Analysis-of-variance (ANOVA) is a standard statistic method for assessment of the influence of various factors on fatigue resistance in the finite life domain. However, the previous research has shown that this method was not capable to determine with sufficient confidence if the build orientation, the thickness of allowance for machining, and the position in the production chamber affect fatigue resistance of Maraging steel MS1 products made by direct metal laser sintering (DMLS) technology. To contribute to a better understanding of the subject, the results of fatigue test experiments were used for training of four types of artificial neural networks (ANN) for assessment of fatigue resistance in the finite life domain. Each ANN had different structure of inputs, which corresponded to a different combination of the factors of DMLS production process. The differences between the predictive abilities of the ANN were attributed to influences of the respective factors on the fatigue resistance of the material in the finite life domain. The approach was verified by the agreement with the conclusive results of ANOVA analyses. Furthermore, in the cases when ANOVA does not lead to a clear result, the analyses of the predictive ability of the ANN strongly suggest that build orientation and thickness of allowance do not influence, while the position of a part in production chamber influences, the fatigue resistance in the finite life domain of Maraging steel MS1 produced by DMLS technology.