The accuracy of manufactured parts holds paramount importance in product assembly. Among various additive manufacturing (AM) techniques, fused deposition modeling (FDM) stands out for its emphasis on accuracy, surface quality, and precision. Hitherto, some investigations have been conducted on dimensional accuracy which are limited to few geometrical shapes and remains inadequately explored. This study adopts an experimental approach to investigate the influence of FDM printing parameters including build orientation (BO), layer thickness (LT), and infill density (ID) on the dimensional accuracy of ABS polymer printed benchmark. Taguchi’s L9 orthogonal array is utilized to assess the correlation between the chosen FDM process parameters and the obtained responses. Measurements of deviations of the printed parts are conducted using a vision measuring machine (VMM) and analyzed employing the ‘nominal-is-the-best’ method. In this study, the most influencing parameter is changing as the geometrical shapes vary. Analysis of variance (ANOVA) is performed to examine the adequacy of the model, considering p-values less than 0.05 at 95% confidence level. The model is found adequate for all the dimensions of the benchmark. The study concludes that the parameter setting will vary with the changes in features. This study will pave the way in selecting the process parameters to get the desired dimensional accuracy for improving the quality of critical applications like prototyping and functional end-use manufacturing.