Hot embossing replica are characterized by the quality of the molded structures and the uniformity of the residual layer. In particular, the even distribution of the residual layer thickness (RLT) is an important issue in hot embossing and the related process of thermal nanoimprint lithography, as variations in the RLT may affect the functionality or further processing of replicated parts. In this context, the paper presents an experimental and simulation study on the influence of three process factors, namely the molding temperature, the embossing force, and the holding time, on the residual layer homogeneity achieved when processing 2 mm thick PMMA sheets with hot embossing. The uniformity of the RLT was assessed for different experimental conditions by calculating the standard deviation of thickness measurements at different set locations over the surface of each embossed sample. It was observed that the selected values of the studied parameters have an effect on the resulting RLT of the PMMA replica. In particular, the difference between the largest and lowest RLT standard deviation between samples was 18 μm, which was higher than the accuracy of the instrument used to carry out the thickness measurements. In addition, the comparison between the obtained experimental and simulation results suggests that approximately 12% of the RLT uniformity was affected by the local deflections of the mold. Besides, polymer expansion after release of the embossing load was estimated to contribute to 8% of the RLT nonuniformity. It is essential to understand the effects of the process parameters on the resulting homogeneity of the residual layer in hot embossing. In this research, the best RLT uniformity could be reached by using the highest considered settings for the temperature and holding time and the lowest studied value of embossing force. Finally, the analysis of the obtained results also shows that, across the range of processing values studied, the considered three parameters have a relatively equal influence on the RLT distribution. However, when examining narrower ranges of processing values, it is apparent that the most influential process parameter depends on the levels considered. In particular, the holding time had the most effect on the RLT uniformity when embossing with the lower values of process parameters while, with higher processing settings, the molding temperature became the most influential factor.