A hierarchical rule-based expert system, called AZEOPERT, to predict the binary azeotrope formation between components in mixtures is presented. Specific knowledge included in AZEOPERT and the implementation of AZEOPERT are described along some of the several components of expert system applications : knowledge representation strategy and levels of knowledge abstraction, inference machine, user interface, explanation facilities. AZEOPERT may be a useful tool for many chemical engineering activities, especially process synthesis and when selecting an entrainer for azeotropic distillation and an appropriate solvent for extractive distillation or extraction.