Thesis and Semester Projects
Urban areas are the biggest consumers of electricity and energy consumption is only likely to increase with rapid urbanization. Out of the urban building stock residential buildings require continuous supply of energy for space heating and appliances. To answer to this demand in a sustainable way, policy makers need to design energy efficiency strategies that must rely on accurate and traceable models. These models estimate energy demand based on a series of building features, out of which building age is of prime importance because it predicts the insulation properties of the building. To support the energy modelling process, we propose a method of automatically identifying building age from spatial data at a large scale. We identify features of buildings that are significant for age prediction and determine which set of features has best prediction power at a national scale, in Germany. It is expected that the accuracy of classification will be strongly related to sampling design and data availability. The final results will be used to identify the impact of misclassification errors on estimating energy use in urban energy models, providing in this manner a measure of the reliability of such models.
Due to increasing pressure on the system, the functioning of public transport networks (PTN) in metropolitan areas is crucial for future mobility. Within these networks, hierarchical levels can be distinguished where different levels have dierent functions. These hierarchical levels can be analyzed but there is no way to quantitatively determine the hierarchy in a PTN. Therefore, this paper presents a metric to quantify the hierarchy in PTN to increase the understanding of these complex networks. In order to determine the hierarchy, a metric is developed based on a combined topological and empirical approach. The metric is a multiplication of three dierent elements which are the topological influence, (non-)redundancy and transfer potential. Together these elements are applied to de fine the hierarchical degree of nodes in a network. Furthermore, to determine the hierarchy of a network as a whole, a hierarchical coefficient, based on the distribution and inequality of the hierarchical degree in the network is developed. The metric has been applied to case-studies for the Dutch cities of Amsterdam and Rotterdam which allows for different state and cross-network comparison. The results show some expected yet non-trivial results identifying dierent patterns in network structures for network states and different spatial distribution of hierarchy between networks. Furthermore, by dividing the network into functional levels, a hierarchical structure can be identifi ed. Throughout this study, a new method to quantify hierarchy in PTN, based on dierent approaches, is developed which can be seen as the most important contribution of this research. While this study explores the implications of this metric, it can be applied in numerous different contexts. Furthermore, in potential the metric has numerous network related applications such as reducing vulnerability and solving bottlenecks.
Changes in climate conditions lead to unanticipated variations in glacial runoffs, snowmelt and precipitation, both significantly changing river flows. An imbalance in river network equilibrium leads to flooding and often ends up causing tremendous damage to society and environment. Regions that are perceived to be downstream from the source of flooding may in fact end up taking the brunt of the river force due to flood cascades. However, most studies cater to flood risk . We propose to devise a novel methodology to map rivers as unidirectional networks using river network geometry and scaling relationships fundamental to its tree-structure. Following that, we aim to convert the unidirectional flow networks to Bayesian belief networks calibrated by precipitation data and changes in glacial terrain at the source. Thus, our goal is to develop a likelihood map of excessive flooding around river networks that takes cascading into account based on drainage basin topography and network effects of river streams. A posterior inference of flooding around river streams would arm policy initiatives with strong evidence to develop safeguarding mechanisms for life and property in good time.
In this study, a data-driven, generic and transfer-based methodology for separation and ranking the PTNs has been put forward. With the hierarchy of a network, this is beneficiary for the management and operation of operators for focusing on the higher level network layer and in turn provide better service for passengers. The study introduces three steps to rank the hierarchy of a PTN: (1) using the passenger journey and ride data to derive transfer flow matrix; (2) applying C-space network representation with community detection method to separate and visualize the PTN layer; (3) performing ranking method, regarding inner- and intra- transfer flow. To this end, the hierarchy of a PTN could be presented with temporal attributes. Different day of week and various time period of a day could potentially yield different hierarchy. The proposed unsupervised learning algorithm is based on passenger transfer flow data, independent from geographic location and the mode of transportation. The study shows that the level is changing based on the selected time slot and can be a mixture of different modes, which is dissimilar from the hierarchy purely based on qualitative method.
Since cities are growing faster than ever, city planning is crucial to maintain a fully functional and efficient infrastructure. However, still no comprehensive city model exists that is able to explain the structure of today’s cities and predict their future. In order to take a small step towards developing such a model, we aim at identifying the basic building blocks of cities. This thesis proposes a data-driven approach towards city modelling using unsupervised clustering techniques. Complete city maps of 251 cities worldwide are analyzed. First, clustering is conducted on scalar features and a similarity measure between cities. We show that although we obtain reasonable clustering results, this approach is unsuitable for the identification of the fundamental elements of cities. In the second part, we focus on network motifs in city graphs and also use latent Dirichlet allocation, a technique from natural language processing, for in-depth city analysis based on network subgraphs and motifs.
With the advent of Automatic Fare Collection in transit networks, there has been a dramatic rise in the availability of transit data. In this work, we show that this data can be used to reveal information about the urban structure and the accessibility of the rail network. Specifically, we analyze the diurnal ridership of the Transit for London underground stations in order to describe the urban structure of London and to quantify the suitability of the network for daily commuters. By removing the conventional dependence on origin-destination trajectories, our methods can be expanded to other transit networks, such as bus and tram. Our work serves as an easily applicable planning tool for urban planners and transit agencies.
Worldwide tourism revenues have tripled in the last decade. Yet, there is a gap in our understanding of how distances shape peoples’ travel choices. To understand global tourism patterns we map the flow of tourists around the world onto a complex network and study the impact of two types of distances, geographical and through the World Airline Network, a major infrastructure for tourism. We find that although the World Airline Network serves as infrastructural support for the International Tourism Network, the flow of tourism does not correlate strongly with the extent of flight connections available worldwide. Instead, unidirectional flows appear locally forming communities that shed light on global travelling behaviour since there is only a $15\%$ probability of finding bidirectional tourism between a pair of countries. We find that most tourists travel to neighbouring countries and mainly cover larger distances when there is a direct flight, irrespective of the time it takes. This may be a consequence of one-way cyclic tourism that we uncover by analysing the triangles that are formed by the network of flows in the International Tourism Network.
The stability and eﬃciency of the global network of commercial airline con-nections has become a vital part of our globalized society. Going beyond empirical analysis, we present here a model trying to understand the forma-tion of the world airline network (WAN) from basic principles. In an iterative algorithm, our model employs two opposing forces: A passenger’s desire to ﬂy on non-stop ﬂights whenever possible and an airline’s strive to maximize proﬁtability of each connection. As a function of a proﬁtability threshold, we identify three distinct families of networks with a fully-connected, a core-periphery, and a tree-like structure, respectively, as outputs of this algorithm. Characterizing the regimes using several diﬀerent metrics, we show that our model is able to recreate the unique core-periphery structure of the empir-ical WAN. Remarkably, in this regime of networks, the passenger load on each ﬂight (airlines’ proﬁtability) is maximized while the average shortest path (passengers’ convenience) stays stable. However, comparing results of a connectivity robustness analysis, we also ﬁnd that the modeled networks are more robust than the real-world network, suggesting that further develop-ment of our model may help to improve the current state of the world airline network.