This thesis is prepared as an outcome of the Energy Engineering Master of Science program at Izmir Institute of Technology, IZTECH, in Turkey. The main purpose of this study is to analyse the effects of water content in diesel fuel spray behaviours using adaptive neuro-fuzzy inference system models (ANFIS) for compression ignition engines. The investigations are carried out using numerical models of ANFIS in MATLAB R2011a, generating simulations from training and test datasets based on recent experimental studies from the literature. The thesis primarily tests the use and the fitness of ANFIS models, modifying the neural network structure so that the simulations acceptably reach the experimental results accurately. Then secondarily, the simulation is used to investigate the effects of parameters originally not available in the related study. The investigation mainly focusses on water in diesel emulsions from pure diesel to an emulsion with 20% water content. Operational conditions such as chamber ambient pressure, injection pressure, chamber ambient temperature are also investigated to find their effects over spray penetration. It was found that the increase of water content in the diesel fuel did not have a relevant effect at very low and very high temperatures, however at medium-high temperatures it increased spray penetration. Furthermore, it was observed that the increase of chamber ambient pressure and chamber ambient temperature reduced the spray penetration as expected.