Search

The ability to reduce energy loss at semiconductor surfaces through passivation or surface field engineering has become an essential step in the manufacturing of efficient photovoltaic (PV) and optoelectronic devices. Similarly, surface modification of emerging halide perovskites with quasi-2D heterostructures is now ubiquitous to achieve PV power conversion efficiencies (PCEs) > 22% and has enabled single-junction PV devices to reach 25.7%, yet a fundamental understanding to how these treatments function is still generally lacking. This has established a bottleneck for maximizing beneficial improvements as no concrete selection and design rules currently exist. Here we uncover a new type of tunable passivation strategy and mechanism found in perovskite PV devices that were the first to reach the > 25% PCE milestone, which is enabled by surface treating a bulk perovskite layer with hexylammonium bromide (HABr). We uncover the simultaneous formation of an iodide-rich 2D layer along with a Br halide gradient achieved through partial halide exchange that extends from defective surfaces and grain boundaries into the bulk layer. We demonstrate and directly visualize the tunability of both the 2D layer thickness, halide gradient, and band structure using a unique combination of depth-sensitive nanoscale characterization techniques. We show that the optimization of this interface can extend the charge carrier lifetime to values > 30 μs, which is the longest value reported for a direct bandgap semiconductor (GaAs, InP, CdTe) over the past 50 years. Importantly, this work reveals an entirely new strategy and knob for optimizing and tuning recombination and charge transport at semiconductor interfaces and will likely establish new frontiers in achieving the next set of perovskite device performance records., Main text: 16 pages, 4 figures. Supporting Information: 47 pages, 26 figures, 6 tables

The fundamental relationship of traffic flow is empirically estimated by fitting a regression curve to a cloud of observations of traffic variables. Such estimates, however, may suffer from the confounding/endogeneity bias due to omitted variables such as driving behaviour and weather. To this end, this paper adopts a causal approach to obtain an unbiased estimate of the fundamental flow-density relationship using traffic detector data. In particular, we apply a Bayesian non-parametric spline-based regression approach with instrumental variables to adjust for the aforementioned confounding bias. The proposed approach is benchmarked against standard curve-fitting methods in estimating the flow-density relationship for three highway bottlenecks in the United States. Our empirical results suggest that the saturated (or hypercongested) regime of the estimated flow-density relationship using correlational curve fitting methods may be severely biased, which in turn leads to biased estimates of important traffic control inputs such as capacity and capacity-drop. We emphasise that our causal approach is based on the physical laws of vehicle movement in a traffic stream as opposed to a demand-supply framework adopted in the economics literature. By doing so, we also aim to conciliate the engineering and economics approaches to this empirical problem. Our results, thus, have important implications both for traffic engineers and transport economists.

In the usual Bayesian setting, a full probabilistic model is required to link the data and parameters, and the form of this model and the inference and prediction mechanisms are specified via de Finetti's representation. In general, such a formulation is not robust to model mis-specification of its component parts. An alternative approach is to draw inference based on loss functions, where the quantity of interest is defined as a minimizer of some expected loss, and to construct posterior distributions based on the loss-based formulation; this strategy underpins the construction of the Gibbs posterior. We develop a Bayesian non-parametric approach; specifically, we generalize the Bayesian bootstrap, and specify a Dirichlet process model for the distribution of the observables. We implement this using direct prior-to-posterior calculations, but also using predictive sampling. We also study the assessment of posterior validity for non-standard Bayesian calculations, and provide an efficient way to calibrate the scaling parameter in the Gibbs posterior so that it can achieve the desired coverage rate. We show that the developed non-standard Bayesian updating procedures yield valid posterior distributions in terms of consistency and asymptotic normality under model mis-specification. Simulation studies show that the proposed methods can recover the true value of the parameter efficiently and achieve frequentist coverage even when the sample size is small. Finally, we apply our methods to evaluate the causal impact of speed cameras on traffic collisions in England.

Crowding valuation of subway riders is an important input to various supply-side decisions of transit operators. The crowding cost perceived by a transit rider is generally estimated by capturing the trade-off that the rider makes between crowding and travel time while choosing a route. However, existing studies rely on static compensatory choice models and fail to account for inertia and the learning behaviour of riders. To address these challenges, we propose a new dynamic latent class model (DLCM) which (i) assigns riders to latent compensatory and inertia/habit classes based on different decision rules, (ii) enables transitions between these classes over time, and (iii) adopts instance-based learning theory to account for the learning behaviour of riders. We use the expectation-maximisation algorithm to estimate DLCM, and the most probable sequence of latent classes for each rider is retrieved using the Viterbi algorithm. The proposed DLCM can be applied in any choice context to capture the dynamics of decision rules used by a decision-maker. We demonstrate its practical advantages in estimating the crowding valuation of an Asian metro's riders. To calibrate the model, we recover the daily route preferences and in-vehicle crowding experiences of regular metro riders using a two-month-long smart card and vehicle location data. The results indicate that the average rider follows the compensatory rule on only 25.5% of route choice occasions. DLCM estimates also show an increase of 47% in metro riders' valuation of travel time under extremely crowded conditions relative to that under uncrowded conditions.

Mobility as a Service (MaaS) is widely expected to make sustainable transport choices more attractive. New approaches to ticketing will be a core part of MaaS, both to attract users and fund services. The associated pricing decisions will be a matter of public policy as much as business objectives, because they can have large social welfare effects. This paper describes options for different pricing structures and their relative efficiency. It considers the potential impact that differing objectives of public and privately-owned transport providers might have on pricing decisions. It concludes with an assessment of the possible effects of Covid-19 on the MaaS market.

Lipids have been recognized as key players in cell signaling and disease. Information on their location and distribution within a biological system, under varying conditions, is necessary to understand the contributions of different lipid species to an altered phenotype. Imaging mass spectrometry techniques, such as time of flight - secondary ion mass spectrometry (ToF-SIMS) and matrix-assisted laser desorption/ionization (MALDI) are capable of revealing global lipid distributions in tissues in an untargeted fashion. However, to confidently identify the species present in a sample, orthogonal analyses like tandem MS (MS/MS) are often required. This can be accomplished by bulk sample analysis with liquid chromatography (LC)-MS/MS, which can provide confident lipid identifications, at the expense of losing location-specific information. Here, using planarian flatworms as a model system, we demonstrate that imaging gas cluster ion beam (GCIB)-ToF-SIMS has the unique capability to simultaneously detect, identify, and image lipid species with sub-cellular resolution in tissue sections. The parallel detection of both, intact lipids and their respective fragments, allows for unique identification of some species without the need of performing an additional orthogonal MS/MS analysis. This was accomplished by correlating intact lipid and associated fragment SIMS-images. The lipid assignments, respective fragment identities, and locations gathered from ToF-SIMS data were confirmed via LC-MS/MS on lipid extracts and ultra-high mass resolution MALDI-MS imaging. Together, these data show that the semi-destructive nature of ToF-SIMS can be utilized advantageously to enable both confident molecular annotations and to determine the locations of species within a biological sample.

This paper is concerned with the Wider Economic Impacts (WEIs) of transport improvements that arise via scale economies of agglomeration. It reviews the background theory and empirical evidence on agglomeration, explains the link between transport and agglomeration, and describes a three step procedure to appraise agglomeration impacts for transport schemes within Cost Benefit Analysis (CBA). The paper concludes with a set of recommendations for future empirical work on agglomeration and transport appraisal.

While the interaction between transport and agglomeration economies is widely accepted, there is insufficient research attempting at a direct empirical quantification. Using a balanced panel dataset for US metropolitan areas, we estimate a system of simultaneous equations to measure the indirect effect of urban agglomeration economies which arises through transport provision. Our findings suggest that public transit reinforces the effect of urban agglomeration, whereas road lane miles appear to weaken it. The results highlight the importance of public transit in supporting positive urban agglomeration externalities. info:eu-repo/semantics/publishedVersion

Organic-inorganic perovskites such as CH3NH3PbI3 are promising materials for a variety of optoelectronic applications, with certified power conversion efficiencies in solar cells already exceeding 21%. Nevertheless, state-of-the-art films still contain performance-limiting non-radiative recombination sites and exhibit a range of complex dynamic phenomena under illumination that remain poorly understood. Here we use a unique combination of confocal photoluminescence (PL) microscopy and chemical imaging to correlate the local changes in photophysics with composition in CH3NH3PbI3 films under illumination. We demonstrate that the photo-induced 'brightening' of the perovskite PL can be attributed to an order-of-magnitude reduction in trap state density. By imaging the same regions with time-of-flight secondary-ion-mass spectrometry, we correlate this photobrightening with a net migration of iodine. Our work provides visual evidence for photo-induced halide migration in triiodide perovskites and reveals the complex interplay between charge carrier populations, electronic traps and mobile halides that collectively impact optoelectronic performance.

This paper quantifies the effect of speed cameras on road traffic collisions using an approximate Bayesian doubly-robust (DR) causal inference estimation method. Previous empirical work on this topic, which shows a diverse range of estimated effects, is based largely on outcome regression (OR) models using the Empirical Bayes approach or on simple before and after comparisons. Issues of causality and confounding have received little formal attention. A causal DR approach combines propensity score (PS) and OR models to give an average treatment effect (ATE) estimator that is consistent and asymptotically normal under correct specification of either of the two component models. We develop this approach within a novel approximate Bayesian framework to derive posterior predictive distributions for the ATE of speed cameras on road traffic collisions. Our results for England indicate significant reductions in the number of collisions at speed cameras sites (mean ATE = -15%). Our proposed method offers a promising approach for evaluation of transport safety interventions.

This paper develops a multiply robust (MR) dose-response estimator for causal inference problems involving multivalued treatments. We combine a family of generalised propensity score (GPS) models and a family of outcome regression (OR) models to achieve an average potential outcomes estimator that is consistent if just one of the GPS or OR models in each family is correctly specified. We provide proofs and simulations that demonstrate multiple robustness in the context of multivalued causal inference problems.

The PRACT Project (Predicting Road ACcidents - a Transferable methodology across Europe) aims to develop a European accident prediction model structure that could be applied to different European road networks with proper calibration. An important part of the PRACT project involves the review of the recent and salient literature on Crash Modification Factors (CMFs) with the aim of understanding CMF availability and identifying CMF needs in Europe. In addition, a survey of 23 National Road Authorities (NRAs) from 20 countries was completed to understand CMF use and CMF needs within NRAs. The survey identified a number of CMFs that are useful to NRAs but not readily available. Based on both the literature review and the NRA survey, the PRACT project developed some key CMFs that are currently missing or underrepresented in the literature. CMFs were developed for motorways and 2-way rural roads using data from Italy, England and Germany. A number of CMFs were developed, including for work zones, speed enforcement, high friction wearing course, traffic composition, horizontal curvature and vertical gradient. This paper presents a synthesis of CMF availability and needs based on the survey of NRAs and an extensive review of the literature, and outlines the development process for CMFs within the PRACT Project.

This paper constructs a doubly robust estimator for continuous dose-response estimation. An outcome regression model is augmented with a set of inverse generalized propensity score covariates to correct for potential misspecification bias. From the augmented model we can obtain consistent estimates of mean average potential outcomes for distinct strata of the treatment. A polynomial regression is then fitted to these point estimates to derive a Taylor approximation to the continuous dose-response function. The bootstrap is used for variance estimation. Analytical results and simulations show that our approach can provide a good approximation to linear or nonlinear dose-response functions under various sources of misspecification of the outcome regression or propensity score models. Efficiency in finite samples is good relative to minimum variance consistent estimators.

This paper reviews methods that seek to draw causal inference from non-experimental data and shows how they can be applied to undertake ex-post evaluation of transport interventions. In particular, the paper discusses the underlying principles of techniques for treatment effect estimation with non-randomly assigned treatments. The aim of these techniques is to quantify changes that have occurred due to explicit intervention (or 'treatment'). The paper argues that transport interventions are typically characterized by non-random assignment and that the key issues for successful ex-post evaluation involve identifying and adjusting for confounding factors. In contrast to conventional approaches for ex-ante appraisal, a major advantage of the statistical causal methods is that they can be applied without making strong a-priori theoretical assumptions. The paper provides empirical examples of the use of causal techniques to evaluate road network capacity expansions in US cities and High Speed Rail investments in Spain.

This paper estimates the relationship between agglomeration and multi factor productivity at the one digit industry level and by region using longitudinal firm level data for New Zealand. A key focus of the paper is on methods to represent firm level heterogeneity and non-random sorting of firms. The panel structure of the data allows us to control for it at the level of local industries or enterprises. We obtain a cross-sectional agglomeration elasticity of 0.171, which falls by 70% when we use local industry controls, and by 90% when we impose enterprise fixed effects. Using industry specific production functions, we find that the “within local industry” estimates are similar, though slightly larger than the cross sectional estimates (~0.070), suggesting negative sorting between areas, combined with positive sorting within areas. The within-enterprise estimates yield a small elasticity of 0.010. Our results indicate that the imposition of a common production technology across all industries is not a valid assumption. While cross-sectional estimates may overstate the true impact of agglomeration on productivity in the presence of positive bias from sorting, the within enterprise approach (which is increasingly common in the literature) can suffer from identification problems due to the highly persistent nature of agglomeration variables and may understate the true causal effect of agglomeration on productivity. We thus rely on the “within local industry” estimates as providing the most reliable indication of agglomeration elasticities.

This paper estimates the relationship between agglomeration and multi factor productivity at the one digit industry level and by region using longitudinal firm level data for New Zealand. A key focus of the paper is on methods to represent firm level heterogeneity and non-random sorting of firms. The panel structure of the data allows us to control for it at the level of local industries or enterprises. We obtain a cross-sectional agglomeration elasticity of 0.171, which falls by 70% when we use local industry controls, and by 90% when we impose enterprise fixed effects. Using industry specific production functions, we find that the “within local industry” estimates are similar, though slightly larger than the cross sectional estimates (~0.070), suggesting negative sorting between areas, combined with positive sorting within areas. The within-enterprise estimates yield a small elasticity of 0.010. Our results indicate that the imposition of a common production technology across all industries is not a valid assumption. While cross-sectional estimates may overstate the true impact of agglomeration on productivity in the presence of positive bias from sorting, the within enterprise approach (which is increasingly common in the literature) can suffer from identification problems due to the highly persistent nature of agglomeration variables and may understate the true causal effect of agglomeration on productivity. We thus rely on the “within local industry” estimates as providing the most reliable indication of agglomeration elasticities.

This paper analyzes the impact of agglomeration externalities on hourly earnings using longitudinal worker micro-level data from the Annual Survey of Hours and Earnings over the period 2002- 2006. We find that the effect of agglomeration externalities on wages is sensitive to the estimator used. Controlling for nonzero correlation between workers’ unobservable skills and other covariates halves the size of the wage elasticity of agglomeration externalities. On the contrary, accounting for firms’ unobservable heterogeneity has only a weak contribution to the explanation of wage differentials. Another interesting result is that correcting for reverse causality between productivity and agglomeration does not appear to have a substantial impact on the magnitude of the parameter estimates. Our best estimate for the effect of labour market density (market potential) is 0.8% (5.8%). This means that doubling labour market’s employment density can raise hourly earnings by nearly 1%, while halving the distances to other markets produces an increase of hourly wages of nearly 3%. The last piece of evidence refers to the spatial attenuation of agglomeration externalities. We estimate that a 100,000 increase in the number of jobs within 5 kilometres raises hourly wages by approximately 1.19%; the effect falls sharply thereafter.

This paper analyses the relationship between firms’ multi-factor productivity and the effective employment density of the areas where they operate. Quantifying these agglomeration elasticities is of central importance in the evaluation of the wider economic benefits of transport investments. We estimate agglomeration elasticities using the Statistics New Zealand prototype Longitudinal Business Database: a firm-level panel covering the period 1999 to 2006. We estimate that an area with 10 percent higher effective density has firms with productivity that is 0.69 percent higher, once we control for the industry specific production functions and sorting of more productive firms across industries and locations. We present separate estimates of agglomeration elasticities for specific industries and regions, and examine the interaction of agglomeration with capital, labour, and other inputs.

The case for including agglomeration benefits within transport appraisal rests on an assumed causality between access to economic mass and productivity. Such causality is difficult to establish empirically because estimates may be subject to sources of bias from endogeneity and confounding. They may also be sensitive to the range of sample variance in agglomeration being used. The purpose of this paper is to demonstrate some of the key difficulties that the researcher faces in estimating agglomeration economies and to show how these can affect the calculation of agglomeration benefits for the appraisal of transport projects. The results show a high degree of sensitivity to treatment for unobserved heterogeneity and to differences in the sample variance of agglomeration. A key conclusion is that we are unable to distinguish agglomeration effects from other potential explanations for productivity increases, most notably functional heterogeneity. Consequently, the agglomeration effects of transport investments cannot be interpreted causally.

In this thesis, automated fare collection (AFC) data are used to analyse and quantify transit journey time service quality on the London Underground metro system. The thesis comprises of three main research areas. The first part focuses on characterising passenger journey time variance through the generation of empirical probability distributions of journey times under regular and incident-affected operating conditions. The distributions are parametrically defined, and practical passenger-oriented performance metrics are proposed based on the moments of the distributions. The second area of research involves decomposing total passenger journey times recorded by the AFC data into sub-components that distinguish between the walking and in-vehicle phases of a passenger journey. To achieve this, a Bayesian assignment algorithm is proposed to allocate individual passengers to individual trains. Total journey times are then decomposed into the constituent components of access, on-train, and egress times. In the third area of research, the degree to which different service supply and demand factors influence journey times is analysed. Semiparametric regression methods are applied to quantify the effect of physical station and route characteristics, operational service supply factors, and passenger demand levels for each journey time component. To quantify the effect of individual passenger characteristics on journey times, passenger-level heterogeneity within each journey time component is analysed. As an extension to the access time model, the influence of train headways on passenger wait times at the origin station is also derived. The main outputs of the thesis are the quantification of journey time performance, and the identification of the key service supply and demand factors that impact journey times. The results can be directly applied by operators to guide where potential interventions should be made in order to improve the reliability of journey times for urban rail transit networks. Open Access

The overarching theme of this dissertation is how public transportation impacts the local econ- omy and the wellbeing of society. I concentrate on the wider economic benefits of transportation to provide precise and causal evidence on how large-scale metro projects impact firm location, employment, productivity and women’s feeling of safety. I use spatially disaggregate panel microdata to understand the local impact of metro openings. Developing robust identification strategies and employing panel instrumental variable methodologies ensure the causality of re- sults. There are three empirical investigations in this thesis: the first estimates the impact of the 1999 Jubilee Line Extension on the spatial distribution of firms and employment in Lon- don, and finds empirical evidence for the existence of urban displacement. The second estimates the impact of the same metro opening on firm productivity and develops a project appraisal methodology for the calculation of agglomeration benefits. The estimation finds that within 750 meters to new stations aggregate value added, mean firm productivity and the number of local business units increased. However, areas between 1250 and 2000 meters experienced a decrease in the number of local units but mean firm productivity did not change, leading to decreasing aggregate value added. The third chapter tests the hypothesis that transport con- nectivity improves perceptions of safety for women by investigating whether metro openings in 2018 increased the safety feeling of women around stations in Delhi, India. The analysis finds that the feeling of safety significantly increased within 500 meters to stations. High levels of safety increase women’s freedom of movement, which is a major step in integrating women to the labour force, and these benefits may be separately included in transport project appraisal. Open Access

The overarching theme of this dissertation is the investigation of the relevance of transportation infrastructure for the development and performance of the private sector. I analyse this relationship for both firms located in developing and developed countries to highlight similarities and disparities of the importance of transportation networks for the performance of the firms. For the examination of this relationship, I have chosen two countries that differ substantially on various economic and social factors and their levels of transportation infrastructure: Colombia and Germany. Research on developing countries is often limited by the absence of reliable panel data, this especially applies to data on firms. I propose the use of aggregated Colombian firm data to identify a pseudo-panel of manufacturing firms that can be used to simulate the analysis of firm-level effects. I provide further validity for this method by conducting a Monte Carlo simulation, which provides support for the appropriateness of pseudo-panels in the context of firm-level investigations and further highlights the particular suitability of different panel data estimation techniques for it. The identified pseudo-panel is employed for the subsequent analysis. The elasticity of highway infrastructure on the growth of a firm’s output is significant and positive, however the identified magnitude of this effect of ap- proximately 0.13 to 0.15 is substantially larger than the average effect identified in similar studies for developed countries, hence implying the crucial importance of transportation infrastructure for economic growth in developing countries. The results further show a time lag of this effect of one year, suggesting that firms’ production processes require time to adjust to the transportation infrastructure changes. These results are robust to the inclusion of additional controls and alternative transportation variables used. An additional investigation reveals thattransportation benefits particularly accrue to firms of the heavy industries. A further analysis highlights the role of transportation infrastructure within the context of spatial heterogeneity of this effect across regions. The results indicate the benefits of transportation infrastructure predominantly only accrue to firms located in relatively stronger regions with reported output elasticities of around 0.15. Additional analyses provide no support for the importance of transportation for fostering economic convergence in Colombia. This thesis contains furthermore an examination of the role of the German highway network in influencing firm-level productivity patterns. This work uses German firm-level data and relies on the use of the Levinsohn-Petrin methodology and historical transportation data to conduct a two-step productivity estimation. I further derive two different transportation measurements to investigate the importance of transportation spillover effects. In line with previous studies of the German economic geography, I find spatial differences in productivity levels where larger levels are predominantly found in the districts of the South and West of the country, and lower productivity levels are identified throughout the Eastern districts. The investigation of the whole country identifies positive and significant productivity elasticities of highways where estimates range from 0.03 to 0.07 and 0.061 to 0.23 depending on the choice of method, controls and highway variable. This represents an increase of the firm’s productivity of 0.03 to 0.0.7 per cent following a 1 per cent increase in highways, or alternatively an increase of 0.061 to 0.23 when highway spillover effects are accounted for. This indicates that firms derive benefits from local highways and those located in the surrounding regions and the remainder of the country which points towards large spillover effects. The results further show that while highways are important for a firm’s productivity growth, alternative productivity enhancing factor, predominantly the amount to skilled labour, present relatively more important productivity determinants. The investigation of subsamples designed to capture districts with relatively high and low highway levels points towards diminishing returns of transportation infrastructure and satiation effects. While no consistent differences in the estimated effects could be identified across districts of the Eastern and Western states and for those districts which have recently received new highways under the “Transportation Projects German Unity”, further disaggregate results show that highways particularly benefited firms in rural districts in the West and those located in urban districts in the East. Overall, the robustness of the results of this chapter requires further analyses, however they tentatively point towards productivity enhancing benefits of transportation and additionally reveal a large degree of heterogeneity of this effect. Open Access

Access to the global container liner service network is vital to international trade. Disruptions to this network have major and immediate implications for consumers, industries, markets and national economies. The measurement and understanding of the existing liner service network performance susceptibility to disruptions (transport vulnerability) is therefore of critical importance to public and private stakeholders responsible for ensuring its operability and accessibility. Methodologies that assess transport vulnerability in liner service networks face two key modelling challenges: (i) A lack of historical disruption data on network components (such as ports, canals, and liner services) and (ii) the typical size of realistic liner service networks. The first challenge limits the implementation of methodologies that require prior knowledge of disruption probabilities to quantify the vulnerability of the network. The second challenge creates the need for models capable of capturing key industry practices such as transhipment, empty repositioning, and the ability of vessels to skip disrupted ports while still allowing implementations at realistic global scales. This study addresses the above-mentioned challenges by developing a quantitative framework capable of identifying critical components in large-scale networks with limited or unavailable historical disruption data. The proposed framework consists of a game-theoretic attacker-defender model (ADM) and a cost-based container assignment model (CBCAM) adapted for the analysis of networks under disruptions. The ADM consists of a two-player, zero-sum game between a malevolent agent (attacker) that seeks to maximise disruption costs and an ocean carrier (defender) that aims to minimise routing costs. The CBCAM is used to generate the payoff matrix of the game computing routing costs for the ocean carrier and disruption costs for the attacker on each of their available strategies. Linear program formulations of the joint AD-CBCAM allow for implementations on large-scale realistic liner service networks. This scalability is tested using CPLEX to solve network instances of up to 88 distinct liner services, 230 ports, and 2,648 OD pairs. Results allow establishing performance baselines and test the effectiveness of iterative interventions aimed at increasing the resilience of the system. Open Access

Cities are indispensable for economic development, although urban agglomerations can fail to deliver their potential due to congestion. Metros are key to achieving the benefits of urban densities without the costs of congestion, by means of providing unparalleled fast and high-capacity transportation in densely populated areas. These urban rail systems require large one-off initial investments but also substantial operational costs, in order to provide the regular day-to-day rail service and maintain the metro's fixed and mobile assets in good condition, such as the rail tracks and the trains, respectively. The aim of this thesis is to better understand what determines these operational costs. The analysis is based on a panel of more than 20 large and medium-sized urban rail systems across the world during the period 2005 to 2014. The research is based on the econometric estimation of cost functions for the aggregate operating costs but also for some of its constituents like train service, station service and rolling stock maintenance costs. Likewise, different outputs are considered to evaluate the cost efficiency of the metros, such as car kilometres, passenger journeys and train hours among others. The analysis quantifies the effects of economies of scale and economies of density in metro operations, which can vary significantly between different divisions of the metro operations leading to substantial policy implications. Moreover, the effect of wages, and how this is linked to labour productivity, is also examined. The dissertation also estimates the corresponding cost elasticities for other cost factors such as labour outsourcing, subsidies and network characteristics. Lastly, the research describes the effect of age on urban rail operating costs and also outlines the background time trends on how the metro operating costs could evolve in the future. Open Access

Crowding is a major source of inconvenience for public transport users in densely populated metropolitan areas globally, while eliminating crowding requires costly investments. Crowding can be considered as a cornerstone phenomenon of public transport theory, as the interaction between demand and supply side policies. This PhD thesis aims to improve our understanding of the mechanics behind crowding, using microeconomic modelling techniques. From a demand perspective, the crucial precondition of any objective economic analysis is to reliably quantify the inconvenience caused by crowding. In pursuit of this goal, the thesis develops a statistical model to infer the user cost of crowding from metro passengers' route choice decisions. As an important intermediate research outcome, the thesis delivers a novel passenger-to-train assignment algorithm that recovers the network-level crowding pattern of a metro system. Our method is a unique contribution in the sense that it is based on large-scale automated datasets: we use smart card and automated vehicle location data only. The theoretical part of the thesis provides new insights into crowding pricing and capacity optimisation. One of the key messages of the thesis is that crowding in certain time periods and network segments is an unavoidable feature of optimal public transport provision, when demand fluctuates by time and space, but capacity cannot be differentiated between jointly served markets. We show that pricing can be an efficient tool to tackle the deficiency caused by this technological constraint. The thesis devotes special attention to two policy relevant applications: (i) the external cost of seat occupancy, an externality inversely proportional to the density of crowding, and (ii) the inefficiency of unlimited-use travel passes. Our conclusions may assist researchers and practitioners in better understanding the true cost of public transport usage and the related aspects of optimal policy design, including pricing, subsidisation and capacity provision. Open Access