Within this dissertation a model-based framework and its software implementation as Decision Support System is described which supports decision makers to design free-floating carsharing systems. Free-floating carsharing like DriveNow and car2go emerged as new carsharing concept and shows fast growing rates in Europe and North America. This is expected to continue and expansion to more cities is considered the key industry trend. So far, the local adaptation of the system is based on expert opinions solely. The developed model-based framework comprises the design of the operating area, fleet size, charging concept, fleet mix, and price because these five aspects are identified as key elements when adapting free-floating carsharing to a selected city. In summary, twelve quantitative input variables and five qualitative aspects are modeled as adjusting screws to determine these aspects. The individual models are evaluated where possible and results appear very satisfying. For validation and to ensure applicability, this research also develops a Decision Support System for the execution of the framework. This implementation is demonstrated for Chicago, where a whole new free-floating carsharing system was planned out guided by the developed models. In a first step, this work discusses key ecological, economical, social, and technological trends which lead to the emergence of a new mobility paradigm. Here, battery electric vehicles (BEVs) and carsharing appear as promising future transportation solutions, also in combination. This is followed by a comprehensive description of carsharing, more precisely, free-floating carsharing. Despite all advantages, such as reduction in traffic, resource use, emissions, and ownership-hassles, there are still some challenges that have to be overcome. First, profit potentials are limited as providers are currently only on the edge to profitability and BEVs potentially worsen the situation by inducing additional costs. Second, the diffusion of the concept is still in the beginning phase and even though forecasts are promising, the diffusion process has to be supported through the provision of market adequate offers, also because customers are very inexperienced. Third and lastly, it is identified that market knowledge is currently very limited. Free-floating carsharing is very new and different from previous approaches. The lack of knowledge on the system poses a problem since expansion to more markets will be the key industry trend in future. Accordingly, the research question is derived as follows: ”How shall free-floating carsharing systems be adapted to different markets/cities?” Since every market or city is different, the answer to the research question shall be a model-based framework and its software implementation that supports decision makers to design free-floating carsharing systems city-by-city. The research design of this dissertation derived three sub-questions from the overall research question. This allows for focus on relevant factors and aspects for the development of the model-based framework. Sub-question a) (”which aspects must be locally adapted?”), requires clarification on how carsharing systems are structured. The so-called service marketing mix is chosen as suitable theoretical framework and the debate on adaptation vs. \r\nstandardization gives a guideline for answering the question. To do so, expert knowledge is gathered through interviews. It is revealed that the operating area, fleet size, fleet mix, charging mode, and the price are main aspects that have to be locally adapted. Sub question b) (”how to measure success?”) defines success as adoption, which is an individuals’ decision to make full use of carsharing. This definition is translated into measurable quantitative variables. As analysis of GPS booking and customer data shows, booking density (number of bookings km2 ) appears as the most appropriate measure for success. The analysis of question c) (”what influences success?”) is logically based on the same definition of success and hence success factors are defined as determinants of adoption. Here, interviews with experts, companies, and customers are used for a qualitative determination of factors. These factors are argumentatively explored whether they are relevant for local adaptation or not, followed by a quantitative statistical analysis of the remaining factors with the help of booking and structural city data. The result shows that population density, housing rent, city center distance, and restaurant and hotel density are the most influential success factors. Backed up by these findings the model-based framework is developed, allowing for local adaptation. This is the core result of this dissertation and answers the research question individually for each city as the operating area, fleet size, charging concept, fleet mix, and the price can be determined individually for each city with the help of the framework and in application, its implementation in a Decision Support System for experts. 1. The determination of the operating area relies on a regression model based on population density, city center distance, housing rent, and hotel-/restaurant density. Further, qualitative alignment with a city map to validate forecast results, on-street parking possibilities, and the technology and spatial distribution of existing charging infrastructure must be carried out. 2. The size of the operating area is also a input for the second model concerning the fleet size, together with desired values for utilization and spatial coverage of the system. These the three inputs determine the fleet size, processing results from an agent-based simulation. 3. The third model which determines the charging concept contains a threshold value for the decision of whether decentralized charging is feasible or not. To do so, the input of the operating area as well as the number of charging stations in the operating area is considered. This recommendation must then be aligned qualitatively with the spatial distribution and technology of chargers. 4. Model four concerns the fleet mix which also relies on the number of chargers in the operating area. Furthermore, a potential strategic or regulatory minimum percentage of electric vehicles which might have to be employed is considered. This mix of electric and conventional vehicles should then be reviewed qualitatively whether the respective car models fit in with the local vehicle mix and potentially, the fleet mix must be aligned to suit local conditions. 5. The fifth and final model helps to determine the price per minute for a freefloating carsharing system in a city. Quantitative input factors are a base price for adaptation, local taxi costs, local private car costs, and the exchange rate. The modeled price must then be aligned with the chosen fleet mix because for more useful/valuable cars there could be a price premium. Also, competitor prices must be considered in order to position the product correctly. Further, this dissertation carries out an analysis on the willingness to pay of customers, which could act as base price instead of current market prices. Together, this results in twelve quantitative input variables and five qualitative aspects which should be considered for alignment, being modeled as adjusting screws to determine the five key aspects of this transportation system. The models were evaluated wherever possible and results were very satisfactory. Joined together and considering the interrelatedness of the models, the model-based framework is derived and guides decision makers on the local adaptation of their free-floating systems as it limits complexity and allows to focus on what is important. To bring it in application, a prototype software implementation as Decision Support System is developed. This gives decision makers a powerful tool on hand to plan for the key industry trend, expansion to new markets internationally and out of its niche in transportation. The validity and applicability of the tool was demonstrated on the example of the city of Chicago. In conclusion, this dissertation entered new territory in many respects as very few researchers dealt with the problems posed by free-floating carsharing. Moreover, carsharing was never approached in the context of holistic local adaptation to a city. This holistic approach is probably the greatest strength of this work since it presents a reference for further research in the many fields that were opened. Moreover, it provides practitioners with a strong tool-set to actually design free-floating carsharing systems and therefore optimize current and future systems with regards to the operating area, fleet size, fleet mix, charging mode, and the price.    «

Within this dissertation a model-based framework and its software implementation as Decision Support System is described which supports decision makers to design free-floating carsharing systems. Free-floating carsharing like DriveNow and car2go emerged as new carsharing concept and shows fast growing rates in Europe and North America. This is expected to continue and expansion to more cities is considered the key industry trend. So far, the local adaptation of the system is based on expert opin...    »