The aim of this thesis was to define the functional relationships between the emission structure of the M1 vehicle fleet and greenhouse gas emissions. These relationships were used to develop a model for estimating and managing greenhouse gas emissions from the entire road traffic. The model consists of three sub-models: a sub-model for emission estimation, a forecast sub-model and a sub-model for emission control. The function of the first sub-model for emission estimation is performed by COPERT 5. The second sub-model has two additional sub-models: an additional forecast sub-model and an additional scenario sub-model. In the additional forecast sub-model, regression analysis was used to forecast M1 fleet emission factors, while the ARIMA time series method was used to forecast the number of vehicles in the fleet and the average annual distance travelled. The diffusion of new xEV propulsion systems in the M1 fleet was modelled in the additional scenario sub-model, using diffusion functions. A very strong correlation was observed between CO2 emissions from the M1 fleet and the total greenhouse gas emissions from the entire road traffic in the EU-28. Greenhouse gas emissions from road traffic in the Republic of Croatia were estimated on the basis of these results. It was demonstrated that the three emission indicators introduced help monitor the achievement of set target values for annual emissions from road traffic. Based on historical data for the period between 2007 and 2016, the additional forecast sub-model was used to reconstruct emissions in the 2005 reference year. The scenario analysis applied to six different scenarios indicated how the regular annual fleet renewal should be managed in order to achieve the compulsory 30% reduction in road traffic emissions in the European Union by 2030 compared to the 2005 reference year. The thesis outlines the challenges that Croatia will face in meeting this objective.