The objective of this paper is to present a systematic methodology for performing analysis of sensitivity of process design due to uncertainties in property estimates. The methodology provides the following results: a) list of properties with critical importance on design; b) acceptable levels of accuracy for different thermo-physical property prediction models; and c) design variables versus properties relationships. The application of the methodology is illustrated through a case study of an extractive distillation process and sensitivity analysis of designs of various unit operations found in chemical processes. Among others vapour pressure accuracy for azeotropic mixtures is critical and needs to be measured or estimated with a ±0.25% accuracy to satisfy acceptable safety levels in design.