Abstract: Flux balance analysis (FBA) is a widely adopted computational model in the study of whole-cell metabolisms, often used to identify drug targets, to study cancer, and to engineer cells for targeted purposes. The most common model is a linear program that maximizes cellular growth rate subject to achieving steady metabolic state and to satisfying environmental bounds. Quadratic and integer modifications are also common. Standard stoichiometry decides the preponderance of data in most instances, and hence, the majority of information defining an optimization model is certain. However, several key parts of a model rely on inferred science and are less certain; indeed, the method of deciding several of these values is opaque in the literature. This prompts the question of how the resulting science might depend on our lack of knowledge. We suggest a robust extension of FBA called Robust Analysis of Metabolic Pathways (RAMP) that accounts for uncertain information. We show that RAMP has several mathematical properties concomitant with our biological understanding, that RAMP performs like a relaxation of FBA in practice, and that RAMP requires special numerical awareness to solve.