolution of airport’s market shares based on the allocation of air traffic passengers amongst airports in a chosen MAR. Second, understand the market dynamics of passenger integrated transport systems in a MAR. Forecasting its underlying determinants can provide estimates for the future transport system in a MAR, which accommodate and facilitate smooth future operations of used logistical components. Greater London was chosen as a multi-airport region.

Part one consists of a quantitative model that can compute annual air traffic demand for a multi-airport region, and the allocation of this demand over airports within, for the period 2010 - 2050. Aggregate air traffic demand is based on UK's GDP projections and the allocation model was based on the relative performance of airports in terms of their accessibility, airfares and connectivity. A multivariate regression model was applied correlating historical (2010 - 2019) airport performance to their respective market shares. The results show regional air traffic demand is likely to grow 82\% over the next 30 years. Based on 60 observations, the model can predict its market shares with an R-squared of 0,952.

Part two progresses the quantitative model to generate a forecasting framework for the next 30 years for various important aviation-related variables in the region. This concludes that for London Gatwick, City and Stansted airport the projected growth in air traffic demand will not form any capacity problems, but for London Heathrow, Luton and Southend airport, it will. A major factor influencing this is the rising environmental awareness and increasing costs for carbon emission abatement. London Heathrow Airport, being UK's most important airport, together with the UK government has identified several developments to facilitate the expected growth in air traffic demand.

Overall economic assessments (OEAs) can provide a sound basis for decision-making in the areas of investments in new technologies and the application of existent technologies or operating practices. However, due to their long time horizons and complex nature, OEAs often contain many uncertain inputs, making a deterministic simulation insufficient to reflect the true value of the output. In order to incorporate these uncertainties, a systematic and efficient approach for uncertainty analysis is required. This paper sets out such a process, which consists of an iterative Uncertainty Quantification (UQ) based on importance measures for each uncertainty obtained from a Global Sensitivity Analysis (GSA). Methods for UQ and GSA are generally actively researched and well established in theory, but are infrequently applied on actual problems due to the computational and organisational complexity associated with integrative uncertainty assessments. To address this issue, the process is demonstrated on an interdisciplinary problem, namely the economic valuation of Engine Wash (EW) procedures using the cost-benefit tool LYFE. It is concluded that with this iterative uncertainty quantification procedure, the total uncertainty in the output distribution, measured using the 2.5th and 97.5th percentiles and expressed in terms of the Delta Net Present Value, is reduced from $45K - $983K to $78K - $584K. To achieve this reduction, additional modelling was carried out for only the two most important of the six uncertainties, determined using the GSA results, which illustrates the efficient allocation of modelling resources.

In this paper, the emissions mitigation potential of implementing operational towing at major European airports is evaluated. Using mixed-integer linear programming, a scenario simulating mixed taxiing operations is used to find the relationship between fixed vehicle costs and target emission savings at an airport and European level. Using formulae derived from the ICAO Advanced Emissions Model, fuel burn and emissions are calculated for conventional taxiing, single-engine taxiing (SET), and hybrid and electric operational towing. An assignment optimization model is used to assign vehicles in a vehicle fleet to flights; yielding the minimum fuel to cover a flight schedule with mixed taxiing operations. As the number of vehicles in the fleet increases, the saving potential for each additional vehicle is calculated, linking fuel and emission savings with vehicle costs. Two scenarios are compared; Scenario 1 with hybrid towing vehicles and Scenario 2 with electric towing vehicles. Applying this method to 30 case-study airports, a maximum jet fuel reduction of 66% for hybrid towing and 57% for electric towing is calculated; outperforming SET at 29 of the 30 airports. The average taxi time and aircraft compatibility are the driving factors that influence the maximum fuel savings potential. The total average taxi minutes of compatible flights is thus the best metric to predict an airport's potential, with a coefficient of determination equal to 0.96 for hybrid towing and 0.97 for electric towing. Because of the shorter refuelling downtime, hybrid vehicles can service more jobs per day than electric vehicles. This means a smaller hybrid fleet size can achieve the same CO2 and jet fuel savings on a European level. As the fleet size increases, the marginal savings per hybrid vehicle decrease at a higher rate than electric vehicles. This means that the maximum emissions savings potential can be a misleading metric for comparing strategies. By using this approach to link vehicle fleet size and potential savings, a more extensive trade-off of emission mitigation strategies is possible, allowing stakeholders to find the best strategy to fit a budget or emission savings target.

Aviation is an important sector and one of the key contributors to today’s economy. With a forecasted annual growth rate of 4.4%, the sector is expected to expand even more in the coming years. This expansion is however followed by the environmental burdens aviation causes to the environment. Following the anticipated growth of the aviation industry, action to avoid the same trend in emissions is critical. Following the recent introduction of circularity in inflight services airlines are taking measures often labeled as sustainable, seeking ways to reduce their carbon footprint. It is important to evaluate whether these measures improve the environmental performance of inflight services or not. To achieve that, the cradle-to-grave approach of the Life Cycle Assessment
(LCA) method is recommended. Furthermore, there is little to no research on quantifying the overall climate impact of inflight services. This thesis is therefore aiming to fill this research gap by conducting an LCA on inflight services offered to passengers and the research question is formed as: “How do inflight services, and measures intending to reduce their climate impact, perform in terms of
carbon footprint and to what extent do they enable sustainability?”. Two baseline cases were evaluated to answer the research question, a short-haul, and a long-haul flight, representing conventional inflight service practices. For each case, five scenarios were constructed based on measures aiming to reduce the carbon footprint of inflight services, namely single-use plastic reduction, food waste reduction, lightweight materials introduction, carbon offsetting, and a combination of all the measures. The analysis aimed to quantify the impact
of inflight services, in terms of contribution to climate change, and compare the applied measures to the baseline cases, assessing whether they enable sustainability or not. Overall, the study can conclude that the evaluated inflight service measure scenarios performed better than the business-as-usual inflight service practices. The total carbon footprint of inflight services per passenger on a short-haul flight was calculated as 10.9 kg CO2-Eq and on a long-haul flight as 50.2 kg CO2-Eq. The combined measures carbon footprint reduction potential could reach the impressive 96% for a long-haul flight and 89% for a short-haul flight reduction. Most of this reduction is mainly attributed to carbon offsetting, 82% for the long-haul flight and 69% for the short-haul flight. The second best-performing scenario was the lightweight materials measure which could reduce carbon footprint by 13% and 19% for the long-haul and the short-haul flight respectively. Single-use plastic and food waste reduction measures did not exceed a 0.4% reduction of the impact. Although carbon emissions of inflight services entail a small fracture of the overall aviation carbon footprint, the studied measures can contribute to notable environmental impact reductions with immediate results.

City logistics concern solutions in package movements that face challenges in rapid growth in demand for home deliveries, congestion, and expectations from consumers for sustainable solutions. Providing last mile delivery services with drones is considered a promising solution in literature as a response to the raising challenges in urban areas. In this paper, we propose an innovative framework that allows a logistics operator to design a last mile delivery network. The Network Design Model contains two parts. First, a binary integer linear programming model is presented, allowing the operator to find the optimal location of intermediate facilities. It is a flow model that uses pre-located facilities. Secondly, a sequential framework is presented that considers Facility Assignment, Knapsack Problem, Traveling Salesman Problem, and Vehicle Assignment. The Network Design Model aims to minimize the operating costs measured in delivery duration, vehicle ground time, and number of not-delivered packages. A Monte Carlo Simulation is formulated to assess the impact risk factors, uncertainties, and variation in key network design attributes have on the performance of the network.

Impact on composite structures shows a different damage behaviour compared to metal structures. Due to the current short operational life time of composite aircraft the risks of impact damages on composite structures are unknown. This paper proposes a new method for quantitative risk analysis of low velocity impact damages on composite aircraft structures by combining a conventional risk analysis with the probability of detection of the damages. Real damage data of metal structures is used to estimate impactors and to predict damages on composite structures by means of an aircraft impact damage model. To create a large set of impact events and to conduct a more accurate analysis impactor data estimated from the real metal damage data is augmented. Three fuselage sections with a significant difference in amount of damages have been selected to compare the different risk results. The outcome of a conventional risk analysis and the outcome of the probability of detection of damages show that the section with the highest amount of damages results into the highest risk. However, the proposed method by combining the probability of detection of the damages with a conventional risk analysis shows different and more revealing results. The fuselage section with nearly 50% less damages compared to the section with the highest damages appears to be the highest risk section but the difference with the other fuselage sections is small. The proposed risk analysis method intends to be a useful tool for aircraft maintenance organisations for a different approach of assessing the risks of impact damages on composite structures.

The current focus in the aviation industry for more sustainable designs, could mean the revive of propeller propulsion, due to their relative high propulsion efficiency compared to jets. In addition, the application of wingtip-mounted propellers installed in tractor configuration can be used as tip-vortex attenuating devices, reducing the wing induced drag. The wingtip-mounted propeller configuration is believed to offer a significant aircraft performance benefit from an aerodynamic perspective. So far, studies on wingtip-mounted propellers mainly concentrated on the aerodynamic interaction effects, disregarding the integration with the airframe and wing-structural mass. This thesis presents a methodology to integrate aerodynamic, aero-propulsive, and aero-structural effects of tip-mounted propellers in the context of a typical turboprop featuring hybrid-electric propulsion. The developed methodology is used to assess the effect of wingtip-mounted propellers on both wing and aircraft level, providing new insights in the potential of the configuration.

As on-time performance is one of the main contributors to success in the world of commercial aviation, predictions on flight delays and cancellations can significantly improve operational efficiency and thus quality of service. Since flight delays and cancellations are occasional and infrequent events, operational on-time performance data is inherently imbalanced. This is especially the case for cancellations, as on average 1.6% of flights are cancelled, while about 33% of the flights is delayed. For this research, flight operational data is combined with weather data to predict flight delays and cancellations on prediction horizons of hours to months before the flight, by means of Neural Network and Random Forest machine learning algorithms. Since these algorithms naturally tend towards the usage of balanced data, the need exists to find a systematic approach to deal with the imbalance issues, in order to make accurate predictions. Hence, an imbalanced data approach is proposed, which analyses model performance with indicators such as precision and F1-score on varying data imbalance ratios. The imbalance ratios are obtained through the use of sampling techniques such as Synthetic Minority Oversampling and Random Undersampling. It is concluded that the highest precision is found without any sampling while for the highest F1-score sampling is essential. Additionally, the research confirms that severely imbalanced data, like the cancellation data, yields the worst performance when compared to medium imbalanced data, like the delay data.

As the profit margins of the airline industry are relatively low, it is of utmost importance to keep costs low in order for airlines to stay competitive. An important cost factor is maintenance costs, as it can take up around 10-20 % of the total direct operational costs. Currently, much development is taking place in developing condition-based maintenance (CBM) strategies. These strategies on the one hand leverage remaining useful life (RUL) predictions of components to enable better planning and lower repair costs while on the other hand less preventive maintenance tasks are required due to the increase of useful sensor data available. This paper develops insights in the potential benefits that CBM can have as a function of different prognostic performance levels. This is done by developing cost- benefit models which accept a wide range of parameters being able to simulate prognostic effectiveness on different aircraft fleets. Results are obtained by using real MRO and operator input data. The results show that CBM can be beneficial, given that the model has a sufficient specificity and the component supply chain scales accordingly.

Estimating the RUL (Remaining Useful Life) of machinery is a useful tool for maintenance and performance operations. This results in lower costs, improved safety and operational improvements.
This paper proposes two adaptations to the CNN-LSTM network provided by Li et al. \cite{Li2019APrediction}, as well as exploring reproducibility, accuracy and sensitivity of the original DAG (Directed Acyclic Graph) network. The network at hand is an ensemble network combining LSTM and CNN neural networks to provide an accurate regression RUL prediction using the NASA CMAPSS dataset \cite{NasaNasaReprository}.
The Adaptable Time Window (ATW) adaptation increases the amount of time cycles that can be predicted and increases the accuracy, allowing for earlier predictions and better RUL predictions. Allowing state-of-the-art predictions accuracy for complex datasets. The Sub-network training adaptions did not surpass the accuracy of the original network with the current implementation settings, however is promising for further research.

Cargo airlines' Revenue Management (RM) departments oversee the acceptance/rejection process of incoming bookings. The overarching goal is to accept as many bookings as possible, hence maximizing profit, while avoiding overbooking and offloading of already accepted shipments, to ensure customer satisfaction. This whole process is characterized by great uncertainty, arising from the fact that the exact shipment dimensions and the available aircraft capacity are not always known beforehand. The decision process is further complicated by the fact that loading of shipments is generally performed by experience, resulting to inefficient use of the available space. These factors pose challenges to the acceptance/rejection of an incoming booking request, that can lead to loss of potential revenue for the airline. A novel combined forecasting and stochastic packing model is presented, that tackles the aforementioned issues. The forecasting block of our model, instead of providing point estimates, generates the probability distribution of the predicted capacity and shipment dimensions through the use of the Monte Carlo (MC) dropout technique in Long Short Term Memory (LSTM) and Multi-Layer Perceptron (MLP) network respectively. The stochastic packing block falls into the general category of Knapsack Problems and is solved using the Extreme Point (EP) heuristic within a user-defined confidence level. It uses the generated distributions to select the optimal ULD configuration and palletization strategy, and makes a decision to accept or decline an incoming booking request accordingly. For evaluation purposes, four case studies with real booking data provided by our partner airline are carried out. The results support the validity of the model as well as the efficiency and robustness of the solution method followed.

In this paper, we introduce the Air Cargo Allocation Plan Recovery Problem, where we embrace the perspective of a combination airline that relies on belly space to transport cargo. We present a recovery model that can reallocate bookings that are offloaded because of disruptions in the demand (overbooking) or in the supply (aircraft swap of cancellation) side. The model is based on a set-partitioning mixed integer linear program formulation with an itinerary generation pre-processing step whose goal is to limit the number of possible itineraries per booking and hence the computational effort. The problem is solved for three different cases from a European airline. The three cases are, a flight cancellation, an aircraft swap, and an aircraft swap with ULD configuration. The results show that the model is adaptable to fit the time available, the operational cost of the initial solution can be improved and the larger amount of the objective function value comes from revenue loss.

The MRO market currently spans around 9.5% of the total operating cost of an airline. Of this, 70% is covered by heavy-maintenance. Reduction of these costs and improving efficiency could, therefore, be significant for an airline. A possible solution is the optimization of the long-term schedule of heavy-maintenance checks. Current approaches are found to be reliant on manual input and operator experience. Next to that, revisions to the initial schedule are made continuously due to the inherently stochastic nature of aircraft maintenance through non-routine maintenance. Taking this uncertainty into account could offer more robust schedules, saving cost and improve quality of service.

This study proposes a genetic algorithm that can generate robust and efficient C-check schedules for a fleet of heterogeneous aircraft. Uncertainty in check duration and utilization are taken into account by assessing multiple scenarios through min-max optimization. This study is the first to address the long-term scheduling of heavy-maintenance checks while taking uncertainty into account. The proposed genetic algorithm finds robust and efficient C-check schedules for a case study of a European airline for a fleet of over 40 aircraft in under 30 minutes. The total number of C-checks is reduced by 7% while increasing utilization by 4.4%. This could lead to a reduction of direct annual maintenance costs of $122.5K - $612.5K and an additional $1.8M - $7.1M in annual revenue due to the increased availability of aircraft. Monte Carlo simulations show that with a probability of 41% no adjustments to the schedule are necessary over the planning horizon.

The current operational cost estimate applied by shipping management companies is insufficiently valid and accurate for determining the operational budget. Most shipping management companies include scheduled maintenance jobs in the operational cost calculation. Scheduled maintenance does not take unforeseen maintenance into account. In retrospect, unforeseen maintenance makes up approximately 8\% of the total operational costs. The estimation of the operational costs can be improved by including the unforeseen maintenance costs. Improved operational cost calculations lead to better substantiated maintenance policy decision-making. Therefore, the (unforeseen) maintenance costs over the lifetime of a vessel are modelled. This research focuses on maintenance cost calculations, based on failure behaviour. First, it is explained how the Maintenance Cost Model calculates and compares the costs of different maintenance policies. Thereafter, the model is validated in a case study. The achieved cost reductions resulting from the case study varied from 0\% to 70\%, with a conservative average of 16\%. Next, the case study conclusions are assessed in a sensitivity analysis. The model generates a, system specific, cost based ranking of the maintenance policies. The ranking can be used by shipping management companies, to obtain a better substantiated system specific maintenance policy. The developed model in this research is generally applicable and proves the concept. The research concludes with recommendations for future research, which includes further expansion of the Maintenance Cost Model and the possibilities to increase the likelihood of the number of replacements.