Artemisinin is a potent antimalarial drug; however, it suffers from unstable and insufficient supply from plant source. Here, we established a novel multivariate-modular approach based on experimental design for systematic pathway optimization that succeeded in improving the production of amorphadiene (AD), the precursor of artemisinin, in Escherichia coli. It was initially found that the AD production was limited by the imbalance of glyceraldehyde 3-phosphate (GAP) and pyruvate (PYR), the two precursors of the 1-deoxy-d-xylulose-5-phosphate (DXP) pathway. Furthermore, it was identified that GAP and PYR could be balanced by replacing the phosphoenolpyruvate (PEP)-dependent phosphotransferase system (PTS) with the ATP-dependent galactose permease and glucose kinase system (GGS) and this resulted in fivefold increase in AD titer (11 to 60 mg/L). Subsequently, the experimental design-aided systematic pathway optimization (EDASPO) method was applied to systematically optimize the transcriptional expressions of eight critical genes in the glucose uptake and the DXP and AD synthesis pathways. These genes were classified into four modules and simultaneously controlled by T7 promoter or its variants. A regression model was generated using the four-module experimental data and predicted the optimal expression ratios among these modules, resulting in another threefold increase in AD titer (60 to 201 mg/L). This EDASPO method may be useful for the optimization of other pathways and products beyond the scope of this study.