Refine
Year of publication
Document Type
- Doctoral Thesis (14)
Language
- English (14)
Has Fulltext
- yes (14)
Is part of the Bibliography
- no (14)
Keywords
- Luftverkehr (3)
- Risikomanagement (3)
- Air transport (2)
- Investition (2)
- Investment (2)
- Risk management (2)
- Air traffic (1)
- Aufsatzsammlung (1)
- Automotive industry (1)
- Betrieb (1)
This dissertation investigates the application of machine learning to improve decision making in airline operations. In an introductory overview, we1 discuss the airline industry and the challenges of decision making in airline operations: e. g., complex IT infrastructure, interconnected resources, delay management, fuel price volatility and future environmental regulation. Machine learning can efficiently integrate a large volume of data from a variety of data sources and formats to generate accurate predictions. To assess the viability of using machine learning models for decision making in airline operations, we develop a model based on linear regression and gradient boosting to predict aircraft arrival time. Furthermore, we integrate cost index optimization to model the impact of aircraft speed on arrival time. While we find that machine learning can improve prediction accuracy by more than 30 %, the optimal cost index varies according to fuel cost, flight distance and delay costs. We propose an overall reduction in cost index for short-haul flights to reduce fuel cost while maintaining punctual operations. Arrival time predictions serve as input for daily operations planning and control. For network carriers accurate arrival time predictions are key for efficient hub operations. In a next step, we focus on aircraft arrival time prediction for intercontinental flights. We analyze the accuracy of en-route weather data provided by the flight plan, generate features based on en-route weather data and integrate them in our prediction model. We evaluate three machine learning models: linear regression, random forest and gradient boosting. Through our approach, we can assess the impact of en-route weather data. Overall, an increase in prediction accuracy of 25 % is achieved. By including en-route weather data, prediction accuracy is improved by 5 %. Our model outlines the essential features for intercontinental arrival time predictions and assess the value of en-route weather data. Furthermore, we outline organizational challenges in implementing predictive analytics. Future environmental regulations are a challenge for the airline industry. From 2020 onwards net growth in CO2 emissions is prohibited. Thus, airlines need to focus on initiatives to limit fuel consumption. We develop a prediction model for fuel consumption considering ten different aircraft types. Our results show that fuel consumption can be improved by more than 30 % for short and long-haul flights resulting in an annual reduction of 10.5 million e in fuel costs and a reduction on green house gas emissions of 66.000 tons. Furthermore, we assess the pilots’ willingness to integrate our prediction model in their fuel decision. Our analysis shows that high prediction accuracy, model understanding and a long testing phase are essential for acceptance of the prediction model. The main implication is that decision making in airline operations can be substantially improved through machine learning. Therefore, more prediction.
1The term “we” refers to the authors of the respective chapters as denoted at the beginning of each chapter. For the abstract, this refers to the authors of Achenbach and Spinler (2018a,b), Achenbach et al. (2017).
Hedging retail promotions
(2012)
This dissertation investigates different applications of data analytics in supply chain planning. In the last years, data analytics became more important, because of the increase of computational power and the larger availability of data. Data analytics is used in various domains to improve operations performance, increase customer satisfaction and revenues. However, both the research and the application of data analytics in supply chain management is still lacking behind other industries. We1 analyze the potential of data analytics in the field of supply chain planning in three exemplary fields: demand forecasting, partial defection prediction and price discrimination. In addition, we demonstrate how to deal with three common challenges in the field of data analytics: the manual effort for method selection and hyperparameter tuning, the difficult interpretability of machine learning methods and the risks associated with data collection through randomized experiments. In the first paper, we develop a method selection approach in the field of intermittent demand prediction. Our model combines high predictive performance with automation and calculation efficiency. Unlike common practice, the prediction method gets automatically chosen for each data set without any manual selection. Our results are stable across three different data sets that come from different sources but all contain intermittent demand time series. We showcase the impact of the proposed forecasting approach with a warehouse operation simulation. We thereby prove the financial benefit with empirical data. In the second paper, we deal with partial defection prediction in a business-tobusiness environment in the logistics industry. The predictions must combine predictive performance with interpretability and profit maximization. Our model uses a large variety of customer-based and time-series-based features to predict the probability of partial defection for each customer. We use a data permutation approach to make the best performing, black-box models interpretable. Furthermore, we use a profit assessment to identify the method that leads to the highest revenue through successful retention actions. In the third paper, we study price sensitivity prediction. We do not use any randomized experiments, because the risk of loosing customers through such experiments is too high. Thereby, we address the challenge of data availability
1The term “we” refers to the authors of the respective chapters as denoted at the beginning of each chapter. For the abstract, this refers to the authors of Faber and Spinler (2019a,b,c).
This dissertation investigates capacity and technology choice decisions in maritime container shipping under demand and regulatory uncertainty. In an introductory overview, we discuss the industry and the challenges that complicate investment decisions in shipping: e. g., the multitude of decisions, market volatility, excess capacities and the trend of new environmental regulation. Real option valuation methods can account for strategic options and the uncertainties in capacity decisions in shipping. To assess the impact of chartering on maritime investment, we analyze investment and charter options individually in a continuous-time model. We combine both in a discrete-time approach taking into account key features of the industry: investment with time to build, divestment, chartering, an endogenous charter rate, layup, and demand uncertainty. While we find demand volatility to increase optimal capacities if only investment with time to build is possible, chartering reduces this effect. It adds value to the overall project, should be mainly applied to compensate unexpected capacity shortages and needs to be considered in decision-making. Uncertainty about future eco-regulation is a further challenge for the industry. In an approximate dynamic programming model extension, we account for a stochastic introduction of operating cost-increasing regulation to assess optimal capacity choice under regulatory uncertainty. Regulation can allow for grandfathering, affecting only newly acquired vessels. We find uncertainty about future regulation with grandfathering to induce heavy up-front investment to secure a low cost base even in regulated markets. Such uncertainty may increase excess capacities and industry emissions. Uncertainty without grandfathering, however, reduces overall investment and emissions. In this case, the market may contract as investors resort to chartering until uncertainty is resolved. To also assess technology choice in light of uncertain future regulation, we develop a two-phase regime-switching model. We derive analytical solutions and study the effects of regulatory uncertainty on technology choice in a numerical extension that relaxes restrictive assumptions. We find that such uncertainty can increase optimal capacities and that a single-technology strategy is preferable over a fleet of mixed technologies in most cases. We further extend the model and compare the effectiveness of two different regimes—an emissions cap and an emissions tax. Results suggest that an emissions cap is more effective at reducing overall emissions while a tax causes lower regulation cost. The regime choice also determines the optimal technological fleet composition. The main implication is that chartering and regulatory uncertainty have a strong effect on optimal investment in shipping and should be considered in project valuation. Further, regulatory uncertainty can lead to unintended investment behavior that undermines regulatory goals. Lastly, regulatory regimes are not equally effective/efficient in reducing the environmental footprint of shipping and constitute varying incentives for investing in eco-friendly technology.
In this thesis, we(1) use operations research methods to provide insights into three areas associated with health care operations management. In Chapter 2, we use a discreteevent supply chain simulation to asses if coordination among partners is beneficial in a supply chain with the characteristics of the German pharmaceutical market. We find that the greatest cost savings and service levels could be achieved through a highly integrated collaboration although most of its impact could already be achieved through sharing point-of-sales demand information. Results suggest that coordination is most beneficial in situations where product shelf life is short and demand variation is high.
In Chapter 3 we consider quality-of-life maximizing sequences of prophylactic surgeries for female carriers of a BRCA1/2 genetic mutation, who face a significantly elevated breast and ovarian cancer risk. Using a Markov Decision Process model, we determine the optimal surgery sequence that maximizes the carrier’s expected lifetime qualityadjusted life years (QALYs). Baseline results demonstrate that a QALY-maximizing sequence recommends a bilateral mastectomy between ages 30 and 60 and bilateral salpingo-oophorectomy after age 40 for BRCA1 carriers. Surgeries are recommended later for BRCA2 carriers, as their cancer risk is lower. The model’s structural properties show that when one surgery has already been completed, there exists an optimal control limit after which performing the other surgery is always QALY-maximizing.
In Chapter 4, we develop a two-stage model for optimizing when and where to assign Ebola treatment unit (ETU) beds—across geographic regions—during an infectious disease outbreak’s early phase. The first stage includes a dynamic transmission model that forecasts occurrence of new cases at the regional level, thus capturing connectivity among regions; in this stage we introduce a coefficient for behavioral adaptation to changing epidemic conditions. The second stage includes two approaches to efficiently allocate intervention resources across affected regions. Such an allocation could have prevented up to 3,434 infections over an 18-week period during the 2014 Ebola outbreak in West Africa, a 58% improvement compared with the actual allocation.
(1) In Chapter 2, 3, and 4, the term ’we’ refers to the authors of Nohdurft & Spinler (2016), Nohdurft et al. (2016a), and Nohdurft et al. (2016b), respectively.
This dissertation assesses investment decisions in container shipping. To understand the current state of the industry, key characteristics and challenges, such as overcapacity, eroding margins due to low freight rates, long investment lead times, and frequent changes in alliance structure are introduced.
The nature of the industry motivates the application of real options, hence a real options investment model in oligopolistic competition is presented. An analytic solution in continuous time as well as a dynamic programming solution in discrete time are derived. The model takes into account an endogenous price function, fuel-efficient investment, endogenous lead times, and endogenous price formation in the secondary vessel market. This allows to study the impact of competitive intensity, number of players, volatility, fuel-efficiency, lead time, and variable cost on optimal capacity. An investigation of optimal investment policies shows that strategic action increases firm value and strategic alliances might help alleviate some of the industry’s challenges.
Since the container shipping market is characterized by frequent alliance changes, the performance of the real options model in the context of a cooperative shipping game is assessed. Extending the coalition structure value concept it can be shown that, compared with discounted cash flow, the real options trigger performs better, especially in light of high competitive intensity and freight rate volatility while not exhibiting substantial disadvantages in other settings. A further assessment of a number of drivers for alliance instability finds that alliance complexity cost, freight rate volatility, and competitive intensity increase alliance changes.
To verify the investment approach, a characterization of the container freight rate is provided with an empirical Autoregressive Integrated Moving Average (ARIMA) model. It can be observed that the freight rate exhibits a negative relationship with capacity deployment; hence the oligopoly price function is confirmed. Based on the freight rate characterization, a back testing of the real options investment approach is provided. It shows that if players had applied the presented approach, capacities would have decreased and rates improved. A number of limitations of the real options approach are identified,
i.e. substantial impact of volatility expectation, potentially induced cyclicality from trigger approaches, and the timing impact of investment and divestment lead times.
The implications of this research are that strategic action in the container shipping industry is worthwhile and understanding the market specifics (such as competitive intensity, volatility, and freight rate characterization) is very important. Container carriers should add a real options approach to their investment toolkit and keep an eye on potential overcapacity. Finally, entering strategic alliances is suggested, but complexity should be avoided.
Door-to-door (D2D) air travel is gaining momentum for airlines, airports, and feeder traffic providers. The mobility industry and researchers are broadening their scope to include the entire travel chain, from origin to final destination. Intermodal mobility products are already on the market. At the same time, widespread trends affect transport service providers (as the suppliers) and passengers (regarding demand). Acquiring a better understanding of future D2D air travel trends is crucial for the mobility sector for long-term planning, product adaptation, the services provided and the pricing of these, and improvements in the passenger experience. Focusing on the European market, the overall objective of this doctoral thesis is to identify and understand the future trends of D2D air travel. It is divided into three parts; these provide different perspectives on trends and employ a range of methods that lead to results that develop from each other.
In Part One, the Delphi technique is utilized to identify future travel trends. The study considers projections of European air passengers and their requirements for their entire air travel chain, including airport access, a long-haul flight, and airport egress. The research focuses on 2035 and is based on a two-round Delphi survey involving 38 experts from the transport industry, academia, and consultants. The Delphi survey is supplemented with findings from a preliminary study, consisting of a literature review, interviews with 18 experts in the field of air travel, and a workshop attended by experts. Results reveal that digitalization and personalization will be the main drivers in 2035 and that passengers might demand value-added use of their travel time. In addition, environmentally friendly travel products are considered desirable but only somewhat probable by 2035. Passenger type, age, origins, and travel budget will still be influential factors in 2035. Based on the results from a hierarchical cluster analysis, Part One presents three possible future scenarios: (1) personalized D2D travel, (2) integrated D2D travel, and (3) the game-changer. A technical chapter elaborates on the Delphi technique and individual research steps.
Part Two explores the supply aspect and to what extent transport service providers consider strategically relevant trends. The scope of D2D air travel is adapted by applying multi-labeled text classification models to 52 corporate reports from a sample of transport service providers that operate in the European market. Trends identified in the first Delphi study and from an additional literature review are used to develop seven classes. Two prototype models are developed: a dictionary-based classifier and a supervised learning model using the multinomial naive Bayes and linear support vector machine classifiers. The latter yields the best model output, revealing which trends have a higher, medium, or lower relevance on the supply side. The results show that providers consider environmentally friendly air transport and related products to be highly relevant while disruption management, leveraging passengers’ data, and improving airport feeder traffic through innovative mobility initiatives are considered to be of medium relevance.
Part Three explores air passengers’ preferences and willingness to pay for ancillary services in the current transition into the new normal, brought about by the ongoing COVID-19 pandemic, high uncertainty, and changing market dynamics. A choice-based conjoint analysis is used to test six attributes within a hypothetical travel scenario for a long-haul one-way air trip. Choice data from 269 German business and leisure passengers are analyzed using a hierarchical Bayes estimator. Results reveal that the total ancillary service upgrade price influences passengers’ choices the most, followed by a seat upgrade for greater comfort and the CO2-compensation of a flight. Hygiene-related ancillaries bring low utilities. Female and senior passengers care more for environmentally friendly ancillaries. Confirming previous research, business passengers and frequent flyers care more for onboard comfort.
The combined impact of changing global demand and supply dynamics, extensive trading and speculation as well as global recessionary fears, has led to an environment of unprecedented volatility in worldwide commodity markets. As a result, effective risk management has become an increasingly important topic on the agenda of top management in a broad range of industries. While practical evidence shows that successful firms integrate both operational decision making and financial hedging in a firm-wide, coordinated risk management strategy, this entails managerial challenges. On the one hand, quantifying a firm’s exposure to raw material cost risk necessitates a sound understanding of the stochastic commodity market dynamics. On the other hand, once the exposure to different sources of risk is understood, executives face an intricate optimization problem over their operational and financial decision variables with the ultimate goal to reduce profit variability, while maintaining attractive business opportunities.
In this thesis, the topic of operational and financial risk management is investigated from three different perspectives.
In Chapter 2, a four-factor maximal affine stochastic volatility model of commodity prices is developed, which is consistent with many stylized characteristics of storable commodity markets as well as the historical term structure of commodity futures and option prices. Based on this model, we provide new insights with respect to the structural dynamics of commodity markets and the pricing and hedging of commodity derivatives. As the stochastic model used to describe the uncertain evolution of commodity prices can have important implications also in the valuation and risk management of real assets, a realistic commodity price model is a prerequisite for the integrated risk management models outlined in the subsequent chapters of this dissertation.
Given the previously developed intuition for commodity market dynamics, we model the integrated operational and financial risk management problem of a stylized, single-product industrial firm in Chapter 3. The firm faces risk in the price of commodity inputs and price sensitive, stochastic demand. Within this setting, the firm seeks to maximize inter-temporal utility under downside risk aversion over a multi-period time horizon by dynamically choosing physical procurement volumes, unit selling prices, and a futures hedge. We provide a flexible, simulation-based optimization algorithm, which allows us to solve the firm’s decision problem under realistic, multi-factor commodity price dynamics involving uncertainty in the interest rate and convenience yield as well as stochastic volatility. Based on this model, we characterize the firm’s optimal operating policy and investigate a range of topics including: (a) the value of managerial flexibility and the economic cost of restrictive supply contracts; (b) the importance of accounting for the stochastic nature of costs, interest rates, convenience yields, and volatility in risk management; (c) parameter and estimation risk; (d) the impact of risk aversion and hedging on the distribution of cash flows; and (e) the sensitivity of expected performance to key input parameters.
As opposed to the case of a single-product firm, integrated procurement risk management in a large, multi-divisional organization does not only require the above mentioned cross-functional coordination between, for example, the purchasing, sales, and finance department but also involves a cross-divisional coordination of actions in order to effectively target the firm-wide net risk exposure. To capture the specific aspects of integrated risk management in this type of setting, we extend the above model to a two-product firm in Chapter 4. Within this model, each of the two divisions are subject to cost and demand risk, which can be respectively correlated. Moreover, we allow for dynamic cross-selling to capture the potential complementarity/substitutability of items. The firm has access to futures, call, and put options associated with each of the commodity input markets for financial hedging. Under an intertemporal mean-variance utility function, we are able to provide analytic solutions to the firm’s dynamic procurement, pricing, and financial hedging problem. Based on a complementary numerical study, we analyze the impact of risk correlations and unilateral changes in the market environment of one division on the entire firm. Moreover, we discuss the impact of hedging and risk aversion on optimal policy and assess the effectiveness of different operational and financial hedging strategies for risk reduction.