The doctoral thesis contains a research on transport travel demand model development approach based on machine learning and simulation methods for transport impact analysis. Special attention in the simulation model is paid to the stages of transportation mode choice and generated transport trip number determination. The elaborated transport travel demand simulation model development approach includes: 1) smart growth methods analysis for generated transport trip determination; 2) transportation mode choice technique for transport impact analysis; 3) transport travel demand simulation model calibration procedure for transport impact analysis. In the course of research, the intermediary results referred to the transport travel demand simulation model development approach were achieved: transport travel demand model types and their application in transport impact analysis were investigated; validation procedures of transport travel demand simulation model were analysed; combining of data mining methods was done for the selection of samples in order to decrease noisy data; initial gradient and traversal origindestination matrix evaluation was done for trip assignment; and the transport travel demand simulation model for mixed-use building.