The aim of this dissertation is to evaluate the influence of water content dissolved in crude oil and the meter factor on the uncertainty of the fiscal oil metering using ultrasonic flow meters. The development of this work is motivated by the fact that the Technical Regulation for the Measurement of Oil and Natural Gas (RTM) limits which methodologies can be used for water analysis and what is the maximum permissible uncertainty for the oil flow meter in which it is inserted. However, the RTM doesn t define the uncertainty of the water analysis. There is a risk of adopting one of the permitted methods of analysis and, even so, the measurement uncertainty of the flow meter exceeds the defined by the legislation. In addition, the RTM adopts a repeatability standard when validating the calibration of ultrasonic meters, thus the influence of this on oil uncertainty is also evaluated. The methodology used is the design of experiments and response surface. The results demonstrate that the influence of the water content uncertainty analysis and of the meter factor on the uncertainty of the oil flow meter can be simulated. Finally, the conclusions show that the average currently found for uncertainty of water is adequate to maintain the oil uncertainty within the allowed by the RTM, but small variations may cause this limit to be exceeded. The influence of the meter factor uncertainty was inferior or similar to the water content and, finally, it is suggested the adoption of a standard limit of uncertainty instead of a standard of repeatability for the ultrasonic meters.