Browsing by Author "Krötzl, Julius"

During the last decades, Helsinki and many other cities have begun to restrict parking supply in the city center and in transit-oriented developments, in order to minimize the negative impacts of parking and to restrain growth in housing prices. However, residential parking supply should only be reduced in areas that are well served by public transportation. In last years, novel data sources have been created to simulate the transportation network and land-use distribution in the future. By using computer-processing capacity to combine the travel time and land-use data sources, potential accessibility in the future can be modelled. The aim of this thesis is to provide information on future accessibility by sustainable travel modes, by taking into account the different distance friction characteristics of different land-use opportunities and to estimate car ownership in Helsinki in the year 2030. This thesis has been done as an assignment for the traffic and street planning unit of the City of Helsinki. Methods of this work include distance-based potential accessibility measures, which were computed by combining travel time matrices and land-use data using Python scripts and a geographic information system (GIS). In this work, travel time was used as the transport element of accessibility. For choosing the distance decay functions for the accessibility measures in this thesis, empirical travel data from the Helsinki region travel survey was used, which consists of travel times and trip purposes of the residents’ daily journeys in the Helsinki region in 2012. Travel time and land use estimations for the years 2017 and 2030 from the Helsinki region traffic forecasting system (HELMET) as well as geographic information data from the SeutuCD registers were used as input data for the accessibility analyses. In addition, factors affecting car ownership in the Helsinki region were analyzed and linear regression models were created to estimate future parking demand in Helsinki using accessibility and population density variables. According to the results, potential accessibility measures model the mobility patterns more realistically than cumulative opportunity measures as they weight each feature according to the distance from the origin zone. By comparing potential accessibility results by different means of transport, it can be stated that sustainable transport accessibility in 2030 is, compared to the car still very low. According to the car ownership correlation analysis, the independent variable with the highest correlation coefficient is the percentage of gross floor area of blocks of flats of the entire gross floor area of residential buildings in the zone. The independent accessibility variable with the highest correlation coefficient is the percentage of potential job accessibility by public transport in relation to car, which has a strong negative effect on car ownership (R ≈ -0.8). The highest R-squared value of the multiple linear regression models predicting car ownership is 0.66, meaning that 66 percent of the variation of car ownership can be explained by the independent variables. Thus, the predicting model can be used in estimating future car ownership, if the relationships between car ownership and the predictive variables are assumed to be constant over time.