Road accidents impose serious problems to society in terms of human, economic, property damage and medical costs, as the total number of deaths from road accidents worldwide continues to rise, reaching a high of 1.35 million. The aim of this Ph.D. thesis is to provide integrated support of driver traffic behaviour and safety by smartphone data. Through algorithms and statistical analyses such as Machine Learning, Structural Equation Models (SEMs), etc. appropriate models will be developed to analyse the traffic behaviour and safety of the driver. Specifically, the most critical risk factors that affect the traffic behaviour and safety of drivers (speed, harsh events, driver’s distraction, etc.) will be investigated, while assessing the impact of personalized feedback on driving behavior, as well. The main expected outcomes of this Ph.D. thesis concern both the impact on road safety and other significant socio-economic impacts, as well as new approaches to educating and supporting drivers and safer use of vehicles.