[en] Over the course of the electricity system's transformation, controllable, conventional power plants are increasingly replaced by renewable energy sources. Since a large proportion of renewable electricity is volatile, alternative flexibility options are needed to balance the electricity supply and demand at any given time. This doctoral thesis explores the flexibilization of the electricity demand-side and the extent to which decentrally coordinated flexibility deployment can be beneficial from a local as well as from a systemic perspective. Two challenges result from the volatility and decentralized nature of electricity generated by photovoltaic and onshore wind energy plants: The need for flexibility to integrate volatile power generation, and the increasing complexity of the electricity system due to the high number of power units and resulting bidirectional power flows in the medium- and low-voltage grids. Due to these challenges, decentralized systems are playing a growing role in the discussion on the further development of the electricity system. This thesis addresses the outlined challenges; it consists of four modules, each described in a scientific paper. The first module analyzes the motivation for decentralized electricity systems. A decentralized electricity system is a decentralized approach to the challenges resulting from increasingly distributed (renewable) electricity generation: On the one hand, the challenge is to align local electricity generation more closely with local demand, while, on the other hand, shifting some parts of the electricity demand to balance the available supply, where possible. One of the key drivers for implementing decentralized electricity systems is the concept of regional energy autonomy. Therefore, the first paper examines the effects of increasing energy autonomy on the district level (NUTS-3) in 2030 by expanding renewable electricity (photovoltaic). The analysis is performed for the urban and rural districts of Southern Germany. In this paper, flexibility is considered in the form of battery storage. It can be seen that a significant increase in renewable electricity capacity is economically advantageous in all regions examined. However, achieving higher levels of energy autonomy causes additional costs and leads to overcapacities, although dependence on the overarching power system remains. The second module focuses on electric vehicles as a promising flexibility option. In order to quantify the techno-economic demand response potential of electric vehicles in Germany in the future, a centralized system is assumed within this module. The paper models both the controlled charging of electric vehicles and their feedback of power to the electricity markets. The implications of controlled charging for the system load and the electricity price are explored on the one hand. On the other hand, the avoidance of so-called avalanche effects is analyzed. These occur if a critical mass of electric vehicles react in an uncoordinated manner to a signal. The results indicate that controlled charging of electric vehicles offers substantial flexibility potential and has positive impacts on the system, provided that the incentive signal also considers the (controlled) charging behavior of other electric vehicles. However, the coordinated deployment of flexibility can have a negative impact on the financial attractiveness of controlled charging. From the results of the first two modules, it can be concluded that, in principle, small-scale flexibility resources offer high potential for balancing electricity generation and demand. At the same time, both modules outline the challenges associated with centralized and decentralized coordination of these flexibility resources. The design and implementation of decentralized systems requires high-resolution information on electricity generation, demand and the resulting flexibility potentials. In the third module of the thesis, therefore, a method is developed to create a data set with high spatial and temporal resolution. This method is applied to model the demand and supply for all German regions (again NUTS-3) in 2030. In addition, regional demand response is simulated using the regional hourly energy balance. Furthermore, the regions are grouped based on their structure of electricity demand and supply based on a k-Means clustering algorithm. The cluster analysis highlights the high heterogeneity be-tween regions in terms of the balance of supply and demand and the effectiveness of demand response in reducing electricity imports and exports. In urban regions, the future regional residual load assumes both positive and negative values. Therefore, flexible end users are particularly effective at integrating volatile renewable electricity over the course of the day. Regions with smaller energy-intensive industries or those with medium-sized cities are already comparatively balanced without de-mand response. However, demand-side flexibility can still be used here to further increase the share of regionally consumed electricity by shifting loads over the course of the day. The modeling of decentralized demand flexibility in the third module of the thesis is based mainly on technical parameters. Functioning decentralized concepts also require incentives that promote the participation of decentralized actors, which are taken into account in the fourth module of the thesis: The paper simulates a local energy market for prosumers. The prosumers have the possibility to adjust the flexible fraction of their demand (electric vehicles and battery storage) according to their generation unit as well as the local market price and to minimize their electricity procurement costs individually. The local energy market aggregates the electricity purchase and feed-in of all participating prosumers and, thus, provides an incentive that reflects the local supply and demand balance. For evaluation purposes, the prosumers' costs and revenues are benchmarked against a self-consumption scenario and integration into a central spot market. The evaluation shows that direct participation in a central spot market would be financially more attractive for the prosumers studied than participation in regional trading or self-consumption under the assumed framework conditions. It also represents the most advantageous option for flexibility use with regard to integrating renewable electricity. However, the results also suggest that, from the overall system perspective, a local prosumer market has higher benefits than self-consumption. All four modules of this thesis examine the use of demand-side flexibility and in particular, the effect of balancing electricity generation and demand in decentralized systems. The incentives for this can be non-monetary, such as striving for energy autonomy, or monetary, such as minimizing electricity procurement costs. The results show that sufficient technical flexibility potential exists to at least partially compensate for the volatility of electricity generation from renewables, and thus to contribute to meeting the challenges posed by the energy transition. Balancing supply and demand on a local level has limited economic attractiveness compared to trading on a supra-regional level. However, when compared to the self-consumption of prosumers today, the advantages of decentralized incentives are clear with regard to integrating renewable energies into the system. Establishing decentralized electricity systems is therefore one option to incentivize flexibility.