Due to the rapid increase of waste vulcanized rubber products, the development of low-cost, efficient, and selective devulcanization processes is needed. In this paper, the devulcanization ability of Gordonia desulfuricans DSM 44462(T) was evaluated by a design of experiments. The aim of the experimental design was to investigate the importance of parameters influencing the bacterial growth, such as the glucose concentration (C), dibenzothiophene concentration (DBT), and initial biomass (optical density, OD) in biodevulcanization process. The complex viscosity (eta*) was chosen as experimental response for the experimental design. A multiple linear regression was used to model the relationship between the response and the process variables. In addition, the crosslink density and gel fraction were measured. Furthermore, the automated ribosomal intergenic spacer analysis (ARISA) as a microbiological method was performed to assess the persistence of the inoculated strain during the experiments. Reduced regression models were obtained considering only the significant variables and interactions. The glucose concentration C and OD variables and C-DBT and DBT-OD interactions resulted to the relevant parameters for the process. The fingerprinting showed the persistence of G. desulfuricans DSM 44462(T), despite the presence of other bacterial population after the VGNR sterilization. These results highlight the importance to support the physics analysis with microbiological analyses to evaluate the bacterial persistence during the treatment.