Energy security is one of the three pillars of energy policy next to environmental sustainability and economic efficiency. Despite its importance for policy making, there is no agreement about the definition of energy security and the metrics that could be used to measure the efficiency of different policies. As a result of this, it may be that both a political intervention in one direction and an intervention in the opposite direction could be justified on grounds of energy security. In our thesis we review the multitude of definitions of energy security. They can be characterized according to the sources of risk, the scope of the impacts, and the severity filters in the form of the speed, size, sustention, spread, singularity and sureness of impacts. Using a stylized case study for three European countries, we illustrate how the selection of conceptual boundaries along these dimensions determines the outcome. In order to reduce the overlap between security of supply and other policy objectives we propose the definition of energy security as the continuity of energy supplies relative to demand. The choice of the conceptual boundaries along the other dimensions remains a subjective decision that has to be taken by policy makers in a dialogue with society. Based on the definition of energy security as the continuity of supplies relative to demand, we examine how accurately different modelling approaches and metrics capture this concept. We find most of the wide-spread indicators, such as import shares, concentration measures and composite indicators based on expert elicitation are very rough heuristics that can easily be shown to produce inaccurate results. Simple modelling approaches such as analysis based on portfolio theory, or electricity system simulations offer some improvement, but still suffer from structural shortcomings and limitations in the way they model interdependencies. In this thesis we suggest a modelling approach which allows us to capture the interdependencies between natural, technical and human risk sources and quantify their combined impact on the continuity of energy supplies within a fixed infrastructure system. We use a case study of Italy to compare the outputs of our model with alternative metrics and simplified modelling approaches. Finally, we investigate the degree and the cost at which regulatory interventions in the form of so-called capacity mechanisms may increase the continuity of supplies in the electricity market. In contrast to previous research we find that the choice of a capacity mechanism may both be influenced by the extent to which it should be robust towards different regulatory errors as well as by the question whether it is evaluated from the perspective of consumer cost or from a welfare perspective.