In this paper experimentations were made on shipbuilding aluminium alloy EN AW 5083 thickness of 8 mm in order to investigate influence of variable plasma jet cutting process parameters such as gas pressure, cutting speed, arc current and cutting height on formation of dross on the exit of the plasma jet from the workpiece material. Dross is significant cut quality response in plasma jet cutting process. It can be defined as blown molten metal concentrated on the bottom side of the sheet. Artificial intelligence method fuzzy logic was applied to model relations between input parameters and analysed response. Prediction accuracy of developed fuzzy logic model was checked by comparison between experimental and predicted data. Mean absolute percentage error (MAPE) and coefficient of determination (R2) were used as prediction accuracy measures. After prediction accuracy was proved it was concluded that such defined fuzzy logic model represents good basis for further more detailed experimental research in this area. Furtherly, it allows creation of fuzzy expert system that will enable deeper understanding of process parameters effects on dross formation and quite better response prediction.