Current evaluations of public transport accessibility strongly depend on applied models of accessibility measures and their parameterisation. The project objective is to develop and evaluate new spatial simulation models suitable for more accurate assessment of accessibility in regional and local studies. Models will integrate Monte Carlo stochastic simulation to generate a large volume of random routes according to theoretical models of expected individual behaviour, calibrated using 2011 census data, outputs of local surveys and established model constraints like structure and volume of local population, capacity of services, time limitations etc. Two regional and two local (city) pilot areas will be used for design and optimisation of models. Outputs from new spatial simulation models will be compared with usually used accessibility measures. The project will provide procedures how to distinguish transport and geographical factors in the accessibility evaluation and to promote next research of these factors, and how to measure the real performance of the public transport.