Determining the location of car park by digital modeling

The rapid transformation experienced during the urbanization process has increased individual vehicle use, leading to various structural problems in urban transportation systems. A lack of parking spaces, foremost among these problems, increases traffic congestion and negatively impacts environmental sustainability, urban aesthetics, and pedestrian safety. In this context, the accurate planning of open parking spaces is of strategic importance in urban planning studies. Within the scope of the research, the criteria affecting parking lot selection were classified into five main groups: spatial and technical criteria, natural criteria, social criteria, legal criteria, and economic criteria. These data were processed and analyzed in spatial layers within a Geographic Information Systems (GIS) environment. In the modelling process, an Artificial Neural Network (ANN) model was chosen, which has flexible learning and generalisation capabilities in multi-criteria decision-making environments. The ANN model demonstrates the ability to identify potential parking spaces with high accuracy by learning the complex and nonlinear relationships between spatial criteria. The training and test data used to create the model were fed by five main criteria and twenty-one sub-criteria, and the significance coefficients between these criteria were determined using the analytical hierarchy process (AHP). Pilot areas in the study were selected from the Isparta city centre, and the suitability map obtained from these pilot areas provided concrete output that could contribute to local governments' decision-making processes regarding urban parking planning. The findings demonstrate that artificial neural network-based computational modelling transcends traditional approaches, enabling the development of more dynamic and data-driven planning tools.