In order to meet dominant growth of Salmonella spp. in a composed system of five pathogens for accurate detection, designing an appropriate selective enrichment broth was clearly needed. First, we built a high-throughput assay procedure based on SYBR Green I (TM) real-time PCR, which possessed the necessary specificity for Salmonella spp., a good linear standard curve with typical R (2) value (0.9984) and high amplification efficiency (99.0 %). Further, for the larger target biomass in the mixed microflora, acarbose, LiCl and bile salt were selected to optimize their concentrations using response surface methodology (RSM). A central composite design was employed to collect the data and fit the response. A quadratic polynomial model was derived by computer simulation. Statistical analysis was carried out to explore the action and interaction of the variables on the response. In the end, a novel broth (Sal-5) was formulated to allow the efficient enrichment of Salmonella spp. and inhibit the growth of other tested strains. A detection platform was developed, including selective enrichment in Sal-5, DNA extraction by the boiling lysis method and real-time PCR test based on SYBR Green I (TM). This work could extend the application of RSM and real-time PCR in the design of other selective enrichment media for common pathogens.