pH is an important factor affecting the growth and production of microorganisms; especially, it is effective on the efficiency of ethanologenic microorganisms. It can change the ionization state of metabolites via the change in the charge of their functional groups that may lead to metabolic alteration. Here, we estimated the ionization state of metabolites and balanced the charge of reactions in genome-scale metabolic models of Saccharomyces cerevisiae, Escherichia coli, and Zymomonas mobilis at pH levels 5, 6, and 7. The robustness analysis was first implemented to anticipate the effect of proton exchange flux on growth rates for the constructed metabolic models at various pH. In accordance with previous experimental reports, the models predict that Z. mobilis is more sensitive to pH rather than S. cerevisiae and the yeast is more regulated by pH rather than E. coli. Then, a systemic approach was proposed to predict the pH effect on metabolic change and to find effective reactions on ethanol production in S. cerevisiae. The correlated reactions with ethanol production at predicted optimal pH in a range of proton exchange rates determined by robustness analysis were identified using the Pearson correlation coefficient. Then, fluxes of these reactions were applied to cluster the various pHs by principal component analysis and to identify the role of these reactions on metabolic differentiation because of pH change. Finally, 12 reactions were selected for up and down-regulation to improve ethanol production. Enzyme Regulators of the selected reactions were identified using the Brenda database and 11 selected regulators were screened and optimized via Plackett-Burman and 2-level full factorial designs, respectively. The proposed approach has enhanced yields of ethanol from 0.18 to 0.36 mol/mol carbon. Hence, not only a comprehensive approach for understanding the effect of pH on metabolism was proposed in this work, but also it successfully introduced key manipulations for ethanol overproduction.