The biosynthesis of surfactin has attracted broad interest; however, there is a bottleneck in its low yield in wild strains and the ability to engineer Bacillus producers. Because the key metabolic mechanisms in the surfactin synthesis pathway remain unclear, genetic engineering approaches are all ending up with a single or a few gene modifications. The aim of this study is to develop a systematic engineering approach to improve the biosynthesis of surfactin. First, we restored surfactin biosynthetic activity by integrating a complete sfp gene into the nonproducing Bacillus subtilis 168 strain and obtained a surfactin titer of 0.4 g/l. Second, we reduced competition by deleting biofilm formation-related genes and nonribosomal peptide synthetases/polyketide synthase pathways (3.8% of the total genome), which increased the surfactin titer by 3.3-fold. Third, we improved cellular tolerance to surfactin by overexpressing potential self-resistance-associated proteins, which further increased the surfactin titer by 8.5-fold. Fourth, we increased the supply of precursor branched-chain fatty acids by engineering the branched-chain fatty acid biosynthesis pathway, resulting in an increase of the surfactin titer to 8.5 g/l (a 20.3-fold increase). Finally, due to the preference of the glycolytic pathway for cell growth, we diverted precursor acetyl-CoA away from cell growth to surfactin biosynthesis by enhancing the transcription of srfA. The final surfactin titer increased to 12.8 g/l, with a yield of 65.0 mmol/mol sucrose (42% of the theoretical yield) in the metabolically engineered strain. To the best of our knowledge, this is the highest titer and yield that has been reported. This study may pave the way for the commercial production of green surfactin. More broadly, our work presents another successful example of the modularization of metabolic pathways for improving titer and yield in biotechnological production.